subfolder
stringclasses 367
values | filename
stringlengths 13
25
| abstract
stringlengths 1
39.9k
| introduction
stringlengths 0
316k
| conclusions
stringlengths 0
229k
| year
int64 0
99
| month
int64 1
12
| arxiv_id
stringlengths 8
25
|
|---|---|---|---|---|---|---|---|
1609 | 1609.03227_arXiv.txt | We investigate the chemical composition of {the solar system's} giant planets atmospheres using a physical formation model with chemistry. The model incorporate disk evolution, pebbles and gas accretion, type I and II migration, {simplified} disk photoevaporation and solar system chemical measurements. We track the chemical compositions of the formed giant planets and compare them to the {observed values}. Two {categories of} models are studied: with and without disk chemical enrichment via photoevaporation. Predictions for the Oxygen {and Nitrogen} abundances, core masses, and total amount of heavy elements for the planets are made {for each case}. We find that in the case without disk PE, {both Jupiter and Saturn will have a small residual core and comparable total amounts of heavy elements in the envelopes. We predict oxygen abundances enrichments in the same order as carbon, phosphorus and sulfur for both planets. Cometary Nitrogen abundances does not allow to easily reproduce Jupiter's nitrogen observations.} In the case with disk PE, less core erosion is needed to reproduce the chemical composition of the atmospheres, so both planets will end up with {possibly more massive residual cores, and higher total mass of heavy elements}. {It is also significantly easier to reproduce Jupiter's Nitrogen abundance.} No single was disk was found to form both Jupiter and Saturn with all their constraints in the case without photoevaporation. No model was able to fit the constraints on Uranus \& Neptune, hinting toward a {more complicated formation mechanism for these planets. The predictions of these models} should be compared to the upcoming Juno measurements to better understand the origins of the solar system giant planets. | Twenty years ago, the Galileo probe dove into Jupiter's atmosphere bringing a wealth of informations on its properties, structure and chemical composition. One of Galileo's significant results was the discovery that Jupiter's atmosphere is enriched by a factor of 2-4 with respect to solar value with most volatile elements \citep{atreya2003}, including noble gases and Nitrogen, {elements that condense} at extremely low temperatures. This same trend is also seen in Saturn, Uranus and Neptune, suggesting further that the solar system giant planets all formed via {core accretion, since the alternative gravitational instability model predicts solar abundances for all elements. A {summary} of {the giant} planets chemical compositions {and core masses} can be found in table \ref{table_enri} (from \cite{mousis2014}).} \renewcommand\arraystretch{1.2} \begin{table} \begin{center} \caption{Measured chemical enrichments in the giant planets atmospheres normalized to the protosolar value, followed by the observational constraints on their cores masses and total amount of heavy elements.} \small {\begin{tabular}{lcccccc} \hline \noalign{\smallskip} & \multicolumn{2}{c}{Jupiter} & \multicolumn{2}{c}{Saturn} & \multicolumn{2}{c}{U \& N} \\ Species & E & $\Delta$E$^{\rm(a)}$ & E & $\Delta$E$^{\rm(a)}$ & E \\ \hline C & 4.40 & 1.14 & 9.90 & 1.05 & 20-60\\ N & 4.18 & 2.08 & 2.5--3.8 & -- & ?? \\ O$^{\rm(b)}$ & 0.42 & 0.15 & ?? & ?? & ?? \\ P & 3.34 & 0.36 & 11.54 & 1.35 & ?? \\ S & 2.94 & 0.70 & 12.88 & 1.52 & ?? \\ He & 0.72 & 0.04 & 0.71 & 0.14 & ?? \\ Ne$^{\rm(c)}$ & 0.12 & -- &?? & ?? & ?? \\ Ar & 2.62 & 0.86 &?? & ?? & ?? \\ Kr & 2.23 & 0.61 & ?? & ?? & ?? \\ Xe & 2.18 & 0.61 & ?? & ?? & ?? \\ \hline D/H & 1 & &1 & & 1.9 & \\ \hline M$_c$ (M$_\oplus$) & < 10 & & < 20 & & ?? \\ M$_Z$ (M$_\oplus$) & < 42 & & 16 - 30 & & > 85\% \\ \hline \end{tabular}}\\ $^{\rm(a)}$Error is defined as ($\Delta$E/E)$^2$ = ($\Delta$X/X$_{\rm planet}$)$^2$ + ($\Delta$X/X$_{\rm Protosun}$)$^2$; $^{\rm(b)}$this is a lower limit; $^{\rm(c)}$this is an upper limit. \label{table_enri} \end{center} \end{table} Even though the details of Jupiter's formation remain elusive, several mechanisms --in the framework of core accretion-- has been proposed as the origin of the observed enrichments : 1- The giant planets formation in the cold outer disk where {the temperatures are low enough for the volatiles to condense or get trapped in clathrates-hydrates or amorphous ices \citep{owen1999,gautier2001,hersant2004,alibert2005,mousis20122}}. These models usually assume high trapping efficiency of volatiles in the ices. 2- The giant planets formation in a chemically evolved disk \citep{guillot2006,throop,monga2015}. In these scenarios, the disk evaporation along with dust settling \& solids drift and sublimation enrich the midplane in {metals, where the giant planets will later form and accrete the enriched gas.} The {chemical} enrichment mechanism in the giant planets and specifically Jupiter is therefore still an open question. More detailed discussions on these models can be found in \cite{mousis2014}.\\ In this work, we investigate these proposed mechanisms in the light of new observational and theoretical advancements in planets formation and protoplanetary disks chemistry. For the first time, Argon and molecular Nitrogen abundances have been measured in a comet (67P/C-G by Rosetta \citep{rubin2015,balsiger2015}) giving us valuable insights on the composition of the protosolar nebula. Moreover, the in the last few years, the fast and efficient pebbles accretion models became main stream scenarios for planets formation \citep{lamb1,lamb2}, giving promising solutions to long standing problems on the origin of giant planets. This type of accretion for example can allow for a rapid formation of a giant planet's core even in the outer tenuous nebula. Finally, the \textit{Juno} space probe currently {around} Jupiter will will tentatively measure the elusive Oxygen abundance of the planet in addition to refining its nitrogen abundance and core mass values \citep{bolton2010}. It is crucial hence to have quantitative predictions for these quantities from the different enrichment scenarios (formation in the outer disk vs enrichment via photoevaporation) that are compatible with the known planets formation physics in addition to the chemical constraints we have on the giant planets. The questions {we address in this work are:} \begin{itemize} \item Where/when should Jupiter and Saturn form if their chemical compositions were acquired from a disk with solar composition ? \item How much should the initial solid core get eroded into the envelopes of these planets to reach the observed chemical enrichments ? \item How will the disk chemical enrichment via photoevaporation affect the above ? \item Can Jupiter and Saturn still form late in the disk when photoevaporation has {started}? \item Are these models compatible with the dynamical histories of these planets ? \item Can we disentangle these formation scenarios via Juno measurements ? \end{itemize} In section 2 we detail the planets formation model and our approach to disk chemistry. We discuss the results in section 3 and finally summarize and conclude in section 4. \\ | \subsection{Caveats \& perspectives} Our model incorporate a large number of different disk and planets physical and chemical effects. It is inherently simplified for numerical complexity reasons. The following caveats can hence be identified: \begin{itemize} \item The major caveat of this work is the simplicity of the disk external photoevaporation model. For more robust and quantitative results, proper dynamical coupling of a viscously evolving disk to photoevaporation and planets formation is needed and should be addressed in future works. The case of internally photoevaporated disk is also relevant and should be explored. \item Another possible caveat is the unevolving disk chemistry we assumed. Even though our disk chemistry is based on small bodies and disks observations, it is unlikely that this chemistry was static and unchanged during the disk lifetime. However to date no chemical model was able to fully reproduce the chemical abundances of today's small bodies from first principles. Nonetheless, a more sophisticated disk chemistry can bring valuable insights on the formation of giant planets. \item In interpreting the chemical composition of the giant planets atmospheres as the bulk elemental abundance, we are implicitly assuming these planets to be fully convective. Even though this seems as the most likely case, it is still the subject of an ongoing debate \citep{guillot2014}. \item Many of the disk parameters (dust and pebbles metallicities) and planets parameters (envelope density and opacity) were fixed to a constant value throughout this work to keep the problem tractable. Exploring these parameters should shed more lights on the formation of giant planets, specially exoplanets, and is left for future work. \item In this model we implemented a very simple yes/no indicator to check if Jupiter/Saturn crossed their mutual 2:3 mean motion resonance. For future works however it is important to track the full dynamical evolution of these planets, {along with type III migration,} and to include Uranus and Neptune. \item {This work used the chemical composition of three comets (1P/Halley, 67P/C-G and 81P/Wild) as indicators for the disk's chemistry. In reality, dynamical mixing of the cometary reservoirs have probably meant that both OCCs and JFCs probably sample material formed at a wide variety of orbits \citep{altwegg,brasser}. A deeper understanding of the statistics of comets chemistry is key to improve upon this work.} \item {The models presented in this work assume that the heavy elements in the planet's envelope originates as either material eroded from the core or delivered from a chemically evolved gas. Heavy elements however can also be delivered into the envelope from late contamination by planetesimals. This phenomena however is inefficient. \cite{guillot2000,matter} for example found that the late heavy bombardment could have increased the volatiles enrichment in the atmospheres of Jupiter and Saturn by factors of respectively 0.033 and 0.074. } \item {In this model, planet and disk evolution are calculated separately. We do not take into account the effect of solids removed from the disk into the forming planet on the structure of the disk. This however should should be a second order importance to the effect of the planets itself on the disk. The reason is the nature of pebbles accretion, were pebbles form continuously in the outer disk and drift inward till they get accreted by the core. The structure of the pebbles disk is therefore not being affected by accretion.} \item {Another assumption concerning the disk structure is the constant accretion rate throughout the disk. This is unlikely to be true in the outer parts of the disk where the viscous timescale is comparable to the disk lifetime. Quantifying this effect however necessitate modeling the viscous evolution of the disk, which not taken into account in the disk model of \cite{bitsch2}. A non uniform accretion rate might lead to disk structure affecting the evolution of planets (specially migration). Modeling these effects is left for future works.} \end{itemize} \subsection{Summary} In this work we explored the chemical composition of giant planets atmospheres using a formation model that includes simplified pebbles \& gas accretion, type I and II migration, solar system chemistry observations, \& external photoevaporation, and planet core erosion. We ran population synthesis simulations for individual planets in disks with and without photoevaporation to understand the origins of the solar system giant planets chemical composition, in addition to simulations with both Jupiter and Saturn to check the compatibility of their chemical composition with their dynamical histories. Our conclusions can be summarized as follow: \begin{itemize} \item Since Sulfur and, more probably, Phosphorus were both present as mostly refractory phases throughout the protosolar nebula, and because the uncertainties on their abundances in Jupiter are low, they can hence be used to constrain the fraction of core mass that got eroded into its envelope. The other more volatile (hence location dependent) elements such as Argon and Carbon can be used to constrain where/when in the disk did Jupiter form. {The case with photoevaporation however hints toward a fraction of S and P in volatile phase, since this is the only possibility to form Jupiter in the inner 10 AU.} \item {The highest fraction of Nitrogen in refractories measured so far in today's solar system (14\% solar abundance) along with a 20 K N$_2$ condensation temperature are too low to explain the measured nitrogen abundance in both Jupiter and Saturn if no significant photoevaporation induced disk chemical enrichment took place. This implies that either the Nitrogen chemically evolved significantly in the disk or that it condensed at higher temperatures through trapping in clathrate-hydrate ices.} \item In the case with no disk photoevaporation, Jupiter's chemical composition is only obtained if it formed beyond 24 AU. This model predicts an oxygen enrichment in the order of 3 times solar, a residual core of 2 or 5 M$_{\oplus}$ where 60-80\% of the original core was eroded, and a total mass of heavy elements in the planet (M$_Z$) in the order of 15-17 M$_{\oplus}$. Saturn on the other hand can form in a larger swath of the disk (1-5 and 20-25 AU). The model predicts a 40-60\% eroded core for this planet with M$_c$ around 6-7 M$_{\oplus}$ and M$_Z$ in the order of 16-18 M$_{\oplus}$ and an oxygen enrichment between 5 and 12 times solar. {No single disk was found to form both planets in the right order with all their constraints applied, indicating that either different physics or chemistry are at play, or that this case is not a valid for our solar system formation.} In the case of Uranus \& Neptune, no set of parameters were found to match at the same time their chemical composition and interior structure constraints. \item {In the case with disk photoevaporation, a planet with Jupiter's chemical composition can form all the way in to around 8 AU, due to the disk's photoevaporation enriching it in heavy elements. The gas accreted by the planet is usually enriched due to PE by factors slightly less than 2 with respect to solar value. The rest of the metals in the envelope have to be provided by core erosion. These Jupiters have a residual core mass in the order of 1 to 3 M$_{\oplus}$, M$_Z$ in the order of 14-15 M$_{\oplus}$. This model predicts the same oxygen abundance for Jupiter as in the case with no disk PE. In the case of Saturn, planets respecting all the constraints can be created continuously throughout the disk. These are separated into two populations with core erosion factors between 40\% and 60\%, giving respective lower limits on M$_c$ and M$_Z$ of 7 and 15 M$_{\oplus}$, and upper limits of 15 and 25 M$_{\oplus}$. The predicted water enrichment for Saturn is in the order of 12-15 times solar abundance, even though values as low as 6 are not excluded. The cases with and without PE might then be distinguishable by M$_c$ and M$_Z$ of the giant planets, but degeneracies can be present. In contrast to the case without PE, pairs of Jupiter-Saturn respecting all of the constrains and forming in the right order were found in the case with PE. As in the case with no disk PE, no matches for Uranus \& Neptune have been found. This might be implying {a more complex formation mechanism for these planets similar to the CO ices enhancement proposed by \citep{ali-dibb}}.} \item When checking for the pairs of Jupiter \& Saturn masses planets who crossed their mutual 2:3 resonance while ending up with compatible chemical compositions, non were found in the two cases with and without PE. This might be hinting to a major role of type III migration in our solar system history. \end{itemize} \begin{figure} \includegraphics[scale=0.2]{jup1.pdf} \caption{The final position (Rf) of Jupiter mass planets as a function of the seed's initial position (R0) and injection time (T$_{ini}$). The geometrical forms are the seed's initial mass in M$_{\oplus}$. This is the case with no disk photoevaporation, and showing all planets with no constraints.} \label{fig:jup1} \end{figure} \begin{figure} \includegraphics[scale=0.2]{jup2.pdf} \caption{Same as Figure \ref{fig:jup1}, but with all constraints (chemical composition and core mass) applied.} \label{fig:jup2} \end{figure} \begin{figure*} \includegraphics[scale=0.2]{jup3.pdf} \caption{The enrichment factor of Sulfur, Phosphorus, Nitrogen, Oxygen, Argon and Carbon predicted by this model for the case with no disk photoevaporation for Jupiter. Bottom panels show the predicted core mass and total amount of heavy elements in the planet. The dashed horizontal lines are the observational constraints. } \label{fig:jup3} \end{figure*} \begin{figure} \includegraphics[scale=0.2]{sat1.pdf} \caption{The final position (Rf) of Saturn mass planets as a function of the seed's initial position (R0) and injection time (T$_{ini}$). The geometrical forms are the seed's initial mass in M$_{\oplus}$. This is the case with no disk photoevaporation, and showing all planets with no constraints.} \label{fig:sat1} \end{figure} \begin{figure} \includegraphics[scale=0.2]{sat2.pdf} \caption{Same as Figure \ref{fig:sat1}, but with all constraints (chemical composition and core mass) applied.} \label{fig:sat2} \end{figure} \begin{figure*} \includegraphics[scale=0.2]{sat3.pdf} \caption{The enrichment factor of Sulfur, Phosphorus, Nitrogen, Oxygen, and Carbon predicted by this model for the case with no disk photoevaporation for Saturn. The dashed horizontal lines are the observational constraints. Bottom left panels show the predicted total amount of heavy elements in the planet. Core masses are now shown because all of these model have completely eroded cores.} \label{fig:sat3} \end{figure*} \begin{figure} \includegraphics[scale=0.2]{nep2.pdf} \caption{The final position (Rf) of Neptune mass planets as a function of the seed's initial position (R0) and injection time (T$_{ini}$). The geometrical forms are the seed's initial mass in M$_{\oplus}$. This is the case with no disk photoevaporation, and showing all planets with no constraints.} \label{fig:nep2} \end{figure} \begin{figure} \includegraphics[scale=0.2]{nep1.pdf} \caption{Same as Figure \ref{fig:nep2}, but with the carbon abundance and D/H constraints applied. Adding the informations on the interior structure will remove all matching planets.} \label{fig:nep1} \end{figure} \begin{figure*} \includegraphics[scale=0.2]{jwipo1.pdf} \caption{The final position (Rf) of Jupiter mass planets as a function of the seed's initial position (R0) and injection time (T$_{ini}$) and showing all planets with no constraints. The geometrical forms are the seed's initial mass in M$_{\oplus}$. This is for case with disk photoevaporation, where FUV is the value of $\dot{M}_{FUV}$ used.} \label{fig:jwipo1} \end{figure*} \begin{figure*} \includegraphics[scale=0.2]{jwipo2.pdf} \caption{Same as Figure \ref{fig:jwipo1}, but with all constraints (chemical composition and core mass) applied. This is in the case with disk photoevaporation included.} \label{fig:jwipo2} \end{figure*} \begin{figure*} \includegraphics[scale=0.2]{jwipo3.pdf} \caption{The enrichment factor of Sulfur, Phosphorus, Nitrogen, Oxygen, Argon and Carbon predicted by this model for the case with disk photoevaporation included for Jupiter. Bottom panels show the predicted core mass and total amount of heavy elements in the planet. The dashed horizontal lines are the observational constraints.} \label{fig:jwipo3} \end{figure*} \begin{figure*} \includegraphics[scale=0.2]{satwipo1.pdf} \caption{The final position (Rf) of Saturn mass planets as a function of the seed's initial position (R0) and injection time (T$_{ini}$) and showing all planets with no constraints. The geometrical forms are the seed's initial mass in M$_{\oplus}$. This is for case with disk photoevaporation, where FUV is the value of $\dot{M}_{FUV}$ used.} \label{fig:satwipo1} \end{figure*} \begin{figure*} \includegraphics[scale=0.2]{satwipo2.pdf} \caption{Same as Figure \ref{fig:satwipo1}, but with all constraints (chemical composition and core mass) applied. This is in the case with disk photoevaporation included. The two distinct populations seen correspond to the 40\% and 60\% core erosion factors. These values should hence be seen as lower and upper limits for the parameter being constrained.} \label{fig:satwipo2} \end{figure*} \begin{figure*} \includegraphics[scale=0.2]{satwipo3.pdf} \caption{The enrichment factor of Oxygen, and Carbon predicted by this model for the case with disk photoevaporation for Saturn. Top panels show the predicted total amount of heavy elements in the planet and their core masses. The dashed horizontal lines are the observational constraints. } \label{fig:satwipo3} \end{figure*} \begin{figure*} \includegraphics[scale=0.2]{nwipo1.pdf} \caption{The final position (Rf) of Neptune mass planets as a function of the seed's initial position (R0) and injection time (T$_{ini}$) and showing all planets with no constraints. The geometrical forms are the seed's initial mass in M$_{\oplus}$. This is for case with disk photoevaporation, where FUV is the value of $\dot{M}_{FUV}$ used.} \label{fig:nwipo1} \end{figure*} | 16 | 9 | 1609.03227 |
1609 | 1609.08629_arXiv.txt | We investigate the effects of dense environments on galaxy evolution by examining how the properties of galaxies in the $z=1.6$ protocluster \clustername\ depend on their location. We determine galaxy properties using spectral energy distribution fitting to 14-band photometry, including data at three wavelengths that tightly bracket the Balmer and 4000\AA\ breaks of the protocluster galaxies. We find that two-thirds of the protocluster galaxies, which lie between several compact groups, are indistinguishable from field galaxies. The other third, which reside within the groups, differ significantly from the intergroup galaxies in both colour and specific star formation rate. We find that the fraction of red galaxies within the massive protocluster groups is twice that of the intergroup region. These excess red galaxies are due to enhanced fractions of both passive galaxies (1.7 times that of the intergroup region) and dusty star-forming galaxies (3 times that of the intergroup region). We infer that some protocluster galaxies are processed in the groups before the cluster collapses. These processes act to suppress star formation and change the mode of star formation from unobscured to obscured. | In the local Unvierse there are clear correlations between the environment of a galaxy and its star formation rate (SFR), morphology and colour, such that dense environments tend to host galaxies with suppressed SFRs, red colours and early-type morphologies \citep[e.g.][]{Dressler1980, Balogh1998, Kauffmann2004}. The result is that local clusters and groups are graveyards of passively evolving galaxies, gradually fading as their stellar populations age. As we observe further out in the Universe, the SFR--density, colour--density and morphology--density relations become less clear \citep[e.g.][]{Cucciati2006,Cooper2008,Grutzbauch2011,Ziparo2013}. Beyond $z\sim1.4$, clusters host large numbers of highly star forming galaxies \citep[e.g.][]{Brodwin2013}, which suggest that environmental quenching of star formation does not play as strong a role at these high redshifts. However, these dense environments still host a larger fraction of passive and red galaxies, which indicates that galaxy properties still depend on their environment even at high redshift where environmental quenching is less influential \citep{Chuter2011,Quadri2012}. In this paper we study the protocluster \clustername\ at $z=1.6233$ \citep{Papovich2010,Tran2015}. This protocluster is one of the densest known regions of the early Universe. The forming cluster has an X-ray-derived main halo mass of $5.7\times10^{13}$\Msun\ \citep{Tanaka2010}, and it is likely to grow into a $3\times10^{14}$\Msun\ cluster by the present day \citep{Hatch2016}. The galaxy population of the main group of the \clustername\ protocluster has been extensively studied due to its location in the well observed SXDS-UDS field, and partial coverage by CANDELS {\it HST} observations. The nascent cluster has an enhanced density of star formation \citep{Tran2010}, but also an enhanced passive galaxy fraction \citep{Quadri2012} relative to the field. The member galaxies are likely to be undergoing accelerated galaxy growth through mergers, which is supported by evidence of an enhanced merger rate \citep{Lotz2013} and larger galaxy sizes \citep{Papovich2012} compared to the field. In this paper we investigate not only the main group, but the whole central 10 co-moving Mpc of the protocluster. We improve on previous studies by using an inventory of both star forming and passive protocluster galaxies, as well as detailed maps of the local density within the protocluster, to understand how galaxy properties correlate with environment before the cluster has fully collapsed. In Section \ref{method} we describe the method for locating the protocluster galaxies and determining their properties. In Section \ref{results} we compare the properties of the control field galaxies to the protocluster galaxies, and investigate the properties of galaxies as a function of their location within the protocluster. We discuss which protocluster environments affect galaxy properties and describe these environmental effects in Section \ref{discussion}, with conclusions following in Section \ref{conclusions}. We use AB magnitudes throughout and a $\Lambda$CDM flat cosmology with $\Omega_M=0.315$, $\Omega_\Lambda=0.685$ and $H_0=67.3$ \kmpspMpc\ \citep{Planckcosmology2014}. | \label{conclusions} The goal of this work is to determine the effects of dense environments on galaxy evolution at high redshift. We examined the properties of galaxies in the \clustername\ protocluster at $z=1.6$, and compared them to the field. A third of the protocluster galaxies reside in dense groups, whilst two-thirds reside between the groups in an intermediate-density environment. We present evidence showing that: \begin{itemize} \item[(i)] the protocluster intergroup environment does not greatly influence the masses or colours of the galaxies. Intergroup protocluster galaxies appear to have the same properties as field galaxies. \item[(ii)] the protocluster groups have twice the red galaxy fraction of the intergroup region. This is due to an enhanced passive fraction (by a factor of 1.7) and a higher fraction of star forming galaxies exhibiting red colours (by a factor of 3). The enhancement of red and passive galaxies within groups is most prominent at low stellar masses, but the excess of red star-forming galaxies appears to be at the same level at all masses ($>10^{9.7}$\Msun). \item[(iii)] both the stellar mass density ($\rho_{*}$) and star formation rate density ($\rho_{\rm SFR}$) are greatly enhanced in the protocluster, particularly within the groups where the $\rho_{\rm SFR}$ is $\sim800$ times that of the field. Both $\rho_{*}$ and $\rho_{\rm SFR}$ decrease with distance from the protocluster core, until they reach the field level at $\sim6$\,co-moving Mpc. The total sSFR is suppressed in the central 3\,co-moving Mpc region compared to the outer regions due to the presence of several massive groups within this radius. \end{itemize} We conclude that galaxies are preprocessed within the groups of the protocluster. This dense environment suppresses star formation and enhances the fraction of star forming galaxies that are red. | 16 | 9 | 1609.08629 |
1609 | 1609.04407_arXiv.txt | Together with interstellar turbulence, gravitation is one key player in star formation. It acts both at galactic scales in the assembly of gas into dense clouds, and inside those structures for their collapse and the formation of pre-stellar cores. To understand to what extent the large scale dynamics govern the star formation activity of galaxies, we present hydrodynamical simulations in which we generalise the behaviour of gravity to make it differ from Newtonian dynamics in the low acceleration regime. We focus on the extreme cases of interacting galaxies, and compare the evolution of galaxy pairs in the dark matter paradigm to that in the Milgromian Dynamics (MOND) framework. Following up on the seminal work by Tiret \& Combes, this paper documents the first simulations of galaxy encounters in MOND with a detailed Eulerian hydrodynamical treatment of baryonic physics, including star formation and stellar feedback. We show that similar morphologies of the interacting systems can be produced by both the dark matter and MOND formalisms, but require a much slower orbital velocity in the MOND case. Furthermore, we find that the star formation activity and history are significantly more extended in space and time in MOND interactions, in particular in the tidal debris. Such differences could be used as observational diagnostics and make interacting galaxies prime objects in the study of the nature of gravitation at galactic scales. | Interactions mark milestones in the evolution of galaxies by modifying their mass, stellar, gaseous and chemical contents, morphology, kinematics and dynamical properties \citep[see e.g.][among many others]{Arp1966, Sanders1996, Springel1999, Struck1999, Saintonge2012, Duc2013}. These events are often (but not always) associated with burst(s) of star formation such that, in the local Universe, all the most luminous galaxies (e.g. $> 10^{12} \Lsun$ for the ultra luminous infrared galaxies, ULIRGs, \citealt{Houck1985, Kennicutt1998b}) yield the signatures of major interactions \citep{Armus1987, Ellison2013}. Numbers of studies in all wavebands have characterised the properties of interacting systems, in particular their star formation activity, with the aim of pinning down the underlying physical processes \citep[e.g.][]{Schombert1990, Hibbard1995, Bournaud2004, Chien2007, Smith2010, Boquien2011, Saintonge2012, Scudder2012}. In these fast evolving objects with complex geometries, numerical simulations have long been considered as a fundamental complement to observations. Starting with \citet{Toomre1972}, all works point out the paramount role of gravitation on affecting both the galactic scale structures \citep[e.g.][]{Barnes1992, Quinn1993, Dubinski1996, Mihos1998, DiMatteo2007, Hopkins2009, Renaud2009, Moreno2013, Privon2013} and the internal, small scale, physics of interacting galaxies \citep[e.g.][]{Barnes1991, Teyssier2010, Chien2010, Hopkins2013, Renaud2014b}. The scale-free aspect of gravitation makes it indeed a key process at galactic scale in the shaping of galaxies and their structures (spiral, bars, large scale flows), at sub-galactic scales in the assembly of molecular clouds \citep{MacLow2004, Hennebelle2008} and down to the scale of pre-stellar cores in star formation \citep{Bate2005, Bonnell2013}. Galaxy interactions are thus the perfect benchmark to understand the role of gravitation on star formation, at both galactic and sub-galactic scales. The classical framework in which theoretical galactic studies are performed these days is the $\Lambda$-Cold Dark Matter ($\Lambda$CDM) paradigm. However, both the cosmological constant $\Lambda$ and the CDM part of the model could also be related to a modification of gravity. On galaxy scales, the model is indeed plagued by severe problems, the most famous ones being the cusp-core problem (\citealt{deBlok2010, Oman2015}, but see also \citealt{Read2016b}), the too-big-to fail problem \citep{BoylanKolchin2011, Papastergis2015, Pawlowski2015}, or the satellite planes problem \citep{Kroupa2005b, Metz2007, Metz2008, Pawlowski2012, Ibata2013, Ibata2014, Pawlowski2015}. There is also a more general problem linked to the finely tuned relation between the distribution of baryons and the gravitational field in galaxies, as encapsulated in various scaling relations involving a universal acceleration constant $a_0 \approx 10^{-10} \U{m\ s^{-2}}$, including the tight baryonic Tully-Fisher relation \citep{McGaugh2000, Lelli2016a, Papastergis2016}, the diversity of shapes of rotation curves at a given maximum velocity scale \citep{Oman2015}, or the relation between the stellar and dynamical surface densities in the central regions of galaxies \citep{Lelli2016b, Milgrom2016}, and many others \citep{Famaey2012}. All this points to things happening as if the effects usually attributed to CDM on galaxy scales were actually due to a modified force law. The {\it a priori} simplest explanation for this would be that gravity is indeed effectively different in the weak field regime and accounts for the effects usually attributed to CDM. This paradigm is known as Modified Newtonian Dynamics (MOND), or Milgromian Dynamics, suggested more than 30 years ago by \citet{Milgrom1983}. It predicted all the observed galaxy scaling relations well before they were precisely assessed by observations \citep{Famaey2012}. Nevertheless, this paradigm cannot be complete, as a full theory of gravitation, also valid on cosmological scales, has not yet been found. But while successful on galaxy scales, the MOND paradigm has still been far from being fully explored even on these scales where it is currently successful. Hence there is still potential for falsification of this paradigm in its {\it a priori} domain of validity. The main reason for this lack of exploration of all predictions of MOND on galaxy scales is its non-linear nature, and the previous lack of numerical codes devised to model galaxies in this framework. After the pioneering work of \citet{Brada1999}, only a handful of codes have been designed \citep{Nipoti2007, Tiret2007, Llinares2008, Angus2012}, but all with their own caveats, notably regarding the treatment of hydrodynamics. The first treatment of gas in MOND simulations was proposed by \citet{Tiret2008a}, who used sticky particles, but a full hydrodynamical treatment of gas has become possible only recently thanks to the patches of the \ramses code developed by \citet{Lughausen2015} and \citet{Candlish2015}. This is particularly important in order to study star formation in general, and in galaxy interactions in particular. The role of gravitation on star formation is central at both galactic and sub-galactic scales, and modifying it must have potentially observational consequences. For instance, during interactions, tides can induce the formation of tails expanding up to $\sim 100 \kpc$ away from their progenitor galaxies. Some regions of these tails can become unstable, fragment and form stellar objects as massive as dwarf galaxies ($\sim 10^{8-9} \Msun$). However, self-gravity is usually too weak to assemble such tidal dwarf galaxies (TDGs), and an external contribution to the local potential well is required. With $\Lambda$-CDM, only the potential well of the galactic DM halo can have such catalyst effect and thus, it must be sufficiently extended to embed the TDGs along the long tidal tails \citep{Bournaud2003}. But since the tidal debris originates from the discs (and thus contains very little DM, if any) and because the surrounding DM halos are dynamically hot, the DM distribution does not follow the baryonic one, and their external effect on TDG seeds remains mild. In MOND however, the baryonic seeds of TDGs generate their own ``phantom dark matter'' and thus, an additional potential well. Therefore, instabilities leading to the formation of TDG seeds are strengthened by the local MOND potential which amplifies them and allow them to grow. As a consequence, the formation of TDGs is eased in the MOND framework, compared to the Newton case \citep{Tiret2008b,Combes2010}. Since it provides a test of the gravitation paradigm, the nature of observed TDG candidates is intensively debated, in particular in the context of the the Tully-Fisher relation \citep[see e.g.][]{Gentile2007, Lelli2015, Flores2016} and regarding the potential origin of satellite galaxies sitting on satellite planes in the context of MOND, which could actually be old TDGs instead of primordially formed dwarf galaxies \citep[e.g.][]{Kroupa2010, Kroupa2015}. In this paper, we thus use the MOND framework to analyse the role of gravitation in enhancing star formation in interacting galaxies. We characterise the starburst activity associated to interactions in the context of MOND, and compare to that obtained when they are surrounded by halos of particle dark matter. We present our suite of simulations in Sect.~2 and the results in Sect.~3. Conclusions are drawn in Sect.~4. | Using hydrodynamical simulations, we present a comparison as directly as possible between interacting galaxies within the Newtonian and MOND frameworks. Our main results are: \begin{itemize} \item Replacing dark matter with a MOND formalism induces differences in dynamical friction and angular momentum transfer during the galactic encounters. As a result, the tidal debris spans a much larger volume in Milgromian models than in Newtonian ones for a given set of initial conditions. \item Comparable merger morphologies can still be obtained in Newtonian and Milgromian frameworks. We find that merely tuning down the relative velocity of the progenitors to balance the weak dynamical friction in MOND and reach comparable velocities at the first pericentre in both frameworks provides a reasonable correspondence between the models, at least on large scales. Thus, without a fine tuning of the parameters, the overall morphology and kinematics of an Antennae-like system can be reproduced in the MOND framework. This complements the pioneer work of \citet{Tiret2008b} who reproduced an Antennae-like morphology in MOND for the first time. \item While the global morphology and kinematic structure can be reproduced with Milgromian dynamics, small scale differences from the Newtonian model exist: mainly, the formation of stellar and gaseous clumps along the tidal tails and the spatial extent of the discs at the moment of the second encounter. By generating their own phantom dark matter and thus a deep potential well on small scales ($\lesssim 1 \kpc$), the tidal tails favour the collapse of dense gas structures and thus the formation of stars. A significant fraction of the total star formation occurs in the tidal debris in Milgromiam gravitation, while such activity is negligible in the Newtonian case. The Milgromian models thus lead to significantly more sub-structures in the tidal tails than the Newtonian models. \item The resulting star formation activity is thus significantly more extended in space and time in Milgromian than in Newtonian gravity. \end{itemize} The results presented here originate from a handful of orbital configurations and for only one galaxy model. Generalizing our conclusions to other systems, over the broad range of observed parameters (mass, mass ratio, relative velocity, inclination, spin-orbit coupling etc.) would require a more complete survey of simulations. Such a study is proposed in the Newtonian framework by the {\tt GALMER} project \citep{DiMatteo2007}, where hundreds of configurations are explored. An equivalent in Milgromian gravitation would allow us to broaden our conclusions on the response of Milgromian galaxies to interactions and mergers. In particular, we expect that the SFH of retrograde encounters (i.e. orbital and disc angular momenta being anti-aligned) in Milgromian dynamics would be more similar to their Newtonian counterparts than for the cases we presented here. During the first pericentre passage of a retrograde encounter, the orbital angular momentum is inefficiently transferred to the discs and tidal features are less pronounced and shorter than in prograde cases \citep[see e.g.][]{Duc2013}. As a result, a large gas reservoir remains available at the late stages of the interaction, in particular at coalescence, where cloud-cloud collisions and nuclear inflows can efficiently trigger an intense episode of star formation. In that case, the SFH would only yield one major peak at coalescence. The main differences between Newtonian and Milgromian models noted above in term of space and time extent of star formation during the separation phase might not be found in a retrograde encounter. Our results show that the observational detection or the absence of a sustained star formation activity in interacting galaxies, in particular over large volumes in tidal debris, would provide a strong hint on the nature of gravitation on galactic scales, as a complement to the study of rotation curves for isolated discs. Even when focussing on the merger remnant only, we show that the Milgromian paradigm favours a long and approximately continuous episode of star formation, starting at the first encounter and ending a few $100 \Myr$ after coalescence. Oppositely, the Newtonian formalism rather supports distinct bursts of star formation at high efficiency, associated with the close encounters \citep[see also][]{DiMatteo2007}. Providing that observational techniques would be able to tell apart these two star formation histories with sufficient confidence in the required age range \citep[$\approx 200 \Myr \mh 1 \Gyr$, see][on the difficulties of dating post-starburst episodes]{Lancon2001, Maraston2001, Maraston2004, Simones2014}, such interacting systems could be used as clear diagnostics to test different paradigms for gravitation on galactic scales. | 16 | 9 | 1609.04407 |
1609 | 1609.06919_arXiv.txt | The original ALHAMBRA catalogue contained over 400,000 galaxies selected using a synthetic F814W image, to the magnitude limit AB(F814W)$\approx$24.5. Given the photometric redshift depth of the ALHAMBRA multiband data (<z>=0.86) and the approximately $I$-band selection, there is a noticeable bias against red objects at moderate redshift. We avoid this bias by creating a new catalogue selected in the $K_s$ band. This newly obtained catalogue is certainly shallower in terms of apparent magnitude, but deeper in terms of redshift, with a significant population of red objects at $z>1$. We select objects using the $K_s$ band images, which reach an approximate AB magnitude limit $K_s \approx 22$. We generate masks and derive completeness functions to characterize the sample. We have tested the quality of the photometry and photometric redshifts using both internal and external checks. Our final catalogue includes $\approx 95,000$ sources down to $K_s \approx 22$, with a significant tail towards high redshift. We have checked that there is a large sample of objects with spectral energy distributions that correspond to that of massive, passively evolving galaxies at $z > 1$, reaching as far as $z \approx 2.5$. We have tested the possibility of combining our data with deep infrared observations at longer wavelengths, particularly Spitzer IRAC data. | Astronomical surveys are one of the key elements in the advancement of our knowledge of celestial objects. From the earliest times astronomers have charted stars and observed their basic properties, namely their positions and apparent brightnesses. This task increased exponentially in complexity over the last centuries with the successive arrivals of the telescope, the photographic plate and the electronic detector. In our time some of the most successful astronomical surveys have aimed at covering ever larger fractions of the phase space that includes area in the sky, photometric depth and spectral information. For the moment being (and in any foreseeable future) no project will cover satisfactorily and simultaneously all of those dimensions. For example, the Sloan Digital Sky Survey \citep[SDSS,][]{2000AJ....120.1579Y} and the Two Degree Field Galaxy Redshift Survey \citep[2dFGRS,][]{2001MNRAS.328.1039C} have obtained spectral information for $\sim 10^5 \mbox{--} 10^6$ objects each, by observing large areas (approximately 1/4 of the whole sky) down to a relatively shallow limit (apparent magnitudes $AB \approx 19$). Their photometric counterparts cover areas in the sky of the same size, but reach ten times deeper, out to a typical magnitude $AB \approx 21 \mbox{--} 22$. At the other end of survey space, deep surveys like the Hubble Deep Fields \citep{2000ARA&A..38..667F} cover tiny areas of the sky (of the order of $10^{-3}$ square degrees or even less) but do include spectroscopy out to $AB \approx 25 \mbox{--} 26$ and multi-band photometry out to $AB \approx 28$ and even deeper. A different "axis" defining cosmic surveys is that of spectral completeness. In the most basic end, early surveys like the Palomar Observatory Sky Survey \citep[POSS, ][]{1963PASP...75..488M,1991PASP..103..661R}, included only photometric information in two different bands (\textit{i.e.} one colour) for each object. In the opposite end, spectroscopic surveys include a full spectrum for each target, with all that implies in terms of information content regarding measurements of redshift, star formation history, mass, metallicity, etc. Since the advent of the Hubble Deep Fields \citep{2000ARA&A..38..667F} and other surveys at the end of the last century it has become commonplace to obtain images through multiple filters both in the optical and the near infrared in order to measure at least some spectral properties of the targets, which should allow for basic estimation of some of the physical quantities that would otherwise need a full spectral analysis. The use of photometric redshift techniques has grown and become standard based on this kind of studies \citep{1999ApJ...513...34F,2000ApJ...536..571B,2000A&A...363..476B}. Over the last few years some surveys have been explicitly designed having these techniques in mind (COMBO-17, \citealt{2003A&A...401...73W}; ALHAMBRA, \citealt{moles2008}) and have proved the case for even larger surveys with multiple medium-band filter images (J-PAS, \citealt{2014arXiv1403.5237B}). Early-type galaxies dominate the bright end of the luminosity function at low and moderate redshifts \citep{1997ApJ...475..494L}, in particular they include the most massive galaxies that inhabit the largest overdensities in those epochs. They represent the most massive and evolved objects in the second half of the life of the Universe, and their study is basic to understand how star formation proceeded and its interrelations with many other cosmic processes: black hole formation and evolution, galaxy clustering and the formation of large-scale structures, galactic interactions and mergers, and the AGN phenomenon \citep[][ and references therein]{2014ARA&A..52..589H}. Due to their intrinsically red colours, early-type galaxies are selected against in magnitude-limited surveys selected at optical wavelengths at all those redshifts where the Balmer break and associated absorption features around $\lambda = 4000$ \AA\ are redshifted into the detection band and redwards of it. Over the last years the development of several surveys that detect objects in near infrared (NIR) bands has significantly helped in the analysis of the evolution of early-type galaxies at moderate and high redshift, e.g. the Newfirm Medium Band Survey \citep[NMBS, ][]{2011ApJ...735...86W}, UKIDSS-Ultra Deep Survey \citep{2007MNRAS.379.1599L}, WIRCam Deep Survey \citep[WIRCDS, ][]{2012A&A...545A..23B}, and Ultra VISTA \citep{2012A&A...544A.156M,2013ApJS..206....8M}. \begin{table} \centering \caption{Comparison with other photometric $K$-band selected surveys.}\label{surveycomp} \begin{tabular}{lcc} \hline Survey & Area & AB Magnitude \\ & & ($5\sigma$ Limit) \\ \hline MUSYC& 0.015 $\mathrm{deg^2}$ & $K_s \approx 22.5 $ \\ NMBS& 0.44 $\mathrm{deg^2}$ & $K \approx 24.2 $\\ UKIDSS-UDS& 0.77 $\mathrm{deg^2}$ & $K \approx 24.6 $\\ WIRCDS & 2.03 $\mathrm{deg^2}$& $K_s\approx 24.0 $\\ UVISTA & 1.50 $\mathrm{deg^2}$ & $K_s\approx 23.8$ \\ ALHAMBRA $K_s$-band & 2.47 $\mathrm{deg^2}$ & $K_s \approx 21.5$\\ \hline \end{tabular} \end{table} In the particular case of the ALHAMBRA survey, where detection is performed over a synthetic image that emulates the Hubble Space Telescope F814W filter, this selection effect that creates a bias against red galaxies begins to be noticeable at $z\approx 0.8$, and is dominant at $z\geq 1.1$, as has already been noticed by \citet{2014MNRAS.441.1783A}. A typical early-type spectral energy distribution at $z\approx0.8$ has a colour $(I-K_s) \approx 1.8$, whereas the same galaxy at $z\approx 1.4$ shows $(I-K_s)\approx 3.1$, and reaches $(I-K_s) \geq 4.5$ at redshift $z=2$. This means that, even if the optical detection image is, as is the case in ALHAMBRA, deeper than the corresponding $K_s$ band, at least some of the incompleteness produced by the selection effects can be avoided by using the $K_s$ band to provide the detection image. In this work we present a new $K_s$-band selected catalogue of galaxies in the ALHAMBRA survey that has been compiled in order to partially overcome the selection bias described above. With this catalogue we will be able to extend some of the works that have already been performed with the ALHAMBRA data to higher redshifts $z>1$, namely: calculation of the general and type-segregated correlation functions \citep{2014MNRAS.441.1783A,2016ApJ...818..174H}, search for groups and clusters \citep{2015MNRAS.452..549A}, analysis of the clustering signal encoded in the cosmic variance \citep{2015A&A...582A..16L}, stellar populations of galaxies \citep{2015A&A...582A..14D}, detection of high-redshift galaxies \citep{2015A&A...576A..25V}, and possibly also the morphological analysis of some of the brightest targets \citep{2013MNRAS.435.3444P}. We will also use this catalogue to produce large, well-defined, samples of massive galaxies at intermediate redshifts over the redshift range $1<z<2.5$, as well as Balmer jump selected galaxies at $z>1$ \citep[similarly to what was done in ][]{2016A&A...588A.132T}. The organization of the paper is as follows: we briefly introduce the ALHAMBRA survey in \S2, and describe the construction of the catalogue in \S3. Section 4 presents the catalogue and its most basic properties. In \S5 we discuss some of the immediate applications of the catalogue, with particular attention to how its use will be important in order to complete (either in terms of redshift or in terms of galaxy types) some of the analyses that have already been published based on the original ALHAMBRA catalogue. Finally \S6 contains our conclusions. In what follows all magnitudes are given in the AB system \citep{1983ApJ...266..713O}, and we use a cosmology with $H_0=100\ h \mathrm{\ km\ s}^{-1}\ \mathrm{Mpc}^{-1},\ \Omega_\mathrm{M}=0.28,\ \Omega_\Lambda=0.72$ \citep{2015arXiv150201589P}. | We have presented in this paper the photometric and photometric redshift catalogue of sources detected in the ALHAMBRA $K_s$-band images. The catalogue includes photometry for 94,182 sources distributed over seven fields, covering a total area of 2.47 deg$^2$. This catalogue is different from the original ALHAMBRA catalogue presented in \citet{2014MNRAS.441.2891M} because that sample was selected based on a synthetic F814W image, similar to an $I$-band selection. Such a selection is biased against intrinsically red galaxies at redshift $z \gtrsim 1$, an effect that became noticeable in several of the recent works based on the ALHAMBRA survey. This issue sparked our interest in producing a new catalogue where this bias would be avoided by selecting in the reddest band available. Source detection and photometry was performed using SExtractor in dual mode. We estimated the photometric errors using the method presented by \citet{2003AJ....125.1107L}, and used an adapted version of the masks created in \cite{2014MNRAS.441.1783A} to define the survey window. Star-galaxy separation was performed using a colour-colour diagram, and tested with the SEDs of the NGSL stellar library. We calculated detailed completeness functions for every pointing using the deeper UltraVISTA catalogue of the ALHAMBRA-4/COSMOS field as reference. We applied this completeness functions to extend our magnitude limit and the number counts to magnitude $K_s \approx 21.9$. Two separate tests were performed to check the photometric accuracy of our catalogue: an internal test against the photometry of the objects common to our catalogue and the original ALHAMBRA catalogue presented in M14, and an external test using the objects common to our catalogue and UltraVISTA. In both cases the cross-catalogue accuracy has been shown to be compatible with that expected from the respective uncertainties, and with no significant bias. We completed our catalogue by running the BPZ2.0 code over our sample, including the zeropoint photometric recalibration option that uses a spectroscopic redshift sample to refine, at the same time, both the photometry and the photometric redshift accuracy. Using a spectroscopic redshift sample with 3736 galaxies, and the normalized median absolute deviation (NMAD) as a estimator of the accuracy of our results, we obtain $\sigma_{\rm NMAD}=0.011$, and a catastrophic error rate $\eta_1 \sim 1.3\%$, both comparable to the ones obtained by \cite{2014MNRAS.441.2891M}. We performed a second comparison, in this case with the photometric redshifts in the ALHAMBRA F814W-selected catalogue. This comparison yields $\sigma_{\rm NMAD}=0.009$ with a catastrophic error rate $\eta_1 \sim 0.58\%$. As expected, because of the motivation of our work, the photometric redshift distribution segregated by galaxy type shows that many of the new $K_s$-selected sources fill the dearth of early-type galaxies in the F814W-selected sample at $z \gtrsim 1$. We will present, in forthcoming works, a detailed analysis of this population, including the combination of our data with catalogues covering other wavelengths, in particular the {\it Spitzer}-IRAC public catalogues that overlap several of the ALHAMBRA fields. We have presented in this work some examples of how these extra photometric bands add to the information content of our catalogue. | 16 | 9 | 1609.06919 |
1609 | 1609.01014_arXiv.txt | We present a measurement of the spatial clustering of massive compact galaxies at $1.2\le z \le 3$ in CANDELS/3D-HST fields. We obtain the correlation length for compact quiescent galaxies (cQGs) at $z\sim1.6$ of $r_{0}=7.1_{-2.6}^{+2.3}\ h^{-1}Mpc$ and compact star forming galaxies (cSFGs) at $z\sim2.5$ of $r_{0}=7.7_{-2.9}^{+2.7}\ h^{-1}Mpc$ assuming a power-law slope $\gamma =1.8$. The characteristic dark matter halo masses $M_H$ of cQGs at $z\sim1.6$ and cSFGs at $z\sim2.5$ are $\sim7.1\times 10^{12}\ h^{-1} M_\odot$ and $\sim4.4\times10^{12}\ h^{-1} M_\odot$, respectively. Our clustering result suggests that cQGs at $z\sim1.6$ are possibly the progenitors of local luminous ETGs and the descendants of cSFGs and SMGs at $z>2$. Thus an evolutionary connection involving SMGs, cSFGs, QSOs, cQGs and local luminous ETGs has been indicated by our clustering result. | Massive ($M_\star\ge 10^{10}M_\odot$), quiescent galaxies (QGs) at high redshift ($z\sim 2$) have been found to have $3-5$ times smaller effective radii than their local counterparts (e.g., Daddi et al. 2005; Trujillo et al. 2006; van der Wel et al.2008; van Dokkum et al. 2008; Damjanov et al. 2009; Newman et al. 2012; Szomoru et al. 2012; Zirm et al. 2012; Fan et al. 2013a, b). Since massive compact quiescent galaxies (thereafter cQGs) in the local Universe are rare (e.g., Poggianti et al. 2013), a significant structural evolution has been required. Therefore, there raised two questions: (1) how do these cQGs evolve into local luminous early-type galaxies (ETGs) with larger size? and (2) how did these cQGs form at higher redshift? There are two physical mechanisms which have been proposed to explain the observed structural evolution of cQGs at $z\ge 1$. One is dissipationless (dry) minor mergers (Naab et al. 2009; Oser et al. 2012; Oogi et al. 2016). The other is "puff-up" due to the gas mass loss by AGN (Fan et al. 2008, 2010) or supernova feedback (Damjanov et al. 2009). The recent evidence has shown the inside-out growth of massive cQGs at $z>2$, which indicates that dry minor mergers may be the key driver of structural evolution (Patel et al. 2013). However, whether dry minor mergers are sufficient for the size increase, especially at $z\ge 1.5$, is still under debate (Newman et al. 2012; Belli et al. 2014). Possible mechanisms for the formation of cQGs include gas rich mergers (Hopkins et al. 2008), violent disk instability fed by cold stream, or both (Ceverino et al. 2015). Whatever mechanism governs the formation of cQGs, their precursors should be expected to experience a compact and active phase: compact star forming galaxies (cSFGs) or compact starburst galaxies (i.e, sub-millimeter galaxies, SMGs). Barro et al. (2013) found a population of massive cSFGs at $z\sim2$. They proposed that cSFGs could be the progenitors of cQGs at lower redshift, suggested by the comparison of their masses, sizes, and number densities. Toft et al. (2014) showed that SMGs at $z>3$ are consistent with being the progenitors of $z\sim2$ cQGs by matching their formation redshifts and their distributions of sizes, stellar masses, and internal velocities. They suggested a direct evolutionary connection between SMGs, through compact quiescent galaxies to local ETGs. In this evolutionary scenario, star formation quenching has been proposed to be either due to gas exhaustion or quasar (QSO) feedback. The latter is essential in many models of the evolution of massive galaxies (e.g., Granato et al. 2004; Hopkins et al. 2010). In this paper, we analyze the clustering properties of cQGs and cSFGs at $1.2\le z\le 3$ and compare them to other populations: high-$z$ QSOs, SMGs and local ETGs in order to investigate the possible connection between cQGs, cSFGs, SMGs, QSOs and local ETGs. All our data come from the CANDELS and 3D-HST programs (Grogin et al. 2011; Koekemoer et al. 2011; Skelton et al. 2014). The CANDELS/3D-HST programs have provided WFC3 and ACS images, spectroscopy and photometry covering $\approx 900$ arcmin$^{2}$ in five fields: AEGIS, COSMOS, GOODS-North, GOODS-South and the UDS. The large survey areas and the depth of the HST WFC3 camera enable us to make more accurate clustering measurement than in narrower, shallower fields. We emphasize that it is essential to use the high-resolution HST WFC3 imaging to investigate the compact structure of massive galaxies at high redshift. Throughout this paper, we adopt a flat cosmology (see Komatsu et al. 2011) with $\Omega_{M} = 0.3,\ \Omega_{\Lambda} = 0.7,\ H_{0} = 70\ km s^{-1} Mpc^{-1}$. We assume a normalisation for the matter power spectrum of $\sigma_{8} = 0.84$. All quoted uncertainties are 1 $\sigma$ (68\% confidence). All magnitudes are in the AB magnitude system. | In this paper, we measure the cross-correlation between massive compact galaxies and comparison galaxies at $1.2\le z\le 3$ in CANDELS/3D-HST fields. We obtain the correlation length for cQGs at $z\sim1.6$ of $r_{0}=7.1_{-2.6}^{+2.3} h^{-1}Mpc$ and cSFGs at $z\sim2.5$ of $r_{0}=7.7_{-2.9}^{+2.7} h^{-1}Mpc$. The characteristic DM halo masses of cQGs at $z\sim1.6$ and cSFGs at $z\sim2.5$ are $\sim7.1\times 10^{12}\ h^{-1} M_\odot$ and $\sim4.4\times10^{12}\ h^{-1} M_\odot$, respectively. The observed clustering suggests that cSFGs and SMGs at $z>2$ could be the progenitors of cQGs at $z<2$. We estimate both the co-moving space densities and the corresponding lifetimes of cSFGs/cQGs and find that cSFGs have similarly short lifetime as SMGs. Our clustering results support such an evolutionary sequence involving compact starbursts (SMGs or cSFGs), cQGs and ETGs (see also Toft et al. 2014). | 16 | 9 | 1609.01014 |
1609 | 1609.08135_arXiv.txt | The precision of photometric and spectroscopic observations has been systematically improved in the last decade, mostly thanks to space-borne photometric missions and ground-based spectrographs dedicated to finding exoplanets. The field of eclipsing binary stars strongly benefited from this development. Eclipsing binaries serve as critical tools for determining fundamental stellar properties (masses, radii, temperatures and luminosities), yet the models are not capable of reproducing observed data well either because of the missing physics or because of insufficient precision. This led to a predicament where radiative and dynamical effects, insofar buried in noise, started showing up routinely in the data, but were not accounted for in the models. PHOEBE (PHysics Of Eclipsing BinariEs; {\tt http://phoebe-project.org}) is an open source modeling code for computing theoretical light and radial velocity curves that addresses both problems by incorporating missing physics and by increasing the computational fidelity. \rev{In particular, we discuss triangulation as a superior surface discretization algorithm, meshing of rotating single stars, light time travel effect, advanced phase computation, volume conservation in eccentric orbits, and improved computation of local intensity across the stellar surfaces that includes photon-weighted mode, enhanced limb darkening treatment, better reflection treatment and Doppler boosting.} Here we present the concepts on which PHOEBE is built on and proofs of concept that demonstrate the increased model \rev{fidelity}. | Eclipsing binary stars (EBs) serve as cornerstones of stellar astrophysics. Their uniquely important role in determining fundamental stellar parameters \citep{popper1980}, distances \citep{pietrzynski2013} and providing rigorous tests for stellar evolution models \citep{torres2010} have been widely appreciated. While the underlying principles that govern the modeling of EBs are simple (Newtonian mechanics and straight-forward geometry considerations), a plethora of complications arises from subtler effects: surface distortions, intensity variations due to gravity darkening, limb darkening, reflection and surface prominences. As the precision of the data increases, so does the number of subtleties that a reliable model needs to take into account. This decade has provided us with a range of ground-breaking surveys and missions: {\sl MOST} \citep{walker2003}, {\sl CoRoT} \citep{baglin2003}, {\sl Kepler} \citep{borucki2010}, {\sl Pan-Starrs} \citep{kaiser2002}, {\sl Gaia} \citep{debruijne2012} and {\sl LSST} \citep{tyson2002}, to name just a few. These yielded a vast number of EBs (\citealt{prsa2011a} estimated $\sim$7 million from {\sl LSST} alone), thousands to unprecedented quality (down to $\sim$20 ppm for {\sl Kepler} light curves; cf.~Fig.~\ref{fig:keplerlcs} for several examples of attained precision). We are seeing phenomena that have been up until recently only theorized, but now they are appearing routinely: modern observations provide us with a glimpse into the micromagnitude scale. With such a tremendous boost in both quantity and quality, the tools we use to reduce, analyze and interpret data need to be able to cope with this literal firehose of observations. Yet from all the data currently available, \citet{torres2010} find only $\sim$100 EBs with the uncertainties in the masses and radii smaller than 3\%. For such benchmark objects, this number is several orders of magnitude too low. \begin{figure}[t!] \includegraphics[width=\textwidth]{kepler_lc.pdf} \\ \caption{ \label{fig:keplerlcs} {\sl Kepler} time series (left) and phased light curves (right) of KIC 5513861 (P=1.51012-d, detached EB), KIC 8074045 (P=0.53638-d, semi-detached EB), and KIC 3127873 (P=0.67146-d, contact EB), top-to-bottom. } \end{figure} Having superb quality data in abundance has clear repercussions on our modeling capability. For the first time we observe astrophysical objects in a near-uninterrupted regime, with uncertainties that dip below 20\,ppm. State-of-the-art models such as the renowned Wilson-Devinney code \citep{wilson1971,wilson1979,wilson2008,wilson2014}, ELC \citep{orosz2000}, and PHOEBE \rev{\citep[paper I]{prsa2005}} are showing systematics in the derived values of fundamental stellar parameters. This is partly because of embedded approximations, partly because of the inconsistent use of physical units and constants (cf.~the IAU 2015 resolution B3; \citealt{prsa2016}) and partly because of the missing and/or inadequate physics built into these models. The residuals \rev{of the best-fit models and observed data} are no longer Gaussian, and we cannot assume that these effects are buried in noise; rather, we need to account for their signatures in light and radial velocity curves explicitly. Furthermore, the codes provide minimization algorithms that fit model curves to the data. EB data fitting is a highly non-linear problem that suffers from degenerate solutions: the \emph{right} combination of the \emph{wrong} parameters can often fit the observed data as well as the actual solution. The algorithms currently in use, namely Differential Corrections (DC), \citet{powell1964}'s direction set method, \citet{nelder1965}'s Simplex method (NMS), and genetic algorithms \citep{attia2009}, have all met with success, but cannot be run robustly without experienced human intervention, making the tools fully manual. In this paper we discuss several deficiencies of our models and present advancements in improving the reliability of binary star solutions. \rev{The improvements result from including missing instrumental and astrophysical phenomena in the model explicitly, thus reducing systematical errors in correlated parameters that no longer need to compensate for the missing aspects in the code.} The paper is accompanied by the release of the new version of the open source modeling code \phoebe \rev{2.0, available from {\tt http://phoebe-project.org}, along with extensive documentation and tutorials.} The layout of the paper is as follows. In Section \ref{s:revision} we provide background on the \phoebe project; Section \ref{s:discretization} introduces the new triangulation scheme for surface discretization; Section \ref{s:dynamics} focuses on dynamical and temporal aspects of the model; Section \ref{s:local_quantities} gives the rationale for computing all local quantities across the mesh; in Section \ref{s:limitations} we discuss current limitations and provide concluding remarks. | \label{s:limitations} This paper presented the first step towards the increased fidelity of computed eclipsing binary models. \rev{Triangulation replaces trapezoidal meshing, which provides a robust surface coverage by near-equilateral, near-isometric surface elements and the ability to discretize any 3-D body irrespective of its shape, whereas trapezoidal meshing relied on close-to-spherical shapes. To attain high mesh precision with small number of triangles, we offset mesh elements so that the discretized volume is identical to the analytical equipotential volume. We introduce a new type of body: rotating stars, which eliminates lengthy Roche computations when they are not necessary for the problem at hand. The treatment of dynamics has been updated by including light travel time delay for all bodies in a system, such that time is measured w.r.t.~the barycenter of the system. Time-to-phase conversion accounts for all temporally changing quantities explicitly. In case of eccentric orbits, we provide an argument of why volume conservation is appropriate and we propose a test based on the eccentric ellipsoidal variable systems known as heartbeat stars. Passband intensities are based on the \citet{castelli2004} model atmospheres and values for both energy-weighted and photon-weighted intensities are stored in lookup tables for fast operation. The old treatment of limb darkening caused severe artifacts near the limb of the star because of undersampling; this has been mitigated by providing the option to use actual intensity values as a function of $\mu$ instead of relying on 1-, 2- or 4-parameter models. Reflection effect has been restated to include Lambertian scattering, which paves the way for future inclusion of heat redistribution across the irradiated surface. Doppler boosting (comprised of Doppler shift of the spectrum, time dilation and relativistic beaming) is now fully taken into account. Local radial velocities are now corrected for the gravitational redshift. Eclipse and horizon computation is significantly improved by replacing the old Fourier-based approach with an algebraic scheme that is as accurate as the mesh itself. Finite integration time acts as a smoothing filter on light curves, which is now supported on the synthetic end as well. Finally, flux computation in absolute units has been tested against the Earth-Sun system and the test demonstrates that the fidelity has indeed increased.} There are several limitations of \phoebe 2.0 that we aim to address within the ongoing project: \begin{description} \item[Integrated flux dependence on the number of surface elements.] \rev{Ideally, the computed flux should not depend on the computational aspects such as the fineness of the mesh, but this is an inherent limitation for all numerical schemes. Flux values converge to the analytical value linearly ($\propto 1/N$ where $N$ is the number of surface elements), which is quite slow and is further impacted by other second-order effects, most notably limb darkening. The exact convergence properties are difficult to assess because of the non-linearity of the parameter space, but tests show that for a moderate number of triangles ($\sim 5000$) the attained relative accuracy is of the order of $10^{-3}$. The impact on \emph{relative} light curve shapes (i.e.~flux ratios) is much smaller because all phase points are affected equivalently, but whenever flux values are sought, their dependence on the mesh size should be kept in mind. We are considering acceleration schemes that model the dependence of integrated flux on the number of surface elements by evaluating it for several mesh sizes and then extrapolating it to $N \to \infty$, but no robust solutions have been worked out yet.} \item[Contact binaries.] The trapezoidal model to modeling contact binaries is likely suitable only for systems in thermal equilibrium, since it introduces unphysical discontinuities in the neck area of systems with unequal temperatures (cf.~Section \ref{s:local_quantities}). Even for systems in thermal equilibrium, the atmosphere tables used to derive the normal emergent intensities are computed assuming single, spherical stars, which introduces additional inconsistencies in the overall treatment of contact systems. We are currently working on the implementation of a more feasible model of contact binaries by studying the radiative transfer in the common envelope based on the structure of the whole star. The structure and radiative transfer computations are performed independently which allows for the exploitation of a wide range of hydro-thermodynamical models and the testing of the surface intensity distribution they give rise to. The full description of this novel model will be the topic of \rev{the paper by} Kochoska et al., in preparation. \item[Misaligned binaries.] \phoebe is currently based on the Roche geometry, which assumes perfect alignment between the orbital and the rotational axes and handles each component's tidal distortion as due to the companion's point mass. However, careful ground-based studies have been able to discern that the components of close binaries in a number of systems show misaligned rotational and orbital axes \citep{albrecht2011}. Observationally this is most easily discerned via the Rossiter-McLaughlin effect. The misalignment has also been found for a number of ''hot Jupiters`` \citep{hebrard2008}. Consequently, both EBs and transiting exoplanets bring into question the initial conditions for the formation of both binaries and planetary systems. The generalized Roche potential for binary systems where the stellar rotation is not aligned with the orbital revolution is fundamentally different to the currently implemented aligned potential. The modifications to the potential have been derived by \citet{limber1963}, \citet{kruszewski1967} and \citet{kopal1978}. The properties of the critical equipotential lobe and Lagrangian points for circular orbits have been studied in detail by \citet{avni1982}. We are implementing a fully numerical determination of equipotential surfaces in misaligned binaries. These determine the shapes and radiative properties of components in binary stars and will be presented by Horvat et al., in preparation. \item[Graphical User Interface.] It might come as a surprise that the current version of \phoebe does not come with a standalone graphical user interface. We are working on a web-based user interface that is substantially more flexible than the original interface of \phoebe 1. Its main characteristic is the combination of an interactive shell and a graphical interface that can be either run locally or remotely. The interface will be discussed by Conroy et al., in preparation. \item[Fitting.] The absence of the fitting interface is by no means an oversight. Deterministic minimizing programs used in the past, most notably Differential Corrections (DC; \citealt{wilson1971}), Nelder \& Mead's Simplex method (NMS; \citealt{prsa2005b}) and Powell's direction set method \citep{prsa2007}, are not suited for the high fidelity demand of modern data, and do not lend themselves to automation that is necessary to address the firehose of data coming our way from surveys such as Gaia or LSST. We consider \phoebe models as \emph{likelihood functions} that, accompanied with an appropriate noise model, should feed probabilistic samplers, such as the Markov Chain Monte Carlo methods. The High Perfomance Computing (HPC) wrapper for \phoebe is available on \phoebe website and should be preferred to deterministic minimizers. The rationale for fitting will be discussed by Pr\v sa et al., in preparation. \item[Computational time cost.] The computational infrastructure of \phoebe is implemented in the low-level C language for speed, and the interface part is written in the high-level python language that is inherently slow. A number of causal computations are still linked through python, which causes slowdown of the execution time. Yet the dominant source of slowdown is increased model fidelity. Shortcuts taken before are no longer in effect and, even though effects can be turned off by the user, the impact on the overall runtime is significant. That is why the preferred mode of deployment of \phoebe is on HPC clusters. \end{description} In conclusion, we respectfully invite the community to provide us with any feedback, criticism and suggested improvements to the model, and to help us critically evaluate model robustness in different operational regimes. \rev{We remind that \phoebe is released as open source under the General Public License, is free and will always remain free, so anyone interested to join in our efforts to provide a robust, general modeling code to the community is welcome to join us.} | 16 | 9 | 1609.08135 |
1609 | 1609.04893_arXiv.txt | The Milky Way Galaxy glows with the soft gamma ray emission resulting from the annihilation of $\sim 5 \times 10^{43}$ electron-positron pairs every second. The origin of this vast quantity of antimatter and the peculiar morphology of the 511keV gamma ray line resulting from this annihilation have been the subject of debate for almost half a century. Most obvious positron sources are associated with star forming regions and cannot explain the rate of positron annihilation in the Galactic bulge, which last saw star formation some $10\,\mathrm{Gyr}$ ago, or else violate stringent constraints on the positron injection energy. Radioactive decay of elements formed in core collapse supernovae (CCSNe) and normal Type Ia supernovae (SNe Ia) could supply positrons matching the injection energy constraints but the distribution of such potential sources does not replicate the required morphology. We show that a single class of peculiar thermonuclear supernova - SN1991bg-like supernovae (SNe 91bg) - can supply the number and distribution of positrons we see annihilating in the Galaxy through the decay of $^{44}$Ti synthesised in these events. Such $^{44}$Ti production simultaneously addresses the observed abundance of $^{44}$Ca, the $^{44}$Ti decay product, in solar system material. | The origin of the $\sim5 \times 10^{43}$ positrons that annihilate every second in the Galaxy has been subject to debate over the past almost half a century (see \cite{Prantzos+2011} for a review). The origin of these positrons is difficult to understand as most obvious positron sources can be ruled out either due to the stringent constraints on the injection energy of the positrons ($<3\,\mathrm{MeV}$, \cite{Beacom+2006}), the morphology (the bulge glows almost as brightly as the Galactic disk with positron annihilation radiation) and the implied, absolute rate of positron production. \\ The most recent analysis of positron annihilation radiation shows the morphology of the emitted gamma rays correlates with the stellar populations of the Galaxy with components seemingly tracing the Galactic bulge, Galactic disk (at low surface brightness) and the Galactic nuclear bulge (\cite{Siegert+2015}). The most plausible Galactic positron source matching injection energy constraints is the radioactive decay of $\beta^+$ unstable nuclei synthesised by supernovae (\cite{Prantzos+2011}). Can a transient event explain the observed morphology? \\ | 16 | 9 | 1609.04893 |
|
1609 | 1609.07538_arXiv.txt | Analysis of multiwavelength light curves and VLBI observations of the blazar S5~0716+71 provides different and sometimes contradictory results for periods of long-term variability and kinematics of the parsec-scale jet components. We propose a model that consistently explains observable properties. Obtained results and conclusions: 1) geometrical parameters of helix, pitch-angle, and lower limit on jet components Lorentz-factor, distance from the jet vertex to VLBA core at 15~GHz, 2) distinction in duration of long-term variability periods for position angle of the inner jet ($\text{PA}_\text{in}$), optical and radio emission is caused by non-accelerated non-ballistic motion of jet component, 3) because different jet parts are responsible for observed flux and $\text{PA}_\text{in}$, mutual orientation of these jet parts relative to a plane containing the helix axis and the line of sight influences both the value of correlation coefficient between flux and $\text{PA}_\text{in}$ and the time interval within which this value is observed, 4) non-ballistic motion of jet components, firstly, gives simple explanation for difference apparent speeds of jet features in the inner and outer jet and secondly, forms conditions for existence of high apparent speed up to $37c$. | The BL Lac object S5~0716+71 [$z=0.31$ \citep{Nil08}, $0.2315<z<0.3407$ \citep{Danf13}, $z=0.26$ \citep{Wagner96}, $z\geq0.52$ \citep{STF05}] has been intensively observed in the radio, optical, X-ray, and $\gamma$-ray domains \citep[e.g.][]{Liao14}. The source is variable on time scales from day \citep[]{MontMM06, Stalin06, Wu07, Vol09, Poon09} to a few years \citep[]{Rait03, Nesci05, Dai13}. Nearly all observed blazar emission is produced in its relativistic jet. Therefore, variability of the blazar can be caused not only by physical processes occurring in the jet, for instance, passage of a shock wave \citep[]{mg85} but also by jet Doppler-factor change \citep[see, e.g., ][]{CamKrock92, Abr00}. For more than ten years S5~0716+71 has been observed with VLBI \citep[]{Bach05, Brit09, Rani15} and within the MOJAVE program \citep[e.g.][]{List13}. From these data, the inner jet position angle $\text{PA}_\text{in}$ was found to vary on a decadal time scale \citep[]{Bach05, List13}. It can be easily explained if the jet has a helical structure. Taking it into account, we can derive constraints on geometric parameters of the jet (Section~\ref{sec:geom}). If the jet has a helical structure then periodic variations of the angle between the velocity vector of jet component and the line of sight lead to long-term periodic variability of observed flux. But analysis of long-term observations in the radio and optical bands gives variability periods of about 3 and 6 years \citep[]{Rait03, Liu12}, respectively. These values being differ between themselves are distinct from the 10yr variability period of $\text{PA}_\text{in}$. In this Paper we found connection between these periods (Section~\ref{sec:period}) and provided explanations for alternation of intervals of strong positive and negative correlation between $\gamma$-ray flux of the blazar S5~0716+71 and $\text{PA}_\text{in}$ detected by \citet{Rani14} (Section~\ref{sec:corel}) and for contradictory results of kinematic analysis carried out in papers \citep[]{Rast11, Rani15} and the high apparent speed of jet components \citep[]{Rani15} (Section~\ref{sec:kin}). Discussion of the obtained results and conclusions are given in Sections~\ref{sec:disc} and \ref{sec:conc}, respectively. | \label{sec:conc} We suggest the helical structure of the blazar S5~0716+71 parsec-scale jet. It allows us: \begin{enumerate} \item to explain the observed periodic variations of the inner jet position angle $\text{PA}_\text{in}$, the long-term flux variability in the radio and optical bands; \item to explain different periods of long-term variability; \item to interpret alternation of time intervals of positive and negative correlation between the $\gamma$-ray flux and $\text{PA}_\text{in}$; \item to reconcile contradictory results of kinematic studies of the S5~0716+71 jet components; \item to explain the high apparent speed of jet components. \end{enumerate} Within the considered jet model the following parameters and properties were found: \begin{enumerate} \item the half-opening angle $\xi\!=\! 1^\circ$ of the imaginary cone and angle between the cone axis and the line of sight $\theta_0\!=\!5.3^\circ$. This result does not depend on the motion nature of jet components; \item distance from cone's apex to he VLBA core is about 8.2~pc and to the optical emission production region is 4.7~pc; \item non-ballistic motion of jet components with the pitch angle $p=5.5^\circ$. This gives greater interval of possible values of the $\theta$ than for ballistic motion where $\theta\!=\!\theta_0\pm\xi$; \item low constraint on the physical speed of jet components is $\beta>0.999$; \item it was concluded that the pitch angle differs from the angle $\psi$ between the tangent to the helix and the cone generator at the given point. \end{enumerate} Advantages of the helical jet model: \begin{enumerate} \item it does not depend on mechanisms forming the helix; \item it self-consistently explains connection between long-term variability of blazar S5~0716+71 and kinematics of its parsec-scale jet; \item the model may be applied to other blazars having both similar and different properties as compared to S5~0716+71. For example, a constant value of $\text{PA}_\text{in}$ can be caused by either $\psi\!\!=\!\!p$ or insufficient flux density for detection of jet components which velocity vector makes up larger angle with the line of sight. \end{enumerate} We emphasize that the motion of jet components influences duration of the variability period. In the case of the ballistic jet the variability period increases with decrease of observational frequency when the jet slows down. If components move at some pitch angle $p$ then the period can be either absent at all (when $p=\psi$) or increases with frequency of observation even without a decrease in the physical speed of jet components (when $p\neq\psi$). The helical jet also complicates a correlation of two different observed values and makes it strongly dependent on the jet geometry what hinders the detection of physical connection between these values. The correlation cannot be entirely absent. If it is observed then physical processes make a key contribution to variations of the studied values. | 16 | 9 | 1609.07538 |
1609 | 1609.02401_arXiv.txt | In the present review we discuss the past and present status of the interacting OB-type binary frequency. We critically examine the popular idea that Be-stars and supergiant sgB[e] stars are binary evolutionary products. The effects of rotation on stellar evolution in general, stellar population studies in particular, and the link with binaries will be evaluated. Finally a discussion is presented of massive double compact star binary mergers as possible major sites of chemical enrichment of r-process elements and as the origin of recent aLIGO GW events. | Let us first remark that a statistically significant number of OB-type binary observations is available for the Solar neighborhood only so that conclusions which are based on these observations apply for that region. Whether these conclusions apply for the whole Galaxy, for the whole cosmos is a matter of faith. \subsection{The status before the year 2000} \subsubsection{The primordial O-type binary frequency} By carefully selecting a sample of $\sim$60 O-type stars, Garmany et al. (1980) concluded that $\sim$33\% may be primary of a binary with mass ratio q > 0.2 (q = M$_\mathrm{secondary}$/M$_\mathrm{primary}$) and orbital period P < 100 days. However it was argued by Mason et al. (1998) that it is probable that many more O-type binaries with periods larger than 100 days await to be discovered. At the first massive star conference in 1971 in Buenas Aires, Kuhi (1973) presented a statistical study showing that it is probable that all WR stars are binary components. However, Vanbeveren and Conti (1980) demonstrated that the Kuhi statistics is biased, and that the real WR+OB frequency is no more than 40\%, a binary percentage that has not changed very much over the years till now. To estimate the primordial massive O-type binary frequency, Vanbeveren et al. (1998a) proceeded as follows. A population of massive stars consists of real single stars, un-evolved (=pre-RLOF\footnote{RLOF = Roche lobe overflow}) binaries, evolved (=post-RLOF) binaries, post-supernova rejuvenated binary mass gainers (O+compact companion or disrupted single stars but with binary origin), binary mergers (also single stars but with binary origin), etc. This means that the binary frequency in a region where star formation is continuous in time is smaller than the binary frequency at birth (=primordial binary frequency). Using a detailed binary population code Vanbeveren et al. tried to answer the question: what must be the primordial massive binary frequency f in order to explain the observed results of Garmany et al. and the observed WR binary frequency? The final answer depends on the details of all processes that govern binary evolution but with standard assumptions we concluded that f > 0.7. \subsubsection{The primordial B-type binary frequency} Abt et al. (1990) studied a sample of 109 B2-B5 stars and concluded that 29\% are spectroscopic binaries. Wolff (1978) considered 83 late B-type stars and argued that 24\% are binary with mass ratio q > 0.1 and period P < 100 days. The Bright Star Catalogue (Hoffleit and Warren, 1991) contains 511 B0-B3 stars and after carefully accounting for observational biases Vanbeveren et al. (1998b) promoted a binary frequency $\sim$35\%. Similarly as the O-type star population, the B-type star population consists of pre- and post RLOF stars, real singles but also singles with a binary origin. A similar binary population synthesis exercise as described in Vanbeveren et al. (1998a) for the O-type stars then reveals that in order to explain the $\sim$35\% binary frequency quoted above, the primordial B-type interacting binary frequency should be $\geq$ 50\%. \subsection{The status after the year 2000} A homogeneous sample of 71 O-type stars in 6 Galactic open clusters has been discussed by Sana et al. (2012); 40 (e.g., 40/71 = 56\%) are confirmed spectroscopic binaries. After correction for observational biases the authors conclude that the intrinsic (=primordial) O-type binary frequency $\sim$69\% which is a nice confirmation of the status before the year 2000. Most interestingly, the resulting mass ratio distribution seems to be flat whereas the period distribution is slightly skewed towards small values of P. A radial velocity spectroscopic survey of 250 O-type stars and 540 B-type stars was presented by Chini et al. (2012). If all radial velocity variations are interpreted as being due to binarism, the authors plotted the resulting binary frequencies as function of spectral type (Fig. 3 in Chini et al.). Using the O-type frequencies depicted in that figure and performing a similar binary population synthesis experiment as discussed in section 1.1.1, one arrives at the conclusion that it can not be excluded that all O-type stars are born in binaries. The same is true for the early B-type stars. By considering the whole B-type star set, we obtain an overall B-type primordial binary frequency of 50\% supporting the status before 2000. | We propose the following conclusions that deserve further investigation: \begin{itemize} \item The interacting OB-binary frequency is > 50-70\%. This means that population studies where binaries are ignored may have an academic value but may be far from reality \item At least 20\% of the B-type star population are binary mergers or binary mass gainers but not all are Be-type stars; may be only few Be-type stars are formed via binary interaction \item sgB[e]-stars may be case Br binary mergers \item aLIGO-rates in general, the formation of double BH mergers in particular, depend critically on the physics of the LBV-phenomenon. \end{itemize} | 16 | 9 | 1609.02401 |
1609 | 1609.06708_arXiv.txt | {Quantifying the gas surface density inside the dust cavities and gaps of transition disks is important to establish their origin. } {We seek to constrain the surface density of warm gas in the inner disk of HD~139614, an accreting 9 Myr Herbig Ae star with a (pre-)transition disk exhibiting a dust gap from 2.3$\pm$0.1 to 5.3$\pm$0.3 AU. } {We observed HD~139614 with ESO/VLT CRIRES and obtained high-resolution (R$\sim$90\,000) spectra of CO ro-vibrational emission at 4.7 $\mu$m. We derived constraints on the disk's structure by modeling the CO isotopolog line-profiles, the spectroastrometric signal, and the rotational diagrams using grids of flat Keplerian disk models. } { We detected $\upsilon=1\rightarrow0~^{12}$CO, 2$\rightarrow$1~$^{12}$CO, 1$\rightarrow$0~$^{13}$CO, 1$\rightarrow$0~C$^{18}$O, and 1$\rightarrow$0~C$^{17}$O ro-vibrational lines. Lines are consistent with disk emission and thermal excitation. $^{12}$CO $\upsilon=1\rightarrow0$ lines have an average width of 14 km s$^{-1}$, $T_{\rm gas}$ of 450~K and an emitting region from 1 to 15 AU. $^{13}$CO and C$^{18}$O lines are on average 70 and 100 K colder, 1 and 4 km s$^{-1}$ narrower than $^{12}$CO $\upsilon=1\rightarrow0$, and are dominated by emission at R$\geq$ 6 AU. The $^{12}$CO $\upsilon=1\rightarrow0$ composite line-profile indicates that if there is a gap devoid of gas it must have a width narrower than 2 AU. We find that a drop in the gas surface density ($\delta_{\rm gas}$) at $R<5-6$ AU is required to be able to simultaneously reproduce the line-profiles and rotational diagrams of the three CO isotopologs. Models without a gas density drop generate $^{13}$CO and C$^{18}$O emission lines that are too broad and warm. The value of $\delta_{\rm gas}$ can range from $10^{-2}$ to $10^{-4}$ depending on the gas-to-dust ratio of the outer disk. We find that the gas surface density profile at $1<R<$ 6 AU is flat or increases with radius. We derive a gas column density at $1<R<6$ AU of $N_H=3\times10^{19} - 10^{21}$ cm$^{-2}$ ($7\times10^{-5} - 2.4\times10^{-3}$ g cm$^{-2}$) assuming $N_{\rm CO} = 10^{-4} N_H$. We find a 5$\sigma$ upper limit on the CO column density $N_{\rm CO}$ at R$\leq$1 AU of $5\times10^{15}$ cm$^{-2}$ ($N_H\leq5\times10^{19}$ cm$^{-2}$). } { The dust gap in the disk of HD~139614 has molecular gas. The distribution and amount of gas at R$\leq6$ AU in HD~139614 is very different from that of a primordial disk. The gas surface density in the disk at $R\leq1$ AU and at $1<R<6$ AU is significantly lower than the surface density that would be expected from the accretion rate of HD~139614 ($10^{-8}$ M$_\odot$ yr$^{-1}$) assuming a standard viscous $\alpha$-disk model. The gas density drop, the non-negative density gradient in the gas inside 6 AU, and the absence of a wide ($>$ 2 AU) gas gap, suggest the presence of an embedded $<2$ M$_{\rm J}$ planet at around 4 AU. } | {Transition disks are protoplanetary disks that exhibit a deficit of continuum emission at near- and/or mid-IR wavelengths in their spectral energy distribution \citep[for a recent review, see][]{Espaillat2014}. This deficit of emission is commonly interpreted as evidence of a dust gap, a dust cavity, or a dust hole inside the disk\footnote{We call a dust hole, when no dust emission is detected inside a determined radius in the disk at {\it all} wavelengths. We call a dust cavity, a region where there is a drop in the dust density. Inside the dust cavity radius dust is still present (i.e. continuum emission is detected inside the cavity radius, for instance at IR wavelengths). We call a dust gap when continuum emission is detected at radii smaller and larger than the location of the gap. A dust cavity can have a dust gap inside it.}. Sub-mm interferometry observations have confirmed the existence of dust cavities by spatially resolving the thermal emission from cold large ($\sim$mm) grains at tens of AU in transition disks \citep[e.g.,][]{Pietu2006,Brown2009,Andrews2011,Cieza2012,Casassus2013,Perez2014}}. Observations of scattered light in the near-IR using adaptive optics have further confirmed the dust cavities in micron-sized dust grains. These high-spatial resolution observations show that the cavity size in small grains can be smaller than that in large grains \citep[e.g.,][]{Muto2012, Garufi2013, Follette2013, Pinilla2015J16}. Furthermore, near-IR scattered light imaging and sub-mm interferometry observations have revealed that a large fraction of transition disks has asymmetries in the dust distribution (e.g. spirals, blobs, and horseshoe shapes), although, the presence and shape of asymmetries appear to be different depending on the wavelength of the observations and thus the dust sizes traced \citep[e.g.,][]{Muto2012,vanderMarel2013,Isella2013,Perez2014,Benisty2015,Follete2015}. \begin{table*} \begin{center} \caption{Stellar properties} \begin{tabular}{cccccccccccc} \hline \hline Star & Sp. Type & T${\rm eff}$ & d & Mass & Radius & RV & W2 & Age & i$_{\rm disk}$ & $L_{\rm X}$ & $\dot{M}$ \\ & & [K] & [pc] & $[M_{\odot}]$ & $[R_{\odot}]$ & [km s$^{-1}$] & [mag] & [Myr] & [$^\circ$] & [erg s$^{-1}$] & [M$_{\odot}$ yr$^{-1}$]\\[1mm] \hline {HD~139614} & A7Ve $^a$ & 7600$\pm$300 $^{b}$ & {131$\pm$5} $^{c}$ & 1.76$^{+0.15}_{-0.08}$ $^b$ & 2.06$\pm$0.42 $^b$ &0.3$\pm$2.3 $^{b}$ & 5.1 $^d$ & 8.8$^{ +4.5}_{-1.9}$ $^b$ & 20 $^e$ & 1.2$\times10^{29}$ $^f$ & $10^{-8}$ $^g$ \\[1mm] & & 7850 $^h$ & & 1.7$\pm$0.3 $^h$ & 1.6~$^h$& & & $>$7 $^a$ & \\[1mm] \hline \end{tabular} \tablefoot{ $^a$~\citet[][]{vanBoekel2005}; $^b$~\citet[][]{Folsom2012,Alecian2013}; $^c$~{\citet[][]{Gaia2016}}; $^d$~4.6 $\mu$m, WISE satellite release 2012 \citep[][]{Cutri2012}; $^e$ \citet[][]{Matter2016}; $^f$ G\"udel et al. (in prep.) see Sect.~\ref{origin}; $^g$ \citet[][]{GarciaLopez2006}; $^h$ \citet[][]{vanBoekel2005} stellar properties used in \citet[][]{Matter2016}. \\[3mm]} \label{stellar-properties} \caption{Log of the science and calibrator observations} { \begin{tabular}{lccccccccc} \hline \hline Star & UT Date Obs. & $t_{\rm exp}$ & Airmass & Seeing & RV$_{\rm bary}$$^a$ & PSF$_{\rm FWHM}$$^{b}$ & S/N$^{~b,c}$ & \multicolumn{2}{c}{sensitivity 3$\sigma$$~^{b,d}$} \\ & [y-m-d] & [s] & & [''] & [km s$^{-1}$] & [mas] & & \multicolumn{2}{c}{[10$^{-15}$ erg s$^{-1}$ cm$^{-2}$]}\\[1mm] & & & & & & & & 3.3 km s$^{-1}$& 20 km s$^{-1}$ \\ \hline HD~139614 & 2013-06-15 & 2400 & $1.07 - 1.13$ & $ 0.93 - 1.23$ & $9.74\pm0.02$ & 178$\pm$10 & 160 $-$ 100 & 0.2 $-$ 0.3 & 1.2 $-$ 2.0\\ CAL HIP 76829 & 2013-06-15 & 320 & $1.16 - 1.18$ & $ 0.87 - 1.06$ & $9.36\pm0.01$& 172$\pm$10 & 310 $-$ 200\\[1mm] \hline \end{tabular} \tablefoot{\small $^a$ {Radial velocity due to the rotation of the Earth, the motion of the Earth about the Earth-Moon barycenter, and the motion of the Earth around the Sun}; $^b$ measured in one nod position; $^c$ for the science spectra the S/N is measured in the telluric-corrected spectrum, note that the S/N decreases from chip 1 to chip 4; $^d$ integrated flux sensitivity limits are given for a spectrally unresolved line of width 3.3 km s$^{-1}$ and a line of width 20 km s$^{-1}$.}\\[3mm] } \label{table_observations} \end{center} \end{table*} The origin of the dust cavities and gaps in transition disks is a matter of intense debate in the literature: scenarios such as grain growth (e.g., \citealt{DullemondDominik2005}; but see \citealt{Birnstiel2012}), size-dependent dust radial drift \citep[e.g.,][]{PinteLaibe2014}, dust dynamics at the boundary of the dead-zone \citep[][]{Regaly2012}, photoevaporation \citep[e.g.,][]{Clarke2001,AlexanderArmitage2007,Owen2012}, {giant planet(s) \cite[e.g.,][]{MarshMahoney1992,Lubow1999,Rice2003,Quillen2004,Varniere2006,Zhu2011}}, dynamical interactions in multiple systems \citep[e.g.,][]{ArtymowiczLubow1996,IrelandKraus2008, Fang2014}, and magneto-hydrodynamical phenomena \citep[][]{ChiangMurrayClay2007} have all been proposed. Accretion signatures in many transition disks \citep[e.g.,][]{Fang2009,SiciliaAguilar2013,Manara2014} and emission of warm \cite[e.g,][]{Bary2003,Pontoppidan2008, Pontoppidan2011, Salyk2009, Salyk2011} and cold \citep[][]{Casassus2013,Bruderer2014,Perez2015HD142527,Canovas2015,vanderMarel2015,vanderMarel2016} molecular gas indicate that the dust cavities in accreting transition disks contain gas. Radiative transfer modeling of CO ro-vibrational emission \citep[][]{Carmona2014} and CO pure rotational emission \citep[][]{Bruderer2013,Perez2015HD142527,vanderMarel2015,vanderMarel2016} further suggests a gas surface density drop ($\delta_{\rm gas}$) inside the dust cavity, with $\delta_{\rm gas}$ values varying from 0.1 up to 10$^{-5}$ (see Table~\ref{table_gasdrop}). Some of the transition disks are not accreting and thus do not seem to have gas \citep[][]{SiciliaAguilar2010}. There is also a substantial difference in the global structure and/or disk mass between accreting and non-accreting transition disks, with the non-accreting disks being significantly more evolved (lower masses, flatter disks) as seen with Herschel \citep[][]{SiciliaAguilar2015}. The different spatial locations of dust grains of different sizes, the gas inside the sub-mm dust cavities, together with the different surface density profiles of gas and dust strongly favor the planet(s) scenario. However, we probably witness several coexisting mechanisms, because planet formation might affect the dynamics of the dust in the disk \citep[e.g., ][]{Rice2003, Zhu2011, Pinilla2012, Pinilla2015} or favor the onset of photoevaporation, when the accretion rate has decreased \citep[e.g.,][]{Rosotti2013,Dittkrist2014}. A large portion of studies of transition disks have focused on investigating disks that are bright in the sub-mm and that have large dust cavities of tens of AU \citep[e.g.,][]{Andrews2011,vanderMarel2015b}. Because a single Jovian planet interacting with the disk is expected to open a gap only a few AU wide \citep[e.g.,][]{Kley1999,Crida2007}, multiple (unseen) giant planets have been postulated as a possible explanation for the observed large dust cavities \citep[][]{Zhu2011, Dodson-RobinsonSalyk2011}. {In a recent near- and mid-IR interferometry campaign, \citet[][]{Matter2014,Matter2016} have revealed that the 9 Myr old \cite[][]{Alecian2013} accreting \citep[$10^{-8}$ M$_{\odot}$/yr, ][]{GarciaLopez2006} Herbig A7Ve star HD~139614 has a transition disk with a narrow dust gap extending from 2.3$\pm$0.1 to 5.3$\pm$0.3 AU \footnote{The dust gap limits derived in \citet[][]{Matter2016} are 2.5$\pm$0.1 to 5.7$\pm$0.3 AU. They were calculated using a distance of 140 pc. The values in the text are the values corrected by the new Gaia distance. Both values are consistent within the uncertainties.}. and a dust density drop $\delta_{\rm dust}$ at R$<$6 AU of 10$^{-4}$ (see Table~\ref{stellar-properties} for a summary of the stellar properties). HD~139614 is one of the first objects with a spatially resolved dust gap with a width of only a few AU, thus it might be the case of a transition disk where the dust gap has been opened by a single giant planet. HD~139614 is located within the Sco OB2-3 association \citep[][]{Acke2005} {at a distance of 131$\pm$5 pc} \citep[][]{Gaia2016}. HD~139614 has peculiar chemical abundances in its photosphere \citep[][]{Folsom2012}, with depletions of heavier refractory elements, while C, N, and O are approximately solar. HD~139614 belongs to the group I Herbig Ae stars according to the Spectral Energy Distribution (SED) classification scheme of \citet[][]{Meeus2001}, which suggests that its outer disk is flared. \citet[][]{Matter2016} derived a dust disk mass of 10$^{-4}$ M$_\odot$ based on a fit to the SED. The {\it Spitzer} mid-IR spectra of HD~139614 exhibit a weak amorphous silicate feature at 10~$\mu$m \citep[][]{Juhasz2010} and Polycyclic Aromatic Hydrocarbons (PAH) emission \cite[][]{Acke2010}. The disk's mid-IR continuum has been spatially resolved at 18 $\mu$m ({\it FWHM} of 17 $\pm$4 AU) but it is not resolved at 12 $\mu$m \citep[][]{Marinas2011}. \citet[][]{Kospal2012} reported that the ISOPHOT-S, {\it Spitzer} and TIMMI-2/ESO 3.6m mid-infrared spectra taken at different epochs agree within the measurement uncertainties, thus suggesting that there is no strong mid-IR variability in the source. Emission from cold CO gas in the outer disk of HD~139614 has been reported in JCMT single-dish observations by \citet[][]{Dent2005} and \citet[][]{PanicHogerheijde2009}. Emission of [\ion{O}{I}] at 63 $\mu$m from the disk has been detected by Herschel \citep[][]{Meeus2012,Fedele2013}. The [\ion{O}{I}] 63 $\mu$m line flux of HD~139614 is among the weakest of the whole Herbig Ae sample observed by Herschel. No emission of [\ion{O}{I}] at 145 $\mu$m, [\ion{C}{II}] at 157 $\mu$m, CO, H$_2$O, OH or CH$^{+}$ in the 50 $-$200$~\mu$m region was detected by Herschel \citep[][]{Meeus2012,Meeus2013, Fedele2013}. } In this paper we present the results of high-resolution spectroscopy observations of CO ro-vibrational emission at 4.7 $\mu$m towards HD~139614 obtained with the ESO/VLT CRIRES instrument \citep[][]{Kaufl2004}. Our aim is to use CO isotopolog spectra to constrain the warm gas content in the inner disk of HD~139614 and address the following questions: What is the gas distribution in the inner disk of HD~139614? Does the HD~139614 disk have a gas-hole, a gas-density drop or a gap in the gas? How does the gas distribution compare with the dust distribution? What is the most likely explanation for the observed gas and dust distributions in HD~139614? The paper is organized as follows. We start by describing the observations and data reduction in Sect. 2. In Sect. 3, we present the observational results. In Sect. 4, we derive the CO-emitting region, the average temperature and column density of the emitting gas, and the gas surface density and temperature distribution. In Sect. 5, we discuss our results in the context of the proposed scenarios for the origin of transition disks and compare HD~139614 with other transition disks. Sect. 6 summarizes our work and provides our conclusions. | { We have obtained VLT/CRIRES high-resolution spectra ($R\sim90\,000$) of CO ro-vibrational emission at 4.7 $\mu$m in HD139614, an accreting (10$^{-8}$ M$_\odot$ yr$^{-1}$) Herbig Ae star with a (pre-) transition disk that is characterized by a dust gap between 2.3 and 6 AU and a dust density drop $\delta_{\rm dust}$ of 10$^{-4}$ at R$<$ 6 \citep[][]{Matter2016}. We have detected $\upsilon=1\rightarrow$0~$^{12}$CO, $^{13}$CO, C$^{18}$O, C$^{17}$O, and $\upsilon=2\rightarrow$1~$^{12}$CO ro-vibrational emission. The lines observed are consistent with disk emission and thermal excitation. We find the following: \begin{enumerate} \item The $\upsilon=1\rightarrow$0 $^{12}$CO spectrum indicates that there is gas from 1 AU up to 15 AU, and that there is no gap in the gas distribution. If a gap is present in the gas (i.e., a region devoid of gas) then it should have a width smaller than 2 AU. \item The spectra of $^{13}$CO and C$^{18}$O $\upsilon=1\rightarrow$0 emission are on average colder and emitted farther out in the disk ($R>$6 AU) than the $^{12}$CO $\upsilon=1\rightarrow$0 emission. Keplerian flat-disk models clearly show that a drop in the gas density $\delta_{\rm gas}$ of a factor of at least 100 at $R<5-6$ AU is needed to describe simultaneously the line-profiles and rotational diagrams of the three CO isotopologs. Models without a gas density drop produce C$^{18}$O and $^{13}$CO lines that are too wide and warm to be compatible with the data. If the gas-to-dust mass ratio is equal to 100 in the outer disk, the gas depletion factor $\delta_{\rm gas}$ could be as high as 10$^{-4}$. Moreover, we find that the gas surface density profile in the inner 6 AU of the disk is flat or increases with radius. \end{enumerate} The presence of molecular gas inside 6 AU and the weak X-ray luminosity do not favor photoevaporation as the main mechanism responsible for the inner disk structure of HD~139614. The gas density drop, a flat or increasing gas surface density profile at $R<6$ AU, combined with the non-detection of a gap in the gas wider than 2 AU, suggest the presence of a single Jovian-mass planet inside the dust gap. If a giant planet is indeed responsible for the transition disk shape of HD~139614, then its location would be at around 4 AU and its mass would be lower than two Jupiter masses. Furthermore, if a small gap in the gas (due to a planet) were to be present, a gas surface density profile that increases with radius in the inner disk might lead to a dust trap at the gap inner edge for sub-micron and micron sized grains, which could explain that the dust surface density increases with radius at $R<2.5$ AU, as found in IR interferometry observations. We constrained the gas column density between 1 and 6 AU to $N_H=3\times10^{19} - 10^{21}$ cm$^{-2}$ ($7\times10^{-5} - 2.4\times10^{-3}$ g cm$^{-2}$) assuming $N_{\rm CO} = 10^{-4} N_H$. We derived a 5$\sigma$ upper limit on the CO column density at $R<$1 AU $N_{\rm CO}=5\times10^{15}$ cm$^{-2}$, which suggests an $N_H<5\times10^{19}$ cm$^{-2}$ at $R<$1 AU . The gas surface density in the disk of HD~139614 at $R\leq1$ AU and at $1<R<6$ AU is significantly lower than the surface density that would be expected for the accretion rate of HD~139614, assuming a standard viscous $\alpha$-disk model. Our result, and the low gas surface densities reported in the inner disks of other transition disks, suggests that stellar accretion rates should {\it not} be used as direct proxies to derive the amount of gas left inside the dust cavities of transition disks. An investigation of the topology of the magnetic fields of young stars with transition disks is needed to help address the question of the differences between the accretion rate and the inner disk gas surface density. We have discussed the ensemble of transition disks with current constraints for the gas surface density inside the dust cavity. The sample shows that, in the majority of the sources, the drop in the dust density is larger than the drop in the gas density ($\delta_{\rm dust}<\delta_{\rm gas}$). This suggests that dust is depleted faster than gas in the inner disk. The number of transition disks with a complete set of multi-wavelength observations of gas and dust (ALMA, CRIRES, {\it Herschel}, {\it Spitzer}, {VLTI}, SEDs, HiCiAO, VLT/SPHERE, and GPI) is growing. A homogenous multi-wavelength and multi-technique modeling of gas and dust observations in transition disks would be of great help to understand the variety of gas and dust structures that these disks have, and to study the possible links to planet formation. In that respect, it would be of great help to have spatially resolved measurements of the dust and the rotational transitions of CO isotopologs in the sub-mm for HD~139614 (for example with ALMA\footnote{At the time of writing, HD~139614 has not been observed with ALMA.}). } | 16 | 9 | 1609.06708 |
1609 | 1609.04820_arXiv.txt | {Standard galaxy formation models predict that large-scale double-lobed radio sources, known as DRAGNs, will always be hosted by elliptical galaxies. In spite of this, in recent years a small number of spiral galaxies have also been found to host such sources. These so-called \emph{spiral DRAGNs} are still extremely rare, with only $\sim 5$ cases being widely accepted. Here we report on the serendipitous discovery of a new spiral DRAGN in data from the Giant Metrewave Radio Telescope (GMRT) at 322\,MHz. The host galaxy, MCG+07-47-10, is a face-on late-type Sbc galaxy with distinctive spiral arms and prominent bulge suggesting a high black hole mass. Using WISE infra-red and GALEX UV data we show that this galaxy has a star formation rate of 0.16-0.75 \,M$_{\odot}$\,yr$^{-1}$, and that the radio luminosity is dominated by star-formation. We demonstrate that this spiral DRAGN has similar environmental properties to others of this class, but has a comparatively low radio luminosity of $L_{\rm 1.4\,GHz}$ = 1.12$\times$10$^{22}$ W Hz$^{-1}$, two orders of magnitude smaller than other known spiral DRAGNs. We suggest that this may indicate the existence of a previously unknown low-luminosity population of spiral DRAGNS.} | Spiral DRAGNs \citep[Double-lobed Radio sources Associated with Galactic Nuclei, ][]{Leahy1993} are spiral galaxies that host large-scale double-lobed radio sources. The existence of such sources is in contradiction to our existing models of galaxy formation \citep[e.g.][]{Hopkins2008}, which predict that DRAGNs should be hosted exclusively by elliptical galaxies. Indeed, until recently, observations of DRAGNs in the local Universe confirmed this expectation \citep[e.g.][]{Matthews1964, Urry1995,Best2005}. Elliptical galaxies are formed as a result of mergers, the phenomenology of which also triggers the formation of DRAGNs \citep{Chiaberge2011,Chiaberge2015}. However, a spiral galaxy's structure cannot withstand a major merger. Moreover, morphological transition from spiral to elliptical is thought to be a one-way process, at least in the local Universe. Consequently, the standard galaxy formation model does not predict the existence of spiral DRAGNs. Nonetheless, a number of spiral DRAGNs have been discovered in recent years \citep[e.g.][]{Ledlow2001,Hota2011,Bagchi2014,Mao2015,Singh2015}. While the first three spiral DRAGN discoveries were serendipitous, \citet{Mao2015} performed the first systematic search for these sources. Using Galaxy Zoo \citep{Lintott2008} morphological classifications were cross-matched with the Faint Images of the Radio Sky at Twenty-Centimeters \citep[FIRST, ][]{Becker1995} and the NRAO VLA Sky-Survey \citep[NVSS, ][]{Condon1998}. In this study, only one spiral DRAGN was found above L$_{\rm 1.4\,GHz}$ = $10^{23}$\,W\,Hz$^{-1}$. \citet{Singh2015} performed a similar analysis using the spiral galaxy catalogue of \citet{Meert2015} and reported the identification of four spiral DRAGNs in those data, including one that was previously known and three that were unknown. However, the precise identification of spiral DRAGN hosts remains contentious in the literature and to date the existence of only 5 spiral DRAGNs are widely accepted. DRAGNs with spiral hosts may represent a rare phenomenon of elliptical galaxies transitioning back into spirals through accretion of gas and stars, perhaps from a companion. A key question is whether spiral DRAGNs are a result of non-standard physical properties, a result of their environment, or perhaps a combination of their nature and nurture. Studying spiral DRAGNs, as well as establishing their numbers more exactly, is vital in order to reconcile their role in standard galaxy formation theories. In this Letter, we present the discovery of a new spiral DRAGN at 325\,MHz with the Giant Meterwave Radio Telescope \citep[GMRT; ][]{Swarup1990}. In Sect.~2 we outline the data processing and imaging steps. In Sect.~3, we present the discovery of this new spiral DRAGN with a description of both its radio morphology and that of the host galaxy, as understood from available multi-wavelength data. In Sect.~4, using Infra-red and UV data we demonstrate that the host galaxy is star-forming and we compare its star formation rate to other spiral DRAGNs. Finally, in Sect.~5 we discuss the nature of this object and state our conclusions. In this work we assume a $\Lambda$CDM cosmology with $H_0 = 69.6\, \mathrm{km s^{-1} Mpc^{-1}}$, ${\Omega_\mathrm{m}} = 0.286$ and ${\Omega_\Lambda} = 0.714$ \citep{Bennett2014}, which we use for the calculation of distance, luminosity and star formation rate. At a redshift of z = 0.017, these values result in a conversion of 0.348\,kpc/$\arcsec$. All uncertainties are quoted at 1\,$\sigma$. | This paper presents the discovery of the spiral DRAGN MCG+07-47-10. The host galaxy has clearly defined spiral arms, a prominent bulge, and hosts a 188\,kpc DRAGN. The radio source, first identified in NVSS, was not previously classified as a DRAGN, but rather three separate radio sources. The deep GMRT observation presented in this paper detects the low surface brightness emission connecting the radio components, thus identifying this radio source as a DRAGN for the first time. The central component of the radio emission, from the host galaxy, appears extended in both the NVSS and GMRT data. These new GMRT data show resolved radio emission emanating from the entirety of the host galaxy, as opposed to only the core. This suggests that the radio emission is not solely due to the presence of an AGN. Moreover, the total integrated radio flux density for the host galaxy gives a SFR that is in good agreement with SFRs calculated from the IR emission. Resolved radio emission observed throughout the disk suggests that most, if not all the radio emission across the disk is due to star-formation. We find that the luminosity of this spiral DRAGN ($L_{\rm 1.4\,GHz}$ = 1.12$\times$10$^{22}$\,Watts\,Hz$^{-1}$) to be significantly lower than other spiral DRAGNS. However, without an reliable redshift this could be misleading. If we take the median redshift from the main galaxy sample from SDSS of $z=0.1$ \citep{Strauss2002}, we would calculate a luminosity of $L_{\rm 1.4\,GHz}$ = 4.25$\times$10$^{23}$\,Watts\,Hz$^{-1}$. This is still lower than that found in other spiral DRAGNS such as J1649+2635 with $L_{\rm 1.4\,GHz}$=1.03$\times$10$^{24}$\,Watts\,Hz$^{-1}$ \citep{Mao2015}. A redshift of 0.1 would imply an optical diameter of 57\,kpc for MCG+07-47-10 and the angular linear extent of the entire spiral DRAGN of $\sim 9$\,$\arcmin$ would equate to a physical size of $\approx$ 1\,Mpc. In order for MCG+07-47-10 to have a comparable luminosity to J1649+2635, MCG+07-47-10 would need to have a redshift of $z \approx 0.17$. This redshift would make the physical size of the DRAGN associated with MCG+07-47-10 significantly larger than 1\,Mpc. We note that this would not necessarily be unusual for the class as two previously identified spiral DRAGNS \citep{Hota2011,Bagchi2014} show evidence of mega-parsec structure. However, the host galaxy, MCG+07-47-10, would then have an optical size of approximately 100\,kpc, which would be exceptionally large compared to similar analogues. The low-luminosity and low surface brightness of the DRAGN may suggest that the radio emission is old. One possible scenario leading to the formation of this spiral DRAGN is that the host of a low-luminosity DRAGN has had gas injected onto it, perhaps through a merger, and this gas has triggered star-formation and built up spiral arms. Further modelling and observations at several frequencies are required to test this theory. Assuming that the host galaxy, MCG+07-47-10, is associated with the two galaxy groups in its immediate vicinity (located within the virial radius of NGC\,7618 and UGC\,12491 \citep{Kraft2006}), it would appear to reside in a similarly intermediate density environment to other known spiral DRAGNs \citep{Mao2015}. This supports the idea that spiral DRAGNs require specific environments to form and that a moderately overdense environment is conducive for mergers to trigger a process that transitions ellipticals back to spirals. In conclusion, we have presented the discovery of a new spiral DRAGN from observations with the GMRT at 322\,MHz. We have shown that the host galaxy is a star-forming spiral galaxy. Given currently available information, we have demonstrated that the radio luminosity of this new source is significantly lower than that of other spiral DRAGNs, about two orders of magnitude, with a value of $L_{\rm 1.4\,GHz}$ = 1.12$\times$10$^{22}$\,Watts\,Hz$^{-1}$. This may indicate the existence of a previously unknown population of low-luminosity spiral DRAGNs and we suggest that future radio surveys may be used to expand this sample further. | 16 | 9 | 1609.04820 |
1609 | 1609.08838_arXiv.txt | Photometric data from the Xuyi Schmidt Telescope Photometric Survey of the Galactic Anticentre (XSTPS-GAC) and the Sloan Digital Sky Survey (SDSS) are used to derive the global structure parameters of the smooth components of the Milky Way. The data, which cover nearly 11,000 deg$^2$ sky area and the full range of Galactic latitude, allow us to construct a globally representative Galactic model. The number density distribution of Galactic halo stars is fitted with an oblate spheroid that decays by power law. The best-fit yields an axis ratio and a power law index $\kappa=0.65$ and $p=2.79$, respectively. The $r$-band differential star counts of three dwarf samples are then fitted with a Galactic model. The best-fit model yielded by a Markov Chain Monte Carlo analysis has thin and thick disk scale heights and lengths of $H_{1}=$ 322\,pc and $L_{1}=$2343\,pc, $H_{2}=$794\,pc and $L_{2}=$3638\,pc, a local thick-to-thin disk density ratio of $f_2=$11\,per\,cent, and a local density ratio of the oblate halo to the thin disk of $f_h=$0.16\,per\,cent. The measured star count distribution, which is in good agreement with the above model for most of the sky area, shows a number of statistically significant large scale overdensities, including some of the previously known substructures, such as the Virgo overdensity and the so-called ``north near structure'', and a new feature between 150\degr $< l < $ 240\degr~and $-1$5\degr $< b < $ $-$5\degr, at an estimated distance between 1.0 and 1.5\,kpc. The Galactic North-South asymmetry in the anticentre is even stronger than previously thought. | \label{introduction} One of the fundamental tasks of the Galactic studies is to estimate the structure parameters of the major structure components. \citet{Bahcall1980} fit the observations with two structure components, namely a disk and a halo. \citet{Gilmore1983} introduce a third component, namely a thick disk, confirmed in the earliest Besancon Galaxy Model \citet{Creze1983}. Since then, various methods and observations have been adopted to estimate parameters of the thin and thick disks and of the halo of our Galaxy. As the quantity and quality of data available continue to improve over the years, the model parameters derived have become more precise, numerically. Ironically, those numerically more precise results do not converge (see Table~1 of \citealt{Chang2011}, Table~2 of \citealt{Lopez2014} and Sect.~5 and 6 of \citealt{Bland2016} for a review). The scatters in density law parameters, such as scale lengths, scale heights and local densities of these Galactic components, as reported in the literature, are rather large. At least parts of the discrepancies are caused by degeneracy of model parameters, which in turn, can be traced back to the different data sets adopted in the analyses. Those differing data sets either probe different sky areas \citep{Bilir2006a, Du2006, Cabrera2007, Ak2007, Yaz2010, Yaz2015}, are of different completeness magnitudes and therefore refer to different limiting distances \citep{Karaali2007}, or of consist of stars of different populations of different absolute magnitudes \citep{Karaali2004, Bilir2006b, Juric2008, Jia2014}. It should be noted that the analysis of \citet{Bovy2012}, using the SEGUE spectroscopic survey, has given a new insight on the thin and thick disk structural parameters. This analysis provides estimate of their scale height and scale height as a function of metallicity and alpha abundance ratio. However, it relies on incomplete data (since it is spectroscopic) with relatively low range of Galactocentric radius as for the thin disk is concerned. A wider and deeper sample than those employed hitherto may help break the degeneracy inherent in a multi-parameter analysis and yield a globally representative Galactic model. A single or a few fields are insufficient to break the degeneracy. The resulted best-fit parameters, while sufficient for the description of the lines of sight observed, may be unrepresentative of the entire Galaxy. For the latter purpose, systematic surveys of deep limiting magnitude of all or a wide sky area, such as the Two Micron All Sky Survey (2MASS; \citealt{Skrutskie2006}), the Sloan Digital Sky Survey (SDSS; \citealt{York2000}), the Panoramic Survey Telescope \& Rapid Response System (Pan-Starrs; \citealt{Kaiser2002}) and the GAIA mission \citep{Perryman2001}, are always preferred. Several authors have studied the Galactic structure with 2MASS data at low \citep{Lopez2002, Yaz2015} or high latitudes \citep{Cabrera2005, Cabrera2007, Chang2011}. \citet{Polido2013} uses the model from \citet{Ortiz1993} and rederive the parameters of this model based on the 2MASS star counts over the whole sky area. However, the survey depth of 2MASS is not quite enough to reach the outer disk and the halo. The survey depth of SDSS is much deeper than that of the 2MASS. Many authors (e.g. \citealt{Chen2001, Bilir2006a, Bilir2008, Jia2014, Lopez2014}) have previously used the SDSS data to constrain the Galactic parameters. Those authors have only made use of a portion of the surveyed fields, at intermediate or high Galactic latitudes. \citet{Juric2008} obtain Galactic model parameters from the stellar number density distribution of 48 million stars detected by the SDSS that sample distances from 100\,pc to 20\,kpc and cover 6500\,deg$^2$ of sky. Their results are amongst those mostly quoted. However, in their analysis, they have avoided the Galactic plane. So the constraints of their results on the disks, especially the thin disk, are weak. In their analysis, \citet{Juric2008} have also adopted photometric parallaxes assuming that all stars of the same colour have the same metallicity. Clearly, (disk) stars in different parts of the Galaxy have quite different \citep{Ivezic2008, Xiang2015, Huang2015} metallicities, and these variations in metallicities may well lead to biases in the model parameters derived. In order to provide a quality input catalog for the LAMOST Spectroscopic Survey of the Galactic Anticentre (LSS-GAC; \citealt{Liu2014,Liu2015, Yuan2015}), a multi-band CCD photometric survey of the Galactic Anticentre with the Xuyi 1.04/1.20m Schmidt Telescope (XSTPS-GAC; \citealt{Zhang2013,Zhang2014,Liu2014}) has been carried out. The XSTPS-GAC photometric catalog contains more than 100 million stars in the direction of Galactic anticentre (GAC). It provides an excellent data set to study the Galactic disk, its structures and substructures. In this paper, we take the effort to constrain the Galactic model parameters by combining photometric data from the XSTPS-GAC and SDSS surveys. This is the third paper of a series on the Milky Way study based on the XSTPS-GAC data. In \citet{Chen2014}, we present a three dimensional extinction map in $r$ band. The map has a spatial angular resolution, depending on latitude, between 3 and 9\,arcmin and covers the entire XSTPS-GAC survey area of over 6,000 deg$^2$ for Galactic longitude 140 $< l <$220\,deg and latitude 40 $< b <$40\,deg. In \citet{Chen2015}, we investigate the correlation between the extinction and the \HI~ and CO emission at intermediate and high Galactic latitudes ($|b| >$ 10\degr) within the footprint of the XSTPS-GAC, on small and large scales. In the current work we are interested in the global, smooth structure of the Galaxy. For the Galactic structure, in addition to the global, smooth major components, many more (sub-)structures have been discovered, including the inner bars near the Galactic centre \citep{Alves2000, Hammersley2000, vanLoon2003, Nishiyama2005, Cabrera2008, Robin2012}, flares and warps of the (outer) disk \citep{Lopez2002, Robin2003, Momany2006, Reyle2009, Lopez2014}, and various overdensities in the halo and the outer disk, such as the Sagittarius Stream \citep{Majewski2003}, the Triangulum-Andromeda \citep{Rocha2004, Majewski2004} and Virgo \citep{Juric2008} overdensities, the Monoceros ring \citep{Newberg2002,Rocha2003} and the Anti-Center Stream \citep{Rocha2003,Crane2003, Frinchaboy2004}. They show the complexity of the Milky Way. Recently, \citet{Widrow2012} and \citet{Yanny2013} have found evidence for a significant Galactic North-South asymmetry in the stellar number density distribution, exhibiting some wavelike perturbations that seem to be intrinsic to the disk. \citet{Xu2015} show that in the anticentre regions there is an oscillating asymmetry in the main-sequence star counts on either sides of the Galactic plane, in support of the prediction of \citet{Ibata2003}. The asymmetry oscillates in the sense that there are more stars in the north, then in the south, then back in the north, and then back in the south at distances of about 2, 4 -- 6, 8 -- 10 and 12 -- 16\,kpc from the Sun, respectively. The paper is structured as follows. The data are introduced in Section~2. We describe our model and the analysis method in Section~3. Section~4 presents the results and discussions. In Section~5 we discuss the large scale excess/deficiency of star counts that reflect the substructures in the halo and disk. Finally we give a summary in Section~6. | \begin{figure} \centering \includegraphics[width=0.48\textwidth]{halogrid.eps} \caption{Reduced likelihood surface of the halo parameters $p$ and $\kappa$ space (see Table~2). The best-fitted values and uncertainties are marked as a red plus with error bars. The red contour ellipse shows the likelihood ranges used for estimating the uncertainties.} \label{halog} \end{figure} \begin{table*} \centering \caption{The best-fit values of the disk fit} \begin{tabular}{lcccccccc} \hline \hline Bin & $n_1$ & $L_1$ & $H_1$ & $f_2$ & $L_2$ & $H_2$ & $f_H$ & $Lr$ \\ & $10^{-3}$stars\,$pc^{-3}$ & pc & pc & per\,cent & pc & pc & per\,cent & \\ \hline Joint fit\\ \hline $0.5 \le (g-i)_0 < 0.6$ & 1.25 & 2343 & 322 & 11 & 3638 & 794 & 0.16 & $-$86769 \\ $0.6 \le (g-i)_0 < 0.7$ & 1.20 & & & & & & & \\ $1.5 \le (g-i)_0 < 1.6$ & 0.54 & & & & & & & \\ \hline Individual fit\\ \hline $0.5 \le (g-i)_0 < 0.6$ & 1.31 & 1737 & 321 & 14 & 3581& 731 & 0.16 & $-$43699 \\ $0.6 \le (g-i)_0< 0.7$ & 1.65 & 2350 & 284 & 7 & 3699 & 798 & 0.12 & $-$35774 \\ $1.5 \le (g-i)_0 < 1.6$ & 0.41 & 2780 & 359 & 8 & 2926 & 1014 & 0.50 & $-$4028 \\ \hline stddev & & 429 & 31 & 3 & 360 & 124 & 0.02 & \\ \hline \end{tabular} \end{table*} \begin{figure*} \centering \includegraphics[width=0.95\textwidth]{cross2d.eps} \caption{ Two-dimensional marginalized PDFs for the disk model parameters, $L_1,~H_1,~f_2,~L_2,$ and $H_2$ and the halo-to-thin disk normalization $f_h$, obtained from the MCMC analysis. Histograms on top of each column show the one-dimensional marginalized PDFs of each parameter labeled at the bottom of the column. Red pluses and lines indicate the best solutions. The dash lines give the 16th and 84th percentiles, which denotes only the fitting uncertainties.} \label{cross2d} \end{figure*} \begin{figure*} \centering \includegraphics[width=0.68\textwidth]{rntestmodel.eps} \caption{Star count (per deg$^2$) for the colour bin $0.5 \le (g-i)_0 <0.6$\,mag and magnitude bins, $r_0$ = 15 (left) and 16\,mag (right), of both the XSTPS-GAC (red pluses) and the SDSS (blue pluses) data as a function of the Galactic latitude for example subfields with Galactic longitude 177\degr $<l<$ 183\degr. The black dots are the model predictions.} \label{rnmodel} \end{figure*} \begin{figure*} \centering \includegraphics[width=0.7\textwidth]{datamode.eps} \caption{Difference map for the all subfields $(N_{\rm obs}-N_{\rm mod})/N_{\rm mod}$. The populations in excess in the data are most likely irregular structures of the Milky Way. The squares denote different regions with significant excesses of the star counts.} \label{resd} \end{figure*} \subsection{Fitting results} The best-fit halo model parameters obtained from the halo fit are listed in Table~2 and the reduced likelihood $L_r$ surface for the fit is shown in Fig.~\ref{halog}. The $L_r$ contour dramatically changes along $\kappa$, but relatively mildly along $p$. The best-fit halo model parameters are $\kappa=0.65\pm 0.05$ and $p=2.79\pm 0.17$. They are not surprisingly in good agreement with those values from \citet{Juric2008}, which have $\kappa=0.64$ and $p=2.8$, since the data used for the halo fit in the current work are mainly from the SDSS subfields. In addition, the photometric parallax relation that \citet{Juric2008} used for the blue stars was corrected for low-metallicities ([Fe/H] $\sim$ $-$1.5\,dex), which is of minor difference from the photometric parallax we used for the halo stars. The best-fit values for the disk fit are listed in Table~3. We performed the disk fit using first jointly all three colour bins, then separately for each colour bin. For the joint fit case, all the model parameters, excepted for the local density ($n_0$), are settled to be the same for stars in every bin, as we expect a universal density profile for all stellar populations \citep{Juric2008}. All the parameters resulted from the joint fit appear to be very well constrained. We explore the correlations between different parameter pairs in Fig.~\ref{cross2d}. The Figure shows the marginalized one- and two-dimensional PDFs of the model parameters. The correlations between different parameters are rather weak in general. We identify small correlations between several pairs of the parameters, such as ($H_1$, $H_2$), ($H_1$, $f_2$) and ($l_1$, $f_h$), and a strong degeneracy between the scale height and the local normalised densities ratio of thick disk ($H_2$, $f_2$). For the separate fits, we fit all parameters independently for each color bin. The individual fits can be served as a consistency check of our method. From Table~3, we can declare that those best-fit solutions are generally consistent. The thin disk scale height varies around $H_1 ~ \sim$ 320 pc, by $\pm 40$\,pc, which is consistent with the result from the joint fit. The variations of the disk scale lengths are relatively large, with $L_1~\sim$ 1.7 $-$ 2.8\,kpc and $L_2~\sim$ 2.9 -- 3.7\,kpc for the thin disk and the thick disk, respectively. The large variations could be due to the limited ranges of Galactocentric distances on the Galactic plane for stars from both the XSTPS-GAC (for the limited depth) and the SDSS (for the poor sky coverage in low Galactic latitudes). Data with deeper depth and better sky coverage in the low Galactic latitudes (such as Pan-Starrs) may help to improve the situation. The thick disk normalization $f_2$ and scale height $H_2$ appear also weakly constrained, with $f_2$ ranging from 7 to 14\,per\,cent and $H_2$ from 730 to 1000\,pc. This is mainly due to the strong correlation between these two parameters (see the $f_2$ vs. $H_2$ panel in Fig.~\ref{cross2d}). The halo normalisation $f_h$ is well constrained to $\sim$ 0.15\,per\,cent for the two G-type star bins. While the value of $f_h$ is abnormally large for the late K-type star bin [1.5$ \le (g-i)_0 < $1.6]. This is mainly due to the fact that late K dwarfs in the halo are cut by our limiting magnitude ($r_0<$21\,mag). We plot in Fig.~\ref{rnmodel} the star counts in the colour bin $0.5 \le (g-i)_0 < 0.6$\,mag and magnitude bins, $r_0=$15 and 16\,mag of both the XSTPS-GAC and the SDSS data as a function of the Galactic latitude for example subfields with Galactic longitude 177\degr $ < l < $ 183\degr. The best-fit model is in good agreement with the observations, with some small deviations. In Fig.~\ref{resd}, we show the differences between the observed star counts, integrated from stars in all the three colour bins and $r_0$ from 12 to 21\,mag, and the model predictions as a function of position on the sky. Different colours in the Figure indicate the values of the ratio $(N_{\rm obs}-N_{\rm mod})/N_{\rm mod}$. For most of the fields, we do not see obvious deviation, with residual smaller than 10\,per\,cent, i.e. $|(N_{\rm obs}-N_{\rm mod})/N_{\rm mod}| < 0.1$. \subsection{Systematics} The dispersions of the resultant parameters from both the jointly fit and individual fit are also listed in Table.~3, which could be used to denote the systematic errors of the corresponding parameters. The typical errors of the parameters are about 10\,per\,cent. Some of the parameters, i.e. the thin disk scale length $H_1$ and the thick disk local normalised densities ratio $f_2$, have relatively larger uncertainties, which are mainly due to the limits of our data and the degeneracies between different parameters. Other dominant sources of the errors are (in order of decreasing importance): 1) the systematic distance determination uncertainties, 2) the misidentification of binaries as single stars, 3) the value of distance error, i.e., finite width of the photometric parallax relation, 4) the contamination of non-dwarf stars, and 5) the effects from disk warp and flare. Finally, the structure parameters for different stellar populations (colour bins) would be intrinsically different, which may also contribute to the dispersions. Due to the absolute calibration errors of the photometric parallax relation, the distances of stars could be systematically over- or under- estimated. To check this effects, we redo the fit by changing the distance scale by 15\,per\,cent, i.e., using the parallax relations 0.3\,mag brighter and 0.3\,mag fainter than Equation~(5). The relative differences between the original resultant parameters and those derived after changing the distance scale by 15\,per\,cent are about 10 -- 15\,per\,cent. Comparing to the single stars of the same colour, the binaries are brighter, i.e. of smaller absolute magnitudes. Thus if one model the Galaxy with no or less fraction of binaries, the resultant model would be more `compact' than it truly is. In the current work we select a binary fraction $f_b$=40\,per\,cent, which is an average binary fraction for field FGK stars \citep{Yuan2015b}. To check the possible effects of the binary fraction, we have tried two extreme values, the lowest value, $f_b$=0, assuming that all stars are single stars and the highest value, $f_b$=1, assuming that all stars are binaries, to redo the fit. The relative differences between the original resultant parameters and those derived after changing the binary fraction to the extreme values are about 10\,per\,cent. When simulating the star counts for each colour bin, we assumed an error of distances of 15\,per\,cent, which includes the effect of photometric uncertainties, dispersion of metallicities of disk stars and uncertainties of the \citet{Ivezic2008} photometric parallax relation. We redo the fit by changing the distance dispersion to 0 and 30\,per\,cent, respectively. The results show that changing the uncertainties of distance do not introduce any significant bias in the derived Galactic model parameters. The effects for the model parameters caused by the non-dwarf (i.e. subgiants and giants) contaminations are similar as the binaries. From the Galaxy stellar population synthesis models, BESANCON \citep{Robin2003} and TRILEGAL \citep{Girardi2005}, we find that the fraction of giants in the three chosen colour bins is no more than 1\,per\,cent and the fraction of sub-giants is less than 5\,per\,cent. Thus the systematics caused by the giant and sub-giant contaminations are likely to be negligible. In the current work we have ignored the effects of disk flare, as we assume that the stellar flare is becoming significant at further distances \citep{Lopez2014}. While \citet{Derriere2001} and Amores et al. (2016, submitted) find that the flare starts at about 9 to 10\,kpc. Having a shorter start of the flare could have an impact on the thin ansd thick disk scale lengths that we determine, which would make the values to be underestimated. \subsection{Comparisons with other work} As stated in Section~4.1, our results of the halo model parameters are very similar as those derived from \citet{Juric2008}, because of the similar method (stellar number density fits) and data (the SDSS) adopted in both work. The main differences in the halo fit between those from \citet{Juric2008} and our work are that they adopt the $\chi^2$ fitting while we use the reduced likelihood $L_r$; and they calculate the number densities for the entire sample in $(R,~Z)$ space while we do that separately for each line of sight. The result in our work confirms those from \citet{Juric2008}. When constraining the disk model parameters, our method and data are both quite different from those in \citet{Juric2008} but the results are in agreement at a level of about 10\,per\,cent. However, \citet{Juric2008} admit that their result suffers large uncertainties from the uncertainty in calibration of the photometric parallax relation and the poor sky coverages for the low Galactic latitudes, which is not the case in our work. Generally, our derived value of thin disk scale height, 322\,pc, is in the range of values, 150--360\,pc, which resulted from the recent work by \citet{Bilir2008, Juric2008, Yaz2010, Chang2011, Polido2013, Jia2014} and \citet{Lopez2014}. Specially, this value is in good agreement with the canonical value of 325\,pc \citep{Gilmore1984, Yoshii1987, Reid1993, Larsen1996}. Our derived value of the thin disk scale length, 2.3\,kpc, is consistent with those results found by \citet{Ojha1996, Robin2000, Chen2001, Siegel2002, Karaali2007, Juric2008, Robin2012, Polido2013, Lopez2014} and \citet{Yaz2015}, which ranges between 2 and 3\,kpc. We find a local thick disk normalization of 11 per\,cent. In the different colour bins, this value varies between 7 and 14\,per\,cent, probably because of the degeneracy with the scale height. It is well in agreement with the values in the literature, which ranges between 7 and 13\,per\,cent \citep{Chen2001, Siegel2002, Cabrera2005, Juric2008, Chang2011, Jia2014}. The range of values for the thick disk scale height and scale length from the recent literature are respectively 600 -- 1000\,pc and 3 -- 5\,kpc \citep{Bilir2008, Juric2008, Yaz2010, Chang2011, Polido2013, Jia2014, Lopez2014, Robin2014}. The results deduced here, which have thick disk scale height of $\sim$800\,pc and scale length of 3.6\,kpc, are both in the middle of the ranges of those values reported in the literature. Notice that there is a significant difference with the \citet{Robin2014} result for the thick disc. Their thick disk is modelled with 2 episodes, one of which has very similar parameters as the present result (their old thick disk). But their young thick disk is more compact, with smaller scale height and scale length. The difference can be due to the different shapes used (they use secant squared density laws and they include the flare) while in the present study the thick disk is a simple exponential vertically. Based on the data from the XSTPS-GAC and the SDSS, we have modelled the global smooth structure of the Milky Way. We adopt a three-component stellar distribution model. It comprises two double exponential disks, the thin disk and the thick disk, and a two-axial power-law ellipsoid halo. The stellar number density of halo stars in the colour bin $0.5 < (g-i)_0 < 0.6$\,mag and the $r$-band differential star counts in three colour bins, $0.5 < (g-i)_0 < 0.6$\,mag, $0.6 < (g-i)_0 < 0.7$\,mag and $1.5 < (g-i)_0 < 1.6$\,mag, are used to determine the Galactic model parameters. The best-fit values are listed in Table~2 and 3. In summary, the scale height and length of the thin disk are $H_1$=322\,pc and $L_1$=2343\,pc, and those of the thick disk are $H_2$=794\,pc and $L_2$=3638\,pc. The local stellar density ratio of thick-to-thin disk is $f_2$=11\,per\,cent, and that of halo-to-thin disk is $f_h$=0.16\,per\,cent. The axis ratio and power-law index of the halo are $\kappa=0.65$ and $p=2.79$. Our results are all well constrained and in good agreement with the previous works. By subtracting the observations from our best-fit model, we find three large overdensities. Two of them have been previously identified, including the Virgo overdensity in the Halo \citep{Juric2008}, which located at 240\degr\ $<l<$ 330\degr\ and 60\degr\ $<b<$ 90\degr\ with a distance between 9 -- 11\,kpc, and the so-called ``north near structure'' in the disk \citep{Xu2015}, which located at 170\degr\ $<l<$ 200\degr\ and 10\degr\ $<b<$ 30\degr\ with a distance between 1.5 -- 2\,kpc. The third structure, located at 150\degr\ $<l<$ 210\degr\ and $-$15\degr\ $<b<$ $-5$\degr\ with a distance between 1 -- 1.5\,kpc, is a new identification. Through the Hess diagram examination, we conclude that it could not be a artifact caused by extinction correction or selection effects. This feature, together with the ``north near structure'' confirms the earlier discovery of \citet{Widrow2012} and \citet{Yanny2013} of a significant Galactic North-South asymmetry in the stellar number density distribution. | 16 | 9 | 1609.08838 |
1609 | 1609.01291_arXiv.txt | We present the results of our ALMA observations of three AGN-dominated nuclei in optical Seyfert 1 galaxies (NGC 7469, I Zw 1, and IC 4329 A) and eleven luminous infrared galaxies (LIRGs) with various levels of infrared estimated energetic contributions by AGNs at the HCN and HCO$^{+}$ J=3--2 emission lines. The HCN and HCO$^{+}$ J=3--2 emission lines are clearly detected at the main nuclei of all sources, except for IC 4329 A. The vibrationally excited (v$_{2}$=1f) HCN J=3--2 and HCO$^{+}$ J=3--2 emission lines are simultaneously covered, and HCN v$_{2}$=1f J=3--2 emission line signatures are seen in the main nuclei of two LIRGs, IRAS 12112$+$0305 and IRAS 22491$-$1808, neither of which show clear buried AGN signatures in the infrared. If the vibrational excitation is dominated by infrared radiative pumping, through the absorption of infrared 14 $\mu$m photons, primarily originating from AGN-heated hot dust emission, then these two LIRGs may contain infrared-elusive, but (sub)millimeter-detectable, extremely deeply buried AGNs. These vibrationally excited emission lines are not detected in the three AGN-dominated optical Seyfert 1 nuclei. However, the observed HCN v$_{2}$=1f to v=0 flux ratios in these optical Seyferts are still consistent with the intrinsic flux ratios in LIRGs with detectable HCN v$_{2}$=1f emission lines. The observed HCN-to-HCO$^{+}$ J=3--2 flux ratios tend to be higher in galactic nuclei with luminous AGN signatures compared with starburst-dominated regions, as previously seen at J=1--0 and J=4--3. | According to widely accepted cold dark matter-based galaxy formation scenarios, small gas-rich galaxies collide, merge, and then evolve into massive galaxies, as seen in the present day universe \citep{whi78}. Recent observations have shown that supermassive black holes (SMBHs) are ubiquitously present in the spheroidal components of present-day galaxies, and that there is a correlation between the mass of SMBHs and the spheroidal stellar components \citep{mag98,fer00,gul09,mcc13}. When gas-rich galaxies containing SMBHs at their center collide and merge, both active star formation and mass accretion onto SMBHs (= active galactic nucleus (AGN) activity) are predicted to occur, but while deeply embedded in dust and gas \citep{hop05,hop06}. Observations at low dust extinction wavelengths are necessary to investigate these types of obscured activity in gas-rich galaxy mergers. Molecular rotational J-transition line flux ratios in the (sub)millimeter wavelength range can be a powerful tool for this purpose because dust extinction effects are usually negligible, unless obscuration is extremely high with a hydrogen column density N$_{\rm H}$ $>>$ 10$^{25}$ cm$^{-2}$ \citep{hil83}. In particular, interferometric observations can probe the properties of nuclear molecular gas in the close vicinity of an AGN, by minimizing contamination from spatially extended (kpc-scale) starburst emission in the host galaxy, and have provided an indication, based on the observations of nearby bright starburst and Seyfert (= modestly luminous AGNs) galaxies, that enhanced HCN J=1--0 emission could be a good AGN signature \citep{koh05,kri08}. Based on this, interferometric HCN and HCO$^{+}$ J=1--0 observations have been extensively performed for nearby gas-rich merging luminous infrared galaxies (LIRGs; infrared 8--1000 $\mu$m luminosity L$_{\rm IR}$ $>$ 10$^{11}$L$_{\odot}$), and it has been confirmed that LIRGs with infrared-identified energetically important obscured AGN signatures tend to display higher HCN-to-HCO$^{+}$ flux ratios than starburst-dominated LIRGs \citep{ima04,ima06b,in06,ima07a,ima09}. HCN and HCO$^{+}$ have similar dipole moments ($\mu$ = 3.0 debye and 3.9 debye, respectively), so that they also have similar critical densities at the same J-transitions. It is expected that HCN and HCO$^{+}$ emission arises from similar regions inside galaxies. Interpretations of the observed HCN-to-HCO$^{+}$ flux ratios are less ambiguous than the flux comparison among molecules with largely different dipole moments (e.g., HCN vs. CO). Using ALMA, a similar enhancement of observed HCN-to-HCO$^{+}$ flux ratios was also found at J=4--3 for LIRGs that are infrared-diagnosed to be AGN-important \citep{ima13a,ima13b,ima14b,ion13,gar14,izu15,izu16}. These observations suggest that elevated HCN-to-HCO$^{+}$ flux ratios could be used to identify AGNs, including deeply buried (= obscured in virtually all directions) ones. The physical origin of HCN flux enhancement in an AGN is not yet completely understood. Compared with a starburst (nuclear fusion), an AGN (mass accretion onto a SMBH) shows stronger X-ray emission when normalized to the ultraviolet luminosity \citep{sha11,ran03}. This strong X-ray emission may enhance the HCN abundance, relative to HCO$^{+}$ \citep{mei05,lin06}, and may be responsible for the enhanced HCN emission in AGNs. Next, because the radiative energy generation efficiency of a mass-accreting SMBH in an AGN (6--42\% of Mc$^{2}$) is much higher than the nuclear fusion reaction inside stars in a starburst ($\sim$0.7\% of Mc$^{2}$), an AGN can produce much higher surface brightness emission, and thereby a larger amount of hot dust ($>$ 100 K) in its close vicinity, than a starburst. HCN abundance enhancement due to high gas/dust temperature chemistry \citep{har10} may also be the cause of the HCN flux enhancement in AGNs. It is also argued that the HCO$^{+}$ abundance can decrease in highly turbulent molecular gas in the close vicinity of a strongly X-ray-emitting AGN \citep{pap07}, which may result in an elevated HCN-to-HCO$^{+}$ abundance ratio. In fact, \citet{yam07} and \citet{izu16} made non-LTE calculations of HCN and HCO$^{+}$ emission over a wide parameter range and found that an enhanced HCN abundance is required to account for the high HCN-to-HCO$^{+}$ flux ratios observed in AGNs. HCN flux enhancement by infrared radiative pumping \citep{car81,ziu86,aal95,gar06,sak10} is also suggested. Namely, HCN can be vibrationally excited by infrared 14 $\mu$m photons that are strongly emitted from AGN-heated hot dust, and through its decay back to the vibrational ground level (v=0) the HCN rotational J-transition flux at v=0 could be higher than that of collisional excitation alone \citep{ran11}. However, chemical models also predict that the HCN and HCO$^{+}$ abundances around an AGN can vary strongly, depending on the surrounding molecular gas parameters \citep{mei05,har13}. The proposed decrease in HCO$^{+}$ abundance could also occur in turbulent molecular gas in a starburst with strong cosmic rays \citep{pap07}. In addition, HCO$^{+}$, as well as HCN, can be vibrationally excited by absorbing infrared 12 $\mu$m photons \citep{dav84,kaw85}, and the HCO$^{+}$ J-transition flux at v=0 could also be increased through the action of infrared radiative pumping. Given these remaining theoretical ambiguities, further detailed interferometric observations of galactic nuclei where the energetic roles of AGNs are reasonably well estimated are an important step toward better clarifying whether elevated HCN-to-HCO$^{+}$ flux ratios can indeed operate as good indicators of AGNs. In light of this, we conducted ALMA Cycle 1 HCN J=3--2 (with a rest-frame frequency of $\nu_{\rm rest}$ = 265.89 GHz) and HCO$^{+}$ J=3--2 ($\nu_{\rm rest}$ = 267.56 GHz) observations of AGN-dominated nuclei in band 6 (211--275 GHz). These high-spatial-resolution ALMA observations enable us to minimize the contamination from spatially extended starburst activity in host galaxies, and thus provide a clearer view on whether AGNs indeed show enhanced HCN-to-HCO$^{+}$ flux ratios, compared to starburst galaxies. As spatially resolved pre-ALMA HCN and HCO$^{+}$ J=3--2 data for galaxies are still very limited in the literature \citep{sak10,hsi12,aal15a}, our ALMA Cycle 2 and 3 LIRG data, with various AGN and starburst contributions, are also included so that we can compare the HCN-to-HCO$^{+}$ J=3--2 flux ratios of AGN-dominated nuclei to regions of strong starburst contributions. Observations of HCN and HCO$^{+}$ at J=3--2 have several important advantages compared with other J-transition lines. First, our main targets are nearby LIRGs, AGNs, and starbursts, whose redshifts are as large as $z \sim$ 0.3. HCN and HCO$^{+}$ J=3--2 lines can be observed simultaneously for these targets using ALMA. We note that (1) HCN and HCO$^{+}$ J=1--0 lines cannot be covered by ALMA if the target redshifts exceed $z \sim$ 0.06, because these lines are shifted beyond the frequency coverage of ALMA band 3 (84--116 GHz), and (2) HCN and HCO$^{+}$ J=2--1 line observations of nearby galaxies are impossible in ALMA Cycles 1, 2, and 3, because ALMA band 5 (163--211 GHz) is not yet open. Second, compared with HCN/HCO$^{+}$ J=4--3 observations in ALMA band 7 (275--373 GHz) for nearby galaxies, HCN/HCO$^{+}$ J=3--2 observations in band 6 are less affected by the Earth's atmospheric background noise and thus enable us to obtain higher quality, higher signal-to-noise (S/N) ratio data. Third, the effects of the precipitable water vapor value at the ALMA observing site are smaller in band 6 than in band 7, so that the probability of observation execution is expected to be higher for HCN/HCO$^{+}$ J=3--2 in band 6 than HCN/HCO$^{+}$ J=4--3 in band 7. Finally, in ALMA band 6, in addition to the HCN J=3--2 and HCO$^{+}$ J=3--2 lines, vibrationally excited v$_{2}$=1, l=1f (hereafter v$_{2}$=1f) emission lines of HCN ($\nu_{\rm rest}$ = 267.20 GHz) and HCO$^{+}$ ($\nu_{\rm rest}$ = 268.69 GHz) can also be simultaneously covered in one shot, with the 5 GHz-wide correlator unit. These v$_{2}$=1f emission lines can be used to investigate how infrared radiative pumping works in observed galaxies and affects rotational excitation at v=0 \citep{sak10,ima13b,aal15a,aal15b,ima16a,mar16}. The simultaneous observations of all of these lines make their flux ratios reliable, with the effect of possible systematic uncertainties being minimized. In this paper, we report the results of our ALMA Cycle 1, 2, and 3 observations of AGN-dominated galactic nuclei, starburst-dominated regions, and the nuclei of AGN-starburst composite LIRGs. Throughout this paper, we adopt H$_{0}$ $=$ 71 km s$^{-1}$ Mpc$^{-1}$, $\Omega_{\rm M}$ = 0.27, and $\Omega_{\rm \Lambda}$ = 0.73 \citep{kom09}. Molecular lines without the notation of v (the vibrational level) refer to v=0 (i.e., the vibrational ground level). HCN refers to H$^{12}$C$^{14}$N. | \subsection{Vibrationally excited HCN/HCO$^{+}$ J=3--2 emission lines} \subsubsection{Luminous infrared galaxies} Because the energy levels of the vibrationally excited (v$_{2}$=1) state for HCN ($\sim$1030 K) and HCO$^{+}$ ($\sim$1200 K) are too high to be excited by collisions, infrared radiative pumping is thought to be necessary for vibrational excitation \citep{sak10}. Due to a large amount of AGN-heated hot ($>$ few 100 K) dust emission, the 14 $\mu$m luminosity in an AGN is significantly higher than that in a starburst for the same bolometric luminosity \citep{mar07,veg08}. Thus, HCN vibrational excitation, through the absorption of infrared 14 $\mu$m photons, is expected to occur more efficiently in an AGN than in a starburst. In fact, the HCN v$_{2}$=1f emission lines at J=3--2 and/or J=4--3 have recently been detected in gas/dust-rich LIRGs, which most likely, or plausibly, contain luminous AGNs, i.e., NGC 4418 \citep{sak10,cos15}, IRAS 20551$-$4250 \citep{ima13b,ima16a}, Mrk 231 \citep{aal15a}, and a few further LIRGs \citep{aal15b,mar16}, demonstrating that this infrared radiative pumping mechanism actually works in some AGNs. As the frequencies of HCN v$_{2}$=1f and HCO$^{+}$ v=0 are very close to each other at J=3--2 and J=4--3, we can only clearly separate these lines for galaxies with small molecular line widths, such as NGC 4418 \citep{sak10}, IRAS 20551$-$4250 \citep{ima13b,ima16a}, and IC 860 \citep{aal15b}. For the majority of the other galaxies, these lines are blended. Even if the HCN v$_{2}$=1f J=3--2 or J=4--3 emission line is detected, it is recognized as a tail at the lower frequency side of the much brighter HCO$^{+}$ v=0 J=3--2 or J=4--3 emission line \citep{aal15a,aal15b,mar16}. The four ULIRGs, IRAS 08572$+$3915, IRAS 12112$+$0305 NE, IRAS 22491$-$1808, and IRAS 20414$-$1651, show these profiles. In particular, IRAS 12112$+$0305 NE and IRAS 22491$-$1808 can be categorized as sources that display detectable HCN v$_{2}$=1f J=3--2 emission lines, given the $>$4$\sigma$ detection in the moment 0 maps (Figure 12). For IRAS 08572$+$3915 and IRAS 22491$-$1808, similar HCN v$_{2}$=1f J=4--3 emission tails were not clearly seen in our ALMA Cycle 0 band 7 data \citep{ima14b}; however, this is not surprising due to the improved performance of ALMA Cycle 2 data and intrinsically lower noise in band 6 with lower background emission, than in band 7. For IRAS 12112$+$0305 NE, a similar signature of the HCN v$_{2}$=1f J=4--3 emission line at the lower frequency side of HCO$^{+}$ v=0 J=4--3 was observed in our ALMA Cycle 2 data, despite a lower detection significance than J=3--2, while no significant emission tail is recognizable at the lower frequency side of HCN v=0 J=4--3 \citep{ima16c}. While IRAS 08572$+$3915 is classified as a ULIRG possessing a luminous buried AGN in the infrared spectrum, IRAS 12112$+$0305 NE, IRAS 22491$-$1808, and IRAS 20414$-$1651 display no clear infrared buried AGN signatures. The signatures of the HCN v$_{2}$=1f J=3--2 emission lines in our data suggest the presence of strong mid-infrared 14 $\mu$m continuum-emitting sources at the nuclei of these two LIRGs. The HCN v$_{2}$=1f J=3--2 to infrared luminosity ratios are $>$7 $\times$ 10$^{-9}$ and $\sim$9 $\times$ 10$^{-9}$ for IRAS 12112$+$0305 NE and IRAS 22491$-$1808, respectively. These ratios are several factors higher than that in the Galactic active ($>$10 L$_{\odot}$/M$_{\odot}$) and luminous ($>$10$^{7}$L$_{\odot}$) star-forming region, W49A ($<$1.2 $\times$ 10$^{-9}$) \citep{nag15,ima16a}. A luminous buried AGN is a plausible origin, although the possibility of a very compact extreme starburst cannot be completely ruled out \citep{aal15b}. IRAS 12112$+$0305 NE and IRAS 22491$-$1808 are candidates that contain extremely deeply buried AGNs whose signatures are not seen in infrared 5--35 $\mu$m spectroscopic energy diagnostic methods due to dust extinction, but are revealed by our (sub)millimeter method because of the reduced effects of dust extinction \citep{dra84}. If this is the case, (sub)millimeter observations could be an even more powerful method for detecting extremely deeply buried AGNs in LIRGs. IRAS 20414$-$1651 may also belong to this class, but higher quality data are needed to quantitatively better estimate the HCN v$_{2}$=1f J=3--2 emission line luminosity. The HCO$^{+}$ v$_{2}$=1f J=3--2 emission line is not clearly detected in any of the observed three optical Seyfert 1s and eleven LIRGs. Like HCN, infrared radiative pumping should also work for HCO$^{+}$, because HCO$^{+}$ can be excited to the v$_{2}$=1 level by absorbing infrared 12 $\mu$m photons \citep{dav84,kaw85}. The infrared radiative pumping rate (P$_{\rm IR}$) is \begin{eqnarray} P_{IR} & \propto & B_{v2=0-1,vib} \times F_{\nu (IR)} \times N_{v=0}, \end{eqnarray} where B$_{v2=0-1,vib}$ is the Einstein B coefficient from v=0 to v$_{2}$=1, F$_{\nu (IR)}$ is the infrared flux in [Jy] used for the infrared radiative pumping of HCN and HCO$^{+}$, and N$_{v=0}$ is the column density at the v=0 level. Here, the possible difference in population between HCN and HCO$^{+}$ at rotational J-levels within v=0 and v$_{2}$=1 is not considered. As discussed in \citet{ima16a}, the B$_{v2=0-1,vib}$ values are comparable within 10\% between HCN and HCO$^{+}$. The F$_{\rm \nu (IR)}$ values at 12 $\mu$m and 14 $\mu$m for the observed galaxies are derived from their Spitzer IRS low-resolution spectra \citep{bra06,ima07b,wu09}. For sources with strong 9.7 $\mu$m silicate dust absorption features, power law continua determined from data points outside the broad 9.7 $\mu$m absorption features are utilized to estimate the intrinsic infrared flux at 14 $\mu$m and 12 $\mu$m, which are used for the vibrational excitation to v$_{2}$=1 of HCN and HCO$^{+}$, respectively. In none of the galaxies was the intrinsic 14 $\mu$m flux $>$30\% larger than the intrinsic 12 $\mu$m flux. Therefore, at least for the observed galaxies in this paper, the term B$_{v2=0-1,vib}$ $\times$ F$_{\nu (IR)}$ does not differ greatly between HCN and HCO$^{+}$ and thereby, under similar HCN and HCO$^{+}$ abundance, the infrared radiative pumping rate is comparable between HCN and HCO$^{+}$. If the flux of the HCN v$_{2}$=1f J=3--2 emission line is significantly higher than that of HCO$^{+}$ v$_{2}$=1f J=3--2, HCN should then have a significantly higher column density, and thereby a higher abundance, than HCO$^{+}$, unless excitation conditions at J=3 significantly differ between HCN and HCO$^{+}$ \citep{ima16a}. The upper limit of the HCO$^{+}$ v$_{2}$=1f J=3--2 flux (Table 6) is only 10--20\% lower than the flux of the HCN v$_{2}$=1f J=3--2 emission lines, even for IRAS 12112$+$0305 NE and IRAS 22491$-$1808. Thus, our only constraint is that the HCN abundance is at least comparable to HCO$^{+}$ or possibly higher. \subsubsection{Optical Seyfert galaxies} The original science goal of our ALMA Cycle 1 program was to detect the HCN v$_{2}$=1f J=3--2 emission lines from the AGN-dominated nuclear regions of the three Seyfert 1 galaxies, NGC 7469, I Zw 1, and IC 4329 A. In NGC 7469 and I Zw 1, the clear detection of the HCN and HCO$^{+}$ J=3--2 emission lines at v=0 suggests that at least a modest amount of dense molecular gas is present at the nuclei. If the infrared radiative pumping mechanism is commonly working in AGNs, it is expected that a number of HCN and HCO$^{+}$ v$_{2}$=1f emission lines are produced. We did not detect HCN/HCO$^{+}$ v$_{2}$=1f J=3--2 emission lines in these AGN-dominated Seyfert 1 nuclei, and this requires some quantitative consideration. For the NGC 7469 nucleus, the HCN v$_{2}$=1f J=4--3 emission line was also undetected by ALMA observations \citep{izu15}. For the non-detected v$_{2}$=1f J=3--2 emission lines of HCN and HCO$^{+}$, we use the 3$\sigma$ upper limits from the moment 0 maps tabulated in Table 6 for our discussion. As there is no existing report for the detection of the HCO$^{+}$ v$_{2}$=1f J=3--2 emission line in external galaxies, we focus here on the HCN v$_{2}$=1f J=3--2 line. The observed v$_{2}$=1f to v=0 flux ratios at J=3--2 for HCN are $<$0.02 and $<$0.04 for NGC 7469 and I Zw 1 nuclei, respectively. For LIRGs with detected HCN v$_{2}$=1f emission lines, the observed HCN v$_{2}$=1f to v=0 flux ratios at J=3--2 or J=4--3 are $\sim$0.04 in IRAS 20551$-$4250 and Mrk 231 \citep{ima13b,aal15a,ima16a}, and 0.1--0.2 for the other sources \citep{sak10,aal15b,mar16}. The observed ratios in NGC 7469 ($<$0.02) and I Zw 1 ($<$0.04) are lower than these ratios. We consider that a plausible scenario for the non-detection of the HCN v$_{2}$=1f J=3--2 emission line in NGC 7469 and I Zw 1 is the small line opacity of the HCN v=0 J=3--2 emission. Thus far, the HCN v$_{2}$=1f J=3--2 or J=4--3 emission lines have been detected in LIRGs with buried AGNs whose signatures are unclear in the optical spectroscopic classification, except for Mrk 231 \citep{sak10,ima13b,aal15a,aal15b,mar16}. Although Mrk 231 is classified optically as a Seyfert 1 galaxy due to the detection of broad optical emission lines \citep{vei99,yua10}, the AGN emission in Mrk 231 is estimated to be highly obscured in infrared and X-ray data \citep{arm07,ten14}. It is likely that Mrk 231 is not a bona fide unobscured Seyfert 1 galaxy, but rather an obscured AGN. For the obscured AGN-hosting LIRGs with detectable HCN v$_{2}$=1f emission lines, significant flux attenuation by line opacity of the HCN v=0 emission is indicated \citep{sak10,aal15a,aal15b,ima16a,mar16}. For NGC 4418, IRAS 20551$-$4250, and Mrk 231, the line-opacity-corrected intrinsic HCN v$_{2}$=1f to v=0 flux ratios are quantitatively estimated to be $\sim$0.01 \citep{sak10,aal15a,ima16a}. These ratios are smaller than the upper limits at the NGC 7469 and I Zw 1 nuclei. NGC 7469 and I Zw 1 are classified optically as Seyfert 1 (= unobscured AGNs); thus, the direction along our line of sight in front of the AGN is at least clear of gas and dust. It is likely that molecular gas and dust are present in the close vicinity of the AGNs in the direction perpendicular to our sightline. If the ratio of rotational to random velocity of molecular gas does not differ greatly between unobscured optical Seyfert 1 AGNs and buried AGNs in LIRGs, the column density ratio along the maximum and minimum column density directions is not dissimilar. The presence of a transparent direction suggests that the total amount of nuclear molecular gas in unobscured AGNs is smaller than buried AGNs in LIRGs \citep{ima07b,ima10}. Thus, the flux attenuation of the HCN and HCO$^{+}$ v=0 emission by line opacity is also expected to be smaller in unobscured AGNs. \citet{izu15} estimated the line opacity for the HCN v=0 J=4--3 emission to be $<$3.5 for the NGC 7469 nucleus. Even if unobscured AGNs and buried AGNs show intrinsically similar HCN v$_{2}$=1f to v=0 flux ratios, the {\it observed} HCN v$_{2}$=1f to v=0 flux ratios in buried AGNs can become larger due to higher HCN v=0 flux attenuation. The upper limits of the observed HCN v$_{2}$=1f to v=0 flux ratios at J=3--2 at the NGC 7469 and I Zw 1 nuclei are still consistent with the scenario that the efficiency of infrared radiative pumping in these unobscured-AGN-dominant nuclei is as high as that of buried AGNs with detected HCN v$_{2}$=1f emission lines. If this scenario is indeed the case, then HCN v$_{2}$=1f J=3--2 emission lines should be detected from the nuclei of NGC 7469 and I Zw 1 in data with a factor of 5--10 better sensitivity, even in the case that the line opacity correction of HCN v=0 J=3--2 emission is negligible. Future higher sensitivity observations and line opacity estimates for the HCN v=0 J=3--2 emission line will help to quantify how infrared radiative pumping works in various types of AGNs, including unobscured AGNs in optical Seyfert 1 galaxies and buried AGNs in LIRGs. \subsection{HCN to HCO$^{+}$ J=3--2 flux ratios} \subsubsection{Observed ratios} The HCN-to-HCO$^{+}$ J=3--2 flux ratios at individual positions in individual galaxies are displayed in Figure 14. The NGC 7469 nucleus and I Zw 1 are classified as Seyfert 1s. In NGC 7469, SB1, SB2, SB3, and the SB ring (0$\farcs$8--2$\farcs$5 annular region) are taken to be starburst-dominated. For NGC 1614, all regions are regarded as starburst-dominated. Among the other LIRGs, IRAS 08572$+$3915, The Superantennae, IRAS 12127$-$1412, IRAS 15250$+$3609, PKS 1345$+$12, and IRAS 06035$-$7102 are categorized as obscured AGNs, based on infrared spectroscopic energy diagnostic methods ($\S$2). IRAS 12112$+$0305 NE and IRAS 22491$-$1808 are now classified as infrared-elusive, but (sub)millimeter-detectable, extremely deeply buried AGN candidates ($\S$5.1.1). We tentatively include IRAS 20414$-$1651 in this category as well, because its spectrum in Figure 10 shows a more clearly discernible HCN v$_{2}$=1f J=3--2 emission signature at the lower frequency part of the bright HCO$^{+}$ v=0 J=3--2 emission than other ULIRGs (IRAS 12112$+$0305 SW, IRAS 12127$-$1412, PKS 1345$+$12, IRAS 06035$-$7102, IRAS 13509$+$0442). IRAS 12112$+$0305 SW and IRAS 13509$+$0442 show no AGN signature in either infrared or our new ALMA (sub)millimeter data. Our ALMA Cycle 2 results of the buried-AGN-hosting ULIRG IRAS 20551$-$4250 \citep{ima16a}, and multiple AGN-dominated nuclear regions of the optical Seyfert 2 galaxy, NGC 1068 \citep{ima16b}, are also plotted. In addition to these ALMA data, HCN J=3--2 and HCO $^{+}$ J=3--2 simultaneous observational data for NGC 4418 \citep{sak10} and NGC 1097 \citep{hsi12} taken with Submillimeter Array (SMA), and those for Mrk 231 \citep{aal15a} obtained with IRAM Plateau de Bure Intermerometer (PdBI), are added, by classifying NGC 4418, the NGC 1097 nucleus, the NGC 1097 starburst ring, and Mrk 231 as a buried AGN, Seyfert 1, starburst, and an obscured AGN, respectively. In Figure 14, we see a clear trend for AGNs, including infrared-elusive buried AGN candidates, to show elevated HCN-to-HCO$^{+}$ J=3--2 flux ratios, compared with starburst regions. Multiple starburst regions in NGC 7469 and NGC 1614, and other starburst galaxies (IRAS 12112$+$0305 SW, IRAS 13509$+$0442, and NGC 1097 off-nuclear starburst) consistently show low HCN-to-HCO$^{+}$ J=3--2 flux ratios. Hence, the low ratios are interpreted to be a general property of starbursts, rather than a specific property of a particular starburst region. The excess of the flux ratios in AGN-dominant nuclei in optical Seyferts and LIRGs with luminous obscured AGN signatures, compared with starbursts, is taken to be a robust result. A similar HCN-to-HCO$^{+}$ flux enhancement in AGNs at J=1--0 has been proposed \citep{koh05,kri08,cos11,pri15,izu16}, and was the basis for our ALMA observations. The enhanced HCN-to-HCO$^{+}$ flux ratios in AGNs appear to be common at different J-transition lines. As mentioned in $\S$1, the HCN-to-HCO$^{+}$ flux comparison at J=3--2 is applicable to many interesting nearby LIRGs at z=0.06--0.3 \citep{kim98}, whose HCN and HCO$^{+}$ observations at J=1--0 are not possible with ALMA. \subsubsection{Interpretation} The observed HCN-to-HCO$^{+}$ flux enhancement in AGNs, compared with starbursts, is naturally explained if (1) the HCN abundance is enhanced and/or (2) HCN excitation to J=3 is higher, as discussed by \citet{ima16a}. Regarding scenario (1), it is clear that enhanced molecular abundance generally produces a higher flux of that molecule in the optically thin regime. When line opacity becomes significant, the emission line flux does not increase proportionally to the increased abundance. However, adopting the widely accepted clumpy molecular gas model, where molecular clouds consist of randomly moving clumps with a small volume filling factor and the line opacity is primarily inside each clump, rather than different clumps in the foreground with different velocities \citep{sol87}, an increasing HCN abundance will result in an increased HCN flux even in the optically thick regime, if each clump has a decreasing radial density profile \citep{ima07a}. This behavior differs from molecular clouds with a smooth gas distribution, where the observed molecular line flux saturates at some point when the opacity exceeds a certain threshold. Regarding scenario (2), the critical density of HCN is a factor of $\sim$5 higher than HCO$^{+}$ at the same J-transition \citep{mei07,gre09}, under the same line opacity. For molecular gas with the same temperature and density, HCO$^{+}$ J=3--2 is more easily excited (to close to the thermalized condition) than HCN J=3--2. As the temperature of molecular gas in the close vicinity of an AGN can be higher than that in a starburst due to the AGN's higher emission surface brightness ($\S$1), if HCO$^{+}$ J=3--2 is thermally excited and HCN J=3--2 is only sub-thermally excited in some starbursts, and if AGNs can excite HCN J=3--2 closer to the thermal condition, then the observed HCN-to-HCO$^{+}$ J=3--2 flux ratios in AGNs can show some excess, relative to those in starbursts, even under similar abundance. Among the galaxies observed in our ALMA programs, HCN J=1--0 flux data are available for IRAS 08572$+$3916 by pre-ALMA interferometric observations \citep{ima07a} and for I Zw 1 by single-dish telescope observations \citep{eva06}. The HCN J=3--2 to J=1--0 flux ratios are $\sim$1.1 and $\sim$2.2 for IRAS 08572$+$3915 and I Zw 1, respectively. These values are substantially lower than the ratio of nine, which is expected from thermally excited optically thick molecular gas. The deviation is larger for IRAS 08572$+$3915 than I Zw 1. For IRAS 08572$+$3915, the J=4--3 to J=3--2 flux ratios are 1.3$\pm$0.1 for HCN and 1.4$\pm$0.1 for HCO$^{+}$ \citep{ima14b}, both of which are lower than the 1.8 (=16/9) expected for thermally excited optically thick gas. The significant sub-thermal excitation of IRAS 08572$+$3915 may be partly responsible for the relatively low observed HCN-to-HCO$^{+}$ J=3--2 flux ratio, compared with other AGNs (Figure 14). It has been argued that an infrared radiative pumping mechanism can enhance the observed HCN v=0 J=3--2 flux \citep{car81,aal95,ran11}. Although this is true, such infrared radiative pumping should also work for HCO$^{+}$ ($\S$1). As described in $\S$5.1.1 and \citet{ima16a}, under the same abundance, the rate of infrared radiative pumping to the v$_{2}$=1 level does not differ a great deal between HCN and HCO$^{+}$ for the observed galaxies. Even though this infrared radiative pumping may work more efficiently in AGNs than in starbursts and increase the {\it absolute fluxes} of HCN and HCO$^{+}$ v=0 J=3--2 emission lines compared with collisional excitation alone, it is not clear whether this may be largely responsible for the elevated HCN-to-HCO$^{+}$ v=0 J=3--2 {\it flux ratios} in AGNs. \subsubsection{Line opacity} If HCN abundance enhancement is (at least partly) responsible for the elevated HCN-to-HCO$^{+}$ J=3--2 flux ratios in AGNs, the HCN line opacity could be higher than HCO$^{+}$. Even though some AGNs show high observed HCN-to-HCO$^{+}$ flux ratios, other AGNs may not, due to higher HCN flux attenuation than HCO$^{+}$ by line opacity. Thus, the selection of AGN-important galaxies, based on the observed high HCN-to-HCO$^{+}$ flux ratios, may be incomplete and miss some fraction of AGNs unless HCN line opacity is properly corrected for. In the widely accepted clumpy molecular gas model \citep{sol87} mentioned in $\S$5.2.2, the opacity in a molecular cloud mostly comes from each clump. If the properties of each clump inside a molecular cloud are assumed to be uniform \citep{sol87}, the observed molecular line flux from a molecular cloud is attenuated without significantly changing the observed velocity profiles. An effective way to estimate the HCN line opacity is the comparison of molecular isotopologues such HCN and H$^{13}$CN, assuming a certain intrinsic $^{12}$C-to-$^{13}$C abundance ratio. The H$^{13}$CN J=3--2 emission line was detected for the AGN-hosting LIRG IRAS 20551$-$4250, and it was estimated that line opacity correction causes the {\it intrinsic} HCN-to-HCO$^{+}$ J=3--2 flux ratio to be substantially larger than the {\it observed} flux ratio \citep{ima16a}. Among obscured-AGN-classified LIRGs, PKS 1345$+$12 and IRAS 06035$-$7102 also show not-as-high observed HCN-to-HCO$^{+}$ J=3--2 flux ratios in Figure 14. It is not clear at this stage whether the {\it intrinsic} HCN-to-HCO$^{+}$ J=3--2 flux ratios are similarly non-high or higher than the observed ratios for these two sources. Line opacity correction is definitely required to refine the AGN selection based on the HCN-to-HCO$^{+}$ J=3--2 flux ratios. In the nuclei of some LIRGs, the concentration of molecular gas could be extreme, and so the volume filling factor of molecular gas clumps in molecular clouds could be large. The molecular gas geometry may be better approximated by a spatially smooth distribution \citep{dow93,sco15} rather than the clumpy structure. In this case, double-peaked molecular emission line profiles could be produced through self-absorption by foreground molecular gas, which works preferentially for the most abundant central velocity component \citep{aal15b}. IRAS 12112$+$035 NE and IRAS 20414$-$1651 display such double-peaked emission line profiles for HCN J=3--2 and HCO$^{+}$ J=3--2 (Figures 5 and 9). For IRAS 12112$+$0305 NE, a similar profile is also evident at J=4--3 of HCN and HCO$^{+}$, albeit at lower S/N ratios \citep{ima16c}. The origin of this double-peaked line profile could be (1) self-absorption and/or (2) emission being dominated by molecular gas in a prominent rotating disk. For case (1), we tried single Gaussian fits using data not significantly affected by the central dips. These fits are shown as dotted curved lines in the spectra of IRAS 12112$+$0305 NE and IRAS 20414$-$1651 in Figure 9. The estimated HCN J=3--2 and HCO$^{+}$ J=3--2 emission line fluxes, based on the single Gaussian component fits, are included in Tables 4 and 5, respectively. For IRAS 12112$+$0305 NE, the fluxes based on the two Gaussian component fits of the double-peaked profiles are smaller than those of the single Gaussian fits by $\sim$25\% and $\sim$50\% for HCN J=3--2 and HCO$^{+}$ J=3--2, respectively. If these flux differences are due to self-absorption by foreground molecular gas inside IRAS 12112$+$0305 NE, then it is estimated that the HCN J=3--2 and HCO$^{+}$ J=3--2 fluxes are attenuated by a factor of $\sim$1.3 and $\sim$2, respectively. For IRAS 20414$-$1651, the fluxes based on the two Gaussian fits are smaller than those of the single Gaussian fits by a factor of 3--5 for both HCN J=3--2 and HCO$^{+}$ J=3--2. For IRAS 12112$+$0305 NE and IRAS 20414$-$1651, H$^{13}$CN J=3--2 data have been taken in ALMA Cycle 2 and been marginally detected in the moment 0 maps with 0.28 [Jy beam$^{-1}$ km s$^{-1}$] (3.2$\sigma$) and 0.20 [Jy beam$^{-1}$ km s$^{-1}$] (3.3$\sigma$) \citep{ima16c}, respectively, which correspond to the observed HCN-to-H$^{13}$CN J=3--2 flux ratios with $\sim$30 and $\sim$20. Assuming that the intrinsic $^{12}$C/$^{13}$C abundance ratios in these ULIRGs are 50--100 \citep{hen93a,hen93b,mar10,hen14} and that H$^{13}$CN J=3--2 emission is optically thin, it is suggested that HCN J=3--2 emission is flux-attenuated with a factor of 1.5--3 and 2.5--5 for IRAS 12112$+$0305 NE and IRAS 20414$-$1651, respectively. Hence, if the central dips detected in the HCN J=3--2 and HCO$^{+}$ J=3--2 emission in IRAS 12112$+$0305 NE and IRAS 20414$-$1651 are due to self-absorption, the flux attenuation estimated from the comparison of Gaussian fittings is smaller than that derived from the HCN-to-H$^{13}$CN J=3--2 flux comparison by a factor of 1--2 for IRAS 12112$+$0305 NE, while these two estimates look comparable within uncertainty for IRAS 20414$-$1651. However, despite limited signal-to-noise ratios, double-peaked emission line profiles with similar velocity peaks to the bright HCN J=3--2 and HCO$^{+}$ J=3--2 emission lines or top-flat type line profiles, rather than a centrally-peaked single Gaussian profile, are seen for the H$^{13}$CN J=3--2 and CS J=7--6 emission lines in IRAS 12112$+$0305 NE and for H$^{13}$CN J=3--2 in IRAS 20414$-$1651 \citep{ima16c}. Since the self-absorption effect is expected to be much smaller for the fainter H$^{13}$CN J=3--2 emission line than HCN J=3--2 and HCO$^{+}$ J=3--2, it is not clear whether the observed double-peaked emission line profiles detected in IRAS 12112$+$0305 NE and IRAS 20414$-$1651 are explained solely by the self-absorption. We next consider the second rotating disk scenario. In the intensity-weighted mean velocity (moment 1) maps of IRAS 12112$+$0305 NE and IRAS 22491$-$1651 in Figure 13, the signature of a rotational motion is marginally seen along the north-east to south-west direction, with a velocity difference of $\sim$200--300 km s$^{-1}$ and $\sim$400 km s$^{-1}$, respectively. They are comparable to the observed velocity difference of the double peaks in Figure 9. Figure 15 shows the spectra within the beam size, at 3 pix (0$\farcs$3) north and 3 pix (0$\farcs$3) east (i.e., $\sim$0$\farcs$4 north-east), and at 3 pix (0$\farcs$3) south and 3 pix (0$\farcs$3) west (i.e., $\sim$0$\farcs$4 south-west), relative to the continuum peak positions, for IRAS 12112$+$0305 NE and IRAS 22491$-$1651. It is shown that the red (blue) component is relatively strong at the 0$\farcs$4 north-east (south-west) position, as is expected from the moment 1 maps of both objects. We interpret that compact rotating disks which are not clearly resolved with our ALMA beam size (0$\farcs$5--0$\farcs$8) can also contribute significantly to the observed double-peaked emission line profiles in IRAS 12112$+$0305 NE and IRAS 20414$-$1651. In summary, AGNs tend to show elevated HCN-to-HCO$^{+}$ J=3--2 flux ratios, but some AGNs have non-high observed HCN-to-HCO$^{+}$ J=3--2 flux ratios. This could be explained by a larger flux attenuation caused by line opacity for HCN than HCO$^{+}$, if the HCN abundance is higher than HCO$^{+}$. In this respect, although we may be able to say that the elevated observed HCN-to-HCO$^{+}$ J=3--2 flux ratios are good AGN signatures, not all AGNs are selected based on the observed high HCN-to-HCO$^{+}$ J=3--2 flux ratios. Line opacity correction will make our method even more powerful and complete by reducing the number of missing AGNs. Clear double-peaked HCN J=3--2 and HCO$^{+}$ J=3--2 emission line profiles are seen in IRAS 12112$+$0305 NE and IRAS 20414$-$1651, which we interpret that rotating disks contribute significantly, in addition to a possible self-absorption effect. \subsection{Non-detection of molecular gas in IC 4329 A} The non-detection of the HCN J=3--2 and HCO$^{+}$ J=3--2 emission lines in IC 4329 A was unexpected. In this subsection, we briefly consider its possible causes. The three Seyfert 1 galaxies, NGC 7469, I Zw 1, and IC 4329 A, were selected because their nuclear infrared $L$-band (3--4 $\mu$m) emission is thought to be dominated by AGN-heated hot dust emission and their observed fluxes are high ($\S$2). If dust and molecular gas spatially coexist in the nuclear region, strong collisionally excited molecular gas emission is also expected there. The observed nuclear $L$-band (3--4 $\mu$m) flux of IC 4329 A is about $\sim$5 times higher than those of NGC 7469 and I Zw 1 \citep{iw04,ima11a}. At a first-order approximation, the HCN J=3--2 peak flux in IC 4329 A is expected to be higher than NGC 7469 and I Zw 1. However, the observed HCN J=3--2 emission peak is more than a factor of ten and five smaller than those of NGC 7469 and I Zw 1, respectively (Figure 4). The nuclear HCN J=3--2 emission peak is roughly predicted from the nuclear infrared emission, based on their correlation \citep{ima14b}. If we assume that the observed infrared luminosity in Table 1 originates from the nuclear region, then the expected HCN J=1--0 emission peaks are $\sim$18 mJy, $\sim$3 mJy, and $\sim$4 mJy for NGC 7469, I Zw 1, and IC 4329 A, respectively. If the HCN J=3--2 emission peak is nine times larger than that of HCN J=1--0, which is expected in thermally excited optically thick gas, then the expected HCN J=3--2 flux peaks are $\sim$165 mJy, $\sim$25 mJy, and $\sim$35 mJy, for NGC 7469, I Zw 1, and IC 4329 A, respectively. For NGC 7469, since it is estimated that about one third of the infrared luminosity originates from the nuclear region \citep{gen95}, the expected HCN J=3--2 flux peak from the NGC 7469 nucleus is $\sim$55 mJy. For the NGC 7469 nucleus and I Zw 1, the observed HCN J=3--2 peak fluxes ($\sim$25 mJy and $\sim$10 mJy, respectively) agree with the above expected values within a factor of 2--3. However, the observed HCN J=3--2 peak flux of IC 4329 A is more than an order of magnitude smaller than the above expectation. Possible explanations include (1) infrared emission is spatially extended, rendering the fraction of the nuclear component small, and (2) HCN J=3--2 is only sub-thermally excited, and the HCN J=3--2 to J=1--0 flux ratio is thus considerably (more than an order of magnitude) smaller than nine. With regard to (1), as no clear spatially extended off-nuclear emission is detected at infrared 10 $\mu$m \citep{asm14}, this seems unlikely. Regarding (2), because IC 4329 A contains a luminous X-ray-emitting AGN \citep{bri11,bre14}, this also seems unlikely for nuclear molecular gas. One scenario that could explain the weak HCN and HCO$^{+}$ J=3--2 emission in IC 4329 A is that the observed featureless nuclear infrared $L$-band (3--4 $\mu$m) continuum is not dominated by AGN-heated hot dust emission, but by other emission mechanisms such as synchrotron emission. Figure 16 displays the spectral energy distributions of NGC 7469, I Zw 1, and IC 4329 A in the infrared and radio wavelength ranges. The radio emission at $<$20 GHz is usually dominated by synchrotron emission. The q-value, defined as the decimal logarithm of the far-infrared (40--500 $\mu$m) to radio flux ratio \citep{con91}, is often used to detect radio-loud AGNs, which show stronger synchrotron emission than the majority of radio-quiet AGNs. While the q-values of NGC 7469 and I Zw 1 (Table 10) are within the range of starburst-dominated galaxies and many radio-quiet AGNs (q $\sim$ 2.3--2.4) \citep{con91,bar96,cra96,roy98}, that of IC 4329 A (q $\sim$ 1.5) (Table 10) is substantially lower, suggesting that IC 4329 A contains a radio-loud AGN. However, the observed infrared $L$-band (3--4 $\mu$m or $\sim$10$^{5}$ GHz) flux is well above the extrapolation from the synchrotron emission component at $<$20 GHz. The strong infrared excess at 200--10$^{5}$ GHz in IC 4329 A, as well as NGC 7469 and I Zw 1, suggests that their infrared $L$-band (3--4 $\mu$m) emission is dominated by AGN-heated hot dust emission \citep{alo11,ich15}, rather than synchrotron emission. Thus, this possibility also seems unlikely. A second scenario is that the column density of the obscuring gas and dust surrounding the central AGN of IC 4329 A is very small. The infrared $L$-band (3--4 $\mu$m) continuum emission primarily arises from hot dust with $>$a few 100 K, located at the innermost region of the obscuring material, with limited contribution from outer cooler dust. On the other hand, collisionally excited HCN and HCO$^{+}$ J=3--2 emission lines can still be produced at the outer regions, whose gas/dust temperature is several 10 K to a few 100 K. If the obscuring gas/dust column density around an AGN is substantially smaller in IC 4329 A than in NGC 7469 and I Zw 1, then the smaller than expected HCN and HCO$^{+}$ J=3--2 emission line fluxes from the infrared $L$-band (3--4 $\mu$m) continuum flux could thus be explained. From the infrared spectral energy distribution, \citet{ich15} estimated that the outer to innermost radius ratio of nuclear-obscuring dust in IC 4329 A is a factor of $\sim$3 smaller than NGC 7469, suggesting that this scenario is a possibility. As a third possibility, if the molecular line widths of IC 4329 A are much larger than those of NGC 7469 and I Zw 1, the observed HCN/HCO$^{+}$ J=3--2 emission peak could be small, even if their fluxes are large. If IC 4329 A follows the similar HCN J=3--2 to nuclear infrared luminosity correlation and has similar molecular line widths to NGC 7469 and I Zw 1, the HCN J=3--2 flux peak is then expected to be $>$15 mJy. To explain the actual observed HCN J=3--2 flux peak with $<$1.5 mJy, molecular line widths $>$10 times larger than those of NGC 7469 and I Zw 1, i.e., at least a FWHM $\sim$ 2000--3000 km s$^{-1}$, are required. This is more than a factor of 2 larger than the highly turbulent ongoing major merger ULIRG, The Superantennae \citep{mir91}. Because IC 4329 A is classified as a fairly settled spiral or S0 galaxy with a nuclear dust lane, but with no obvious highly disturbed morphology \citep{mal98}, such an extremely large molecular line width seems unlikely. For IC 4329 A, there has been no molecular gas detection reported in the published literature, even for CO J=1--0 and J=2--1. Future ALMA high-sensitivity observations of bright CO emission at the IC 4329 A nucleus may help to test this scenario, if detection is realized. In Figure 16, the observed flux increases with decreasing frequency from 10$^{5}$ GHz (3 $\mu$m) to 10$^{4}$ GHz (30 $\mu$m). However, the spectral energy distribution is flatter in IC 4329 A than NGC 7469 and I Zw 1, which means that the temperature of the dust thermal emission in this frequency range is higher in IC 4329 A. For given dust thermal radiation luminosity, if the dust temperature is higher, then the required dust mass can be smaller; consequently, the molecular mass becomes smaller, if dust and molecular gas spatially coexist in a similar manner. This could contribute to the observed weaker-than-expected molecular gas emission in IC 4329 A. The 2--10 keV X-ray luminosities of NGC 7469, I Zw 1, and IC 4329 A are $\sim$2 $\times$ 10$^{43}$ ergs s$^{-1}$, $\sim$8 $\times$ 10$^{43}$ ergs s$^{-1}$, and $\sim$6 $\times$ 10$^{43}$ ergs s$^{-1}$, respectively \citep{pin05,bri11,bre14} \footnote{ For these Seyfert 1 galaxies, absorption-corrected and -uncorrected 2--10 keV X-ray luminosities are comparable, due to the estimated low X-ray-absorbing hydrogen column density (N$_{\rm H}$). }. The high 2--10 keV X-ray luminosity of IC 4329 A indicates the presence of a luminous AGN. This luminous AGN, together with the estimated small outer-to-inner-radius ratio for nuclear dust, in IC 4329 A may be related to its higher dust effective temperature derived from the 10$^{4}$--10$^{5}$ GHz (3--30 $\mu$m) data. In summary, the considerably smaller-than-expected HCN and HCO$^{+}$ J=3--2 emission line flux peak in IC 4329 A could be due to some combination of (1) a low column density of obscuring gas/dust around an AGN and/or (2) a small dust mass to infrared $L$-band (3--4 $\mu$m) luminosity ratio, due to a high dust effective temperature. In either scenario, our (sub)millimeter energy diagnostic method is not sensitive to almost bare AGNs with a very limited amount of surrounding molecular gas, such as IC 4329A, because at least a detectable amount of molecular line emission is required for our method to be effective. | 16 | 9 | 1609.01291 |
1609 | 1609.08653_arXiv.txt | The \Gaia~astrometric mission may offer an unprecedented opportunity to discover new tidal streams in the Galactic halo. To test this, we apply \ngc, a great-circle-cell count method that combines position and proper motion data to identify streams, to ten mock \Gaia~catalogues of K giants and RR Lyrae stars constructed from cosmological simulations of Milky Way analogues. We analyse two sets of simulations, one using a combination of $N$-body and semi-analytical methods which has extremely high resolution, the other using hydro-dynamical methods, which captures the dynamics of baryons, including the formation of an {\it in situ} halo. These ten realisations of plausible Galactic merger histories allow us to assess the potential for the recovery of tidal streams in different Milky Way formation scenarios. We include the \Gaia~selection function and observational errors in these mock catalogues. We find that the \ngc~method has a well-defined detection boundary in the space of stream width and projected overdensity, that can be predicted based on direct observables alone. We predict that about \emph{4--13 dwarf galaxy streams can be detected in a typical Milky Way-mass halo with \Gaia+\ngc}, with an estimated efficiency of $>$80\% inside the detection boundary. The progenitors of these streams are in the mass range of the classical dwarf galaxies and may have been accreted as early as redshift $\sim3$. Finally, we analyse how different possible extensions of the \Gaia~mission will improve the detection of tidal streams. | The \Gaia~mission, whose first data release is now publicly available, is expected to revolutionise our knowledge of the formation of the Milky Way, by mapping, for the first time, close to a billion stars in the disc, bulge and halo with exquisite astrometric precision \citep{Perryman2001,deBruijne2012}. It is anticipated that this detailed information will enable a breakthrough in understanding the formation history of the Milky Way. The stellar halo, in particular, holds a wealth of information about the merger history of the Galaxy, being a repository of most of the tidal debris from the past merger events. The number of tidal streams surviving at the present day in the halo, their morphologies, their total luminosities and their chemical abundance patterns, all encode important information from which the series of accretion events can be reconstructed \citep{Helmi1999,Bullock2005,Johnston2008,Cooper2010,Helmi2011}. Tidal streams can also be used to infer the gravitational potential of the Milky Way \citep[e.g.][]{PriceWhelan2013,Sanderson2015,Sanderson2016}. Increasing the number of stream detections can improve this measurement \citep{Deg2014}. While it is expected that \Gaia~will uncover new tidal streams in the halo \citep{Helmi2000,Gomez2010}, quantitative theoretical predictions for the likely number of such discoveries have not been made to date, mainly because of the uncertainties in modelling the physical processes associated with baryons in the framework of hierarchical structure formation, the computational resolution and re-sampling issues associated with producing adequate simulated catalogues at the level of individual stars, and the need to develop algorithms for making such detections by mining the Gaia dataset. In this study, we aim to make progress by employing a series of state-of-the-art simulations of Milky Way-mass haloes from which we construct mock \Gaia~star catalogues, which we search for tidal streams with a robust, quantifiable method. We use two suites of cosmological simulations to produce the mock \Gaia~catalogues: the Aquarius simulations, a set of high resolution dark matter only simulations of Milky Way-mass haloes \citep{Springel2008a}, combined with the GALFORM semi-analytic prescriptions \citep{Cooper2010}; and a second set, called \ljmus, which comprises several medium resolution hydro-dynamical simulations of Milky Way-mass disc galaxies (Font et al., in prep., hereafter F17), the initial conditions of which were extracted from the EAGLE simulation \citep{Schaye2015}. Aquarius allows us to study tidal streams from progenitors that span a wide range of masses and orbits, and hence to test our method on a realistic set of stream luminosities and morphologies. On the other hand, the \ljmus, although of lower resolution than Aquarius, have the benefit of modelling the hydro-dynamical effects of baryons self-consistently. Baryonic effects, including modification of the density profiles of satellites by stellar feedback and interactions between satellites and the central stellar disc may alter the morphology of tidal streams, and, together with the possible presence of an in situ halo, this may change (most likely decrease) the number of streams that can be detected. The objective of this paper is not to perform a detailed comparison between these two simulation techniques, but rather to estimate the detectability of the tidal streams they predict. This work goes beyond earlier studies of tidal stream detection in several ways. For the first time, we make predictions based on fully cosmological simulations of Milky Way-mass galaxies that we combine with the most up-to-date \Gaia~error estimates and selection function. The simulated tidal streams evolve within a realistic gravitational potential (non-axisymmetric and changing in time). Thus, the mock \Gaia~star catalogues constructed here complement existing \Gaia~mocks which do not include substructure in the stellar halo \citep[e.g.][]{Robin2012}. Examining a number of Milky Way-mass haloes with a variety of merger histories helps to make our predictions robust against our ignorance of the details of the Galaxy's accretion history. This is a step forward towards comparing the models and observations on a level playing field. Also, with the \ljmu~simulations, the effect of halo component formed {\it in situ} \citep{Zolotov2010,Font2011a,McCarthy2012,Cooper2015,Pillepich2015} on the detectability of tidal streams can be taken into account. To our knowledge, the contaminating effect of combined {\it in situ} and accreted halo components has only been estimated for \Gaia~by \citet{Brown2005} and \citet{Mateu2011}, who embedded a set of stellar streams in a smooth Galactic background with a constrained luminosity normalization. However, these streams were evolved in a fixed axisymmetric potential and their progenitors selected \emph{ad hoc}. Rather than starting from the information available in the simulations, in which every star particle and hence every stream can be identified unambiguously with a specific progenitor, we first apply an observational stream finding algorithm based on the Great Circle Counts (GC3) method. This method, described in detail below, uses combined positions and proper motions to assign stars to discrete groups with common orbital poles. GC3 methods are an efficient way to search for tidal streams in the Galactic halo. They exploit the fact that streams will be approximately confined to planes in potentials that are close to spherical, by searching for overdensities of stars along great circles (as seen from the Galactic centre). The idea was initially proposed by \citet{LyndenBell1995} and \citet{Johnston1996} and later modified by \citet[][hereafter M11]{Mateu2011} to improve its efficacy by including kinematical information (mGC3), with the Gaia~mission in mind. Its main advantage is that, with the implementation proposed in M11, the GC3 family of methods works directly in observable space (positions, parallax, proper motion, radial velocity), rather than using physical parameters such as energy or angular momentum, greatly reducing the effect of the propagation of observational errors, which \citet{Brown2005} have shown can be quite substantial even for \Gaia. Finally, we assess the efficiency of our stream detection method by using our knowledge of the `true' population of streams in the simulations to determine which progenitors are recovered and with what `purity'. This knowledge of the method's efficiency and selection biases will be a key ingredient in the inverse process of inferring the Galactic accretion history. The paper is structured as follows: Section \ref{s:simulations} summarises the simulations employed in this study. Section \ref{s:mock_catalogues} describes the construction of the mock \Gaia~catalogues and the \Gaia~error simulation. Section \ref{s:gaia_can_see} presents what \Gaia~like surveys would `see' in the simulated stellar haloes based on a selection of specific stellar tracers. Section \ref{s:mgc3} describes the Great Circle method used to identify tidal streams. The appearance of observable tidal streams in the diagnostic space of the method, which we call pole count maps, is explored in detail for a fiducial halo in Section \ref{s:full_PCMs}. Section \ref{s:all_pcms} summarises the results of applying our algorithm to all the other haloes in our sample. In Section \ref{s:recovery_all}, we investigate the properties of progenitors of the streams that are detected in the mock \Gaia~surveys of our simulations. In Section \ref{s:gaia_extensions} we analyse how the detectability of tidal streams changes under various scenarios for extending the lifetime of the \Gaia~mission. Finally, in Section \ref{s:recommendations} we discuss several ways in which this stream finding method can be further improved and give a summary of our conclusion in Section \ref{s:concl}. \section[]{Cosmological Simulations}\label{s:simulations} \subsection{Aquarius Simulations}\label{s:aquarius} Aquarius is a set of six collisionless cosmological `zoom' simulations of individual dark matter haloes of mass $\sim10^{12}\Msun$ \citep{Springel2008a, Springel2008b, Navarro2010}. The simulations assume a ${\Lambda}\rm{CDM}$ cosmogony with parameters determined from the WMAP 1-year results \citep{Spergel2003} and the 2dF Galaxy Redshift Survey data \citep{Colless2001}: $\Omega_{\mathrm{M}}=0.25$, $\Omega_{\mathrm{\Lambda}}=0.75$, $n_{\mathrm{S}}=1$, $\sigma_{8} = 0.9$ and Hubble parameter $h = 0.73$. The six haloes were selected randomly from a parent sample of isolated halos of similar mass in a lower resolution $(100 \, h^{-1})^{3} \mathrm{Mpc^{3}}$ cosmological volume simulation \citep{Gao2008}. Isolation was defined by the absence of any neighbours with more than half the mass of the target halo within $1\,h^{-1}$ Mpc. A Lagrangian region several times larger than the $z=0$ virial radius of each target halo was resimulated with a much larger number of lower-mass particles, coarsely sampling the surrounding large-scale structure with a smaller number of higher mass particles, subject to exactly the same spectrum of initial density perturbations. The Aquarius simulations are labelled Aq-A to Aq-F; we do not use Aq-F in this paper because its recent merger history makes it highly unlikely to be representative of a system like the Milky Way \citep{BoylanKolchin2010, Cooper2010}. We use the level 2 set of simulations, the highest resolution level at which all six haloes were simulated. The particle mass varies slightly between the level 2 simulations in the range $0.6 < m_{\mathrm{p}}(\times10^{4} \,\Msun) < 1.4$. The Plummer-equivalent gravitational softening length is $\epsilon\sim66$~pc. The Aquarius simulations use a single high-resolution particle species to model the collisionless dynamics of both dark matter and baryons. To represent the stellar component, we use the `particle tagging' models described by \citep{Cooper2010}. This technique first uses a semi-analytic galaxy formation model to determine the star formation history of each dark matter halo in the simulation, and then applies dynamical criteria to select subsets of collisionless particles occupying regions in phase space associated with each distinct single-age stellar population at the time of its formation. The \citet{Cooper2010} technique improves on earlier tagging approaches \citep[e.g.][]{Bullock2005} in the use of a single, self-consistent cosmological simulation to treat the dynamics of the satellites and the host halo, and in the use of a galaxy formation model constrained by large cosmological datasets as well as the properties of Milky Way and M31 satellites \citep{Bower2006,Cooper2010,Font2011b}. The five Aquarius simulations we use show considerable diversity in the properties of their stellar haloes, owing to their range of virial masses and, more significantly, to the intrinsically stochastic nature of dwarf galaxy accretion and disruption in $\Lambda$CDM. The particle tagging technique involves a dynamical approximation with clear limitations, and unlike \citet{Bullock2005} the \citet{Cooper2010} simulations do not include the gravitational contribution of a massive stellar disc at the centre of the host potential. The presence or absence of a disk may accelerate the tidal disruption of some satellites. This is likely to affect predominantly those substructures with orbits passing through the inner $\sim$20 kpc of the galaxy after $z>2$, which nevertheless may include satellites and streams located far from the disk at $z=0$. \citet{Errani2017} find the total number of potentially luminous subhaloes disrupted in the inner region of the halo changes by a factor of $\sim$2 when an idealised disk component is added to the potential in one of the Aquarius simulations. They demonstrate that the inner slope of the satellite mass density profile (which depends on the physics of galaxy formation) has an even larger effect on the number of surviving satellites (almost an order of magnitude; the conclusions of Errani et al. relate only to whether or not an identifiable self-bound core survives, rather than to the presence of tidal streams). Likewise, \citep{GarrisonKimmel2017} find a factor 2--5 depletion of massive subhaloes in a dark matter only simulation when they introduce a growing analytic disk potential based on a hydrodynamical realization from the same initial conditions. Our results here concern the disruption of well-resolved satellites with very high mass-to-light ratios, predominantly in the outer halo; for further discussion of related issues we refer the reader to \citet{Cooper2010,Cooper2013,Cooper2016} and \citet{LeBret2015}. The number of these more distant satellite halos surviving at $z=0$ may therefore be considered uncertain by no more than factor of $\sim2$ as the result of neglecting the (still somewhat uncertain) influence of a disk potential. As we describe in the following subsection, we also analyse a suite of lower-resolution gas-dynamical simulations that account self-consistently for the gravitational effects of baryons neglected by the particle tagging approach. This allows us to check for large differences in the number of streams from bright satellites that could be due to the presence of a disk, albeit in the context of only one hydrodynamical model and in different dark matter haloes to Aquarius. If the MW disk has significantly depleted the number of luminous satellite subhaloes surviving to $z=0$, our predictions based on Aquarius are likely to provide a lower limit to the total number of streams that Gaia will discover. \subsection{Gas Dynamical Simulations}\label{s:gas_dynamical} For the gas-dynamical simulations, we use a suite of `zoom' simulations of Milky Way-mass haloes using the high-resolution `Recal' model from the recent EAGLE project \citep{Schaye2015,Crain2015}. The zoom simulations will be described in more detail in a future study (F17), so we provide only a brief description here. We recall that the main aim of the EAGLE project was to simulate, at relatively high resolution (baryon particle mass $\approx10^6 \ {\rm M}_\odot$, softening length of 500 pc), the evolution of the main galaxy population. The stellar and AGN feedback parameters were adjusted so as to reproduce the observed galaxy stellar mass function and the size$-$mass relation of local galaxies. Unfortunately, the resolution of the main EAGLE box (L100N1504) is too low for our purposes, motivating our use of significantly higher-resolution zoom simulations. Note that \citet{Schaye2015} have found that when the resolution is increased, some re-calibration of the stellar and AGN feedback is required to preserve a match to the galaxy stellar mass function. Using this re-calibrated model (called `Recal') they have simulated a 25~Mpc volume with a factor of 8 (2) better mass (spatial) resolution (i.e., L025N0752). This simulation volume served as the parent volume from which several haloes were selected for re-simulation. Specifically, F17 identified a volume-limited sample of 25 haloes which fall in the mass range $7\times10^{11} < M_{200}/{\rm M}_\odot < 3\times10^{12}$ at $z=0$ ($M_{200}$ denotes the mass within the virial radius $r_{200}$). Inspection of the visual morphologies indicates that not all of these systems have significant stellar disc components. While such systems are interesting in their own right (and the intention is to eventually simulate all 25 haloes), priority was given to 10 systems which have the most disc-like morphology. F17 have carried out zoom simulations with a factor of 8 (2) better mass (spatial) resolution than the parent volume (i.e., baryon particle mass of $\approx1.5\times10^{4} \ {\rm M}_\odot$, Plummer-equivalent softening length of 125 pc) using the Recal model\footnote{No additional re-calibration of the model was performed when increasing the resolution beyond that of the Recal-L025N0752 parent volume, but F17 have verified that the stellar masses of the zoomed haloes agree with those of the parent volume to typically better than 10\%.}. For further details of the Recal model, including a description of the employed hydrodynamic solver and subgrid prescriptions for radiative cooling, star formation, stellar and chemical evolution, and feedback we refer the reader to \citet{Schaye2015}. In the present study we analyse a random subset of 5 of the 10 zoom simulations carried out by F17. At $z=0$, this sub-set spans virial masses $7.14\times10^{11} < M_{200}/{\rm M}_\odot < 1.93\times10^{12}$ and stellar masses $7.33\times10^{9} < M_{*}(<30~\rm{kpc})/{\rm M}_\odot < 1.99\times10^{10}$, respectively, similar to the corresponding values of the five Aquarius haloes \citep{Cooper2010}. Apart from the fact that these galaxies resemble the Milky Way in terms of total and stellar mass, the properties of their bound substructure also match the main properties of Milky Way satellites, e.g. the luminosity function and the stellar mass - metallicity relation (see also \citet{Schaye2015} for the properties of low mass galaxies in the Recal model). A more detailed investigation of the properties of these galaxies will be presented in F17. We note, however, that, due to the limited numerical resolution, these gas-dynamical zoom simulations can follow reliably only the properties of satellites in the classical dwarf galaxies regime ($M_{*}\geq 10^7 M_{\odot}$). Following the methods described in \citet{Font2011a}, we construct simple merger histories for each of the simulated galaxies, identifying which star particles were formed `in situ' (i.e., within the main progenitor branch), which were brought in via mergers/tidal disruption of infalling satellites, and which star particles still reside in orbiting satellites at the present day. For the star particles that were/are in satellites, we record the properties of the halo to which the particles belonged just prior to joining the main Friends-of-Friends group. The \ljmu~simulations have the benefit of treating various baryonic physical processes, such as gas infall, star formation and stellar feedback, self-consistently. Anticipating the results, we expect that the gas-dynamical simulations will obtain a somewhat different number of tidal streams and different stream morphologies, than in the case of the particle tagging methodology. For example, the stellar feedback may change the internal spatial and kinematical distributions of stars in satellite galaxies and may transform cuspy density profiles into cored ones. This, in turn, can affect the rate at which material is tidally stripped from satellites which changes the time when tidal streams are formed and their morphological properties. The presence of a disc may influence the spatial distribution of satellites in the inner region of the galaxy, by inducing changes in the orientation of their angular momentum and by accelerating their tidal disruption. Additionally, the hydro-dynamical gas-dynamical simulations have been shown to produce stellar haloes with dual components: accreted and in situ \citep{Zolotov2010,Font2011a,McCarthy2012,Cooper2015,Pillepich2015}. We caution that the origin of the in situ component is still debated, current gas-dynamical simulations suggesting different scenarios: stars being ejected from the disc by disc-satellites interactions, or formed in the wake of the gas stripped from infalling satellites, or formed in cold gas filaments. Understanding the origin of in situ halo stars is crucial for predicting the physical properties of this halo component and, implicitly, for modelling the environment in which tidal streams evolve. Strictly from the point of view of the detectability of tidal streams, the in situ component of the stellar halo is another source of foreground/background contamination, whose effect needs to be assessed. Overall, the additional effects present in gas-dynamical simulations are expected to diminish the number of tidal streams that are dynamically cold at present day and therefore, those that are most likely to be detected. We note, however, that this discussion is mainly qualitative at this point, and a more rigorous assessment of the significance of the various baryonic processes will require an in-depth quantitative investigation. This is, however, beyond the scope of this present paper since the differences in the initial conditions and numerical resolution between the two types of simulations presented here do not allow for a fair comparison. In the case of these two types of simulations, we estimate that the main differences in the number of tidal streams are most likely do to the differences in the numerical resolution. We note, however, that in the range in which the \ljmu~ simulations are able to resolve the halo substructure $-$ roughly, the domain of the classical dwarf galaxies $-$, the two types of simulations predict similar number of surviving satellites and of tidal streams. | \label{s:concl} Tidal streams are widely recognised for their usefulness in the inference of the Galactic accretion history, one of the key science drivers for the Gaia mission \citep{deBruijne2012}. However, any such inference demands a thorough understanding of the selection biases that may affect tidal stream detection methods. Motivated by this, and the prospects that the Gaia mission opens up for all-sky homogeneous stream surveying, we have explored the detectability of tidal streams in Gaia mock catalogues using \ngc, a great-circle cell counts method that uses positional information and proper motions \citep{Abedi2014}. We have built mock catalogues for two standard candle tracers: K giants and RRLSs, reproducing the Gaia selection function and observational errors, and assuming photometric distance errors of 20 and 7\% respectively for each tracer. These mock catalogues were made from a set of 5 haloes from the Aquarius N-body simulations and 5 haloes from the \ljmu~gas dynamical simulations. The diversity of orbits and progenitors in these allows us to characterise the \ngc~method's completeness and detection limits in a realistic setting. We have also explored how the \insitu~stellar halo background in \ljmu~gas dynamical simulations affects the detection of streams, and the improvements in proper motion errors expected for three possible extensions of the \Gaia~mission. We summarise our results as follows: \begin{enumerate} \item The \ngc~method is able to identify realistic tidal streams produced in cosmological N-body and gas dynamical simulations, even when contamination from a smooth halo background is included. \item The method has a \emph{well defined parameter-free detection boundary} in the plane of angular width vs. ratio of observable to PCM background stars, defined in Equation~\ref{e:boundary}. \item Progenitors are recovered up to infall redshifts as large as $z_{\rm{infall}}\sim3$ based on results with the gas dynamical simulations, in which progenitors are more prone to disruption. \item A total of 9 to 12 progenitors, bound and unbound, are expected to be detectable with \Gaia+\ngc~using KIII stars as tracers; and 4 to 10 using RRLS. These correspond respectively to a median 86\% and 80\% of all progenitors inside the detection boundary, below our selected threshold of $\Delta\theta=15\degr$. \label{i:res_i} \item A total of 3 to 8 \emph{streams} would be recovered successfully with Gaia+nGC3 when observed with K giants and 3 to 10 with RRLS. Depending on the specific merger history of the Milky Way this means that \Gaia~ has the potential to almost double the number of known tidal streams in the halo. Also, approximately the same number of streams can be recovered with RRLS as with K giants, even though RRLS probe a substantially smaller volume. \item When results from RRLS and K giants are combined, 4 to 13 streams are recovered successfully ($N_{\rm RR+K}$), which implies a median gain of 2 extra streams when compared to results obtained with K giants alone. \item The stellar masses and luminosities of recovered progenitors go down to $\sim10^6\Msun$ and $\sim4\times10^5\Lsun$ respectively, i.e. similar to the classical dwarf spheroidal MW satellites.\label{i:res_f} \item Our forecasts in items \ref{i:res_i}-\ref{i:res_f} are based on results from the Aquarius simulations alone, since HYDRO-zoom results may be hampered due to their lower mass resolution. \item Progenitors are recovered down to the same stellar mass limit and the same infall redshift range with either tracer, RRLS or K giants. \item Recovered progenitors span the heliocentric distance range from $20$ to $130$ kpc, with the best completeness ($>$80\%) achieved in the range from $\sim$30 to $\sim$90 kpc. \item For streams (i.e. partially unbound progenitors), RRLS probe \emph{the same effective volume} ($\sim$20--90 kpc) and mass range ($\gtrsim10^6\Msun$), with a similar completeness, as KIII stars. For bound progenitors KIII stars probe a larger volume, reaching out to $\sim$130 kpc. \item We analysed the detectability of progenitors also for gas dynamical simulations which naturally include the \insitu~background. Although, as expected, the contamination from this additional background hinders the detections, we find that using a simple cut to exclude the disc stars ($|b|\leqslant10\degr$ \& $R\leqslant20$ kpc) one can recover as many progenitors (or more in one case) as in the case when the in situ component is not taken into account. \item We analysed how the detectability of progenitors would be improved by the smaller proper motion errors resulting from an extension of the \Gaia~mission lifetime. The three scenarios considered were a two-year extension, a five-year extension and a second \Gaia~mission launched in 20 yr. In these scenarios, proper motion errors would be reduced by factors of 0.6, 0.35 and 0.2 respectively. Increases of about one, two and three progenitors respectively are expected in each scenario with respect to the results found for the nominal mission lifetime. \end{enumerate} Finally, the K giant and RRLS \Gaia~mock catalogues produced for both the Aquarius and \ljmu~simulations are publicly available at \href{https://cmateu.github.io/Cecilia_Mateu_WebPage/Gaia_Halo_Mocks.html}{this URL}. These catalogues include, for each star, all the position and velocity information in heliocentric spherical and galactocentric cartesian coordinate systems, with and without simulated \Gaia~errors, including the pole ID indicating to which pole detection (if any) it is associated in the \ngc~PCM. Each pole detection catalogue thus represents a realistic set of stream detections in which streams may overlap, stars will be missing and there will be contamination from the smooth background and foreground. These catalogues will be a useful benchmark for further studies on the inference of the Galactic accretion history and gravitational potential. | 16 | 9 | 1609.08653 |
1609 | 1609.08465_arXiv.txt | We study, by means of numerical simulations and analysis, the details of the accretion process from a disc onto a binary system. We show that energy is dissipated at the edge of a circumbinary disc and this is associated with the tidal torque that maintains the cavity: angular momentum is transferred from the binary to the disc through the action of compressional shocks and viscous friction. These shocks can be viewed as being produced by fluid elements which drift into the cavity and, before being accreted, are accelerated onto trajectories that send them back to impact the disc. The rate of energy dissipation is approximately equal to the product of potential energy per unit mass at the disc's inner edge and the accretion rate, estimated from the disc parameters just beyond the cavity edge, that would occur without the binary. For very thin discs, the actual accretion rate onto the binary may be significantly less. We calculate the energy emitted by a circumbinary disc taking into account energy dissipation at the inner edge and also irradiation arising there from reprocessing of light from the stars. We find that, for tight PMS binaries, the SED is dominated by emission from the inner edge at wavelengths between 1--4 and 10~$\mu$m. This may apply to systems like CoRoT~223992193 and V1481~Ori. | \label{sec:intro} About 50\% of pre--main sequence (PMS) stars in T~associations have been found to be in binary systems (Duch\^ene \& Kraus 2013). Therefore, a large number of PMS stars evolve through accretion from circumbinary discs, and those discs are also the birthplace of planets. Discs around PMS binary systems were first detected at the same time as discs around single stars in the late 1980s and 1990s through the excess emission in the infrared and at submillimetric wavelengths (Bertout, Basri \& Bouvier~1988). The first circumbinary disc to be imaged, using interferometry and then adaptive optics, was that around GG~Tau (Dutrey, Guilloteau \& Simon~1994, Roddier et al.~1996). In 1997, Jensen \& Mathieu reported a deficit of near--infrared emission from a few short--period PMS binaries, confirming the theoretical prediction that binaries clear up a cavity in their circumbinary disc (Lin \& Papaloizou 1979). Recently, several circumbinary discs have been imaged with ALMA (Czekala et al. 2015, 2016), and it is believed that ALMA may be able to detect structures in circumbinary discs produced by the tidal potential of the binary (Ruge et a. 2015). Numerical simulations have shown that the cavity cleared up by the binary does not prevent accretion onto the stars (Artymowicz \& Lubow~1996), as gas spirals in along streams. Those streams connect to circumstellar discs that ultimately are accreted onto the stars. Such circumstellar discs in binary systems have been imaged using adaptive optics (Mayama et al. 2010) and gas in the cavity of the disc around GG~Tau has been imaged with ALMA (Dutrey et al. 2014). Because of the presence of streams and circumstellar discs in the cavity being fed by the circumbinary disc, PMS binaries are complex systems. Emission from such systems is indeed observed to have contribution from various sources and to often be variable, but it is not always understood what produces the emission and the variability (Messina et al. 2016, Ardila et al. 2015, Gillen et al. 2014, Bary \& Petersen~2014, Boden et al. 2009, Jensen et al. 2007). Here we investigate a new source of (variable) emission, namely energy emission from the inner edge of the cavity. Although the clearing of a cavity by the binary has been extensively studied, the physical conditions in the region of the cavity edge have yet to be elucidated. Here we show that the action of the gravitational torque from the binary, which truncates the disc and produces the cavity, is accompanied by energy dissipation at the disc's inner edge. This energy dissipation, for the most part, comes from shocks produced by fluid elements which are ultimately accreted by the binary, and that suffer some oscillations in their distance to the centre of mass of the binary resulting in them impacting the disc's inner edge on their way in. In addition, the edge of the cavity may behave as a hard wall which absorbs the radiation from the stars and re--emits it at longer wavelength. These two sources of energy produce an excess of emission which is in the mid--infrared for spectroscopic binaries. The plan of the paper is as follows. In section~\ref{sec:hydro_sim}, we present hydrodynamic simulations of a disc orbiting around a binary system. We calculate the rate of energy dissipation associated with shocks which are found to be highly localised at the inner edge of the disc. As already noted by previous authors, we find that the rate of accretion onto the binary {tends to be suppressed, compared to what would be expected from the value of the surface density just beyond the cavity edge in the absence of the binary, } when the disc's aspect ratio decreases below 0.05 or so. However, the energy dissipation at the disc's inner edge is not affected by this suppression. In section~\ref{sec:analysis}, we develop an analysis to elucidate the origin of this energy dissipation. We show that it is associated with the tidal torque which maintains the cavity: angular momentum is transferred from the binary to the disc through the action of compressional shocks and viscous friction. The rate of energy dissipation is given by the binary angular velocity times the torque exerted by the binary, and does not directly depend on the actual accretion rate onto the binary. In section~\ref{sec:particle_sim}, we present particle simulations to illustrate the dynamics near the disc's inner edge in more detail and the accretion flow onto the binary. These simulations show that, before being accreted, particles are in general accelerated onto a trajectory that sends them back towards the circumbinary disc. Shocks that result at the disc's inner edge together with the action of viscous friction circularize the orbits of the particles and dissipate energy. We find that the energy released during one collision of a unit mass particle with the disc's inner edge is on the order of the potential energy per unit mass there, leading to agreement with the results of the hydrodynamic simulations. When accretion onto the binary is reduced, the particles which end up being accreted undergo more collisions, such that the resulting total energy dissipation rate stays the same. In section~\ref{sec:SED}, we calculate the spectral energy distribution (SED) of a circumbinary disc taking into account the energy dissipated at the inner edge and also irradiation of the inner edge by the stars. We show that, for tight binaries having a separation $\sim 10$~R$_{\sun}$, emission from the edge completely dominates the SED in the mid--infrared (from 1--4 to 10~$\mu$m). In section~\ref{sec:discussion}, we summarize and discuss our results. We conclude that the processes presented in this paper may help to explain the excess of emission in the mid--infrared observed in some PMS binary systems, like CoRoT~223992193 (Gillen et al. 2014) and V1481~Ori (Messina et al. 2016). | \label{sec:discussion} In this paper, we have shown that energy has to be dissipated in the circumbinary disc for the tidal torque to maintain the cavity. Energy dissipation happens at the inner edge of the disc through compressional shocks produced by fluid that is propelled back to the disc after drifting inwards into the cavity. The rate of energy dissipation does not depend on the actual accretion rate onto the binary, only on what we call the {\em fiducial} accretion rate through the disc, $\dot{M}_f$, which is determined by the state variables in the disc just beyond the cavity edge. This would be the accretion rate onto the central object if it was a single star rather than a binary and the material near the cavity edge was part of a putative exterior steady state disc. Energy dissipation occurs at the fiducial rate even when the accretion rate onto the binary is reduced below $\dot{M}_f$ (which happens when the aspect ratio and viscosity of the disc is reduced). The rate of energy dissipation is typically $G (M_1+M_2) \dot{M}_f/r_{\rm in}$, where $r_{\rm in}$ is the inner radius of the circumbinary disc. Hydrodynamic simulations (see also the MHD simulations performed by Shi \& Krolik 2015) show that the edge of the circumbinary disc is very sharp. Therefore emission of the energy released by the shocks is localised at the edge. In addition, this edge is irradiated by the central stars and this also contributes to the SED of the system. In a tight PMS binary with a separation of $\sim 10$~R$_{\sun}$, emission from the disc's inner edge completely dominates the SED in the mid--infrared, i.e. for $\lambda \sim 1$--4 to 10~$\mu$m. For large values of the accretion rate $\dot{M}_f$ exceeding about $10^{-8}$~$M_{\sun}$~yr$^{-1}$, the emission predominantly comes from the shocks derived from fluid elements being propelled back towards the disc. For lower accretion rates, irradiation from the central star becomes the main source of emission from the inner edge. When shocks dominate, we expect the SED to display some variability, as the shock dissipation rate varies with time (see fig.~[\ref{shockD}]) on a timescale of a few $\omega^{-1}$, where $\omega$ is the angular velocity of the binary, i.e. comparable to the binary period. Even when irradiation dominates, we would expect some variability on a timescale comparable to the binary period as the distance between the edge of the disc and the stars varies as the stars move around their centre of mass. \subsubsection* {Application to CoRoT~223992193:} CoRoT~223992193 is a double-lined PMS eclipsing binary which was discovered by Gillen et al. (2014). Its age has been estimated to be between 3.5 and 6 Myr. The SED of this system presents a mid--infrared excess between 4 and 10~$\mu$m. Gillen et al. (2014) found that this excess could not be produced by a circumbinary disc unless it had an accretion rate of $10^{-7}$~$M_{\sun}$~yr$^{-1}$, four orders of magnitude higher than the accretion rate derived from the H$\alpha$ emission from the stars. It was then proposed that the excess may be due to thermal emission from dust present in the vicinity of the two stars. As little as $10^{-13}$~M$_{\sun}$ of dust would be needed to produce the observed flux. This dust may also be responsible for the out--of eclipse variability observed in this system (Terquem, S\o rensen--Clark \& Bouvier 2015). \noindent Here, we comment that the excess emission observed for CoRoT~223992193 is very similar to what would be expected from the inner edge of a circumbinary disc. In this system, emission would not be dominated by the shocks as the accretion rate onto the stars is only $10^{-11}$~$M_{\sun}$~yr$^{-1}$ and the fiducial accretion rate through the disc is probably not that much larger (as $H/r$ is not expected to be smaller than 0.05 or so in a disc around a PMS binary). Emission would then be produced by irradiation from the stars. With $H_{\rm in}=0.05$, the disc's inner edge would produce a flux with a maximum of about $3.5 \times 10^{-13}$~erg~cm$^{-2}$~s$^{-1}$ at a wavelength between 3 and 4~$\mu$m, and a flux about 3.5 times smaller at 10~$\mu$m (see the lower plots on fig.~[\ref{fig3}]). The maximum value of the flux would be increased up to $10^{-12}$~erg~cm$^{-2}$~s$^{-1}$ if we adopted $H_{\rm in}=0.15$ instead. Such an enhanced value of the disc's aspect ratio near the edge of the cavity is consistent with 3D MHD simulations (Shi \& Krolik~2015) and may be produced by magnetic fields that get relatively stronger in the cavity. The peak of the excess measured for CoRoT~223992193 is about $3 \times 10^{-12}$~erg~cm$^{-2}$~s$^{-1}$ at 4~$\mu$m and decreases by the same factor as in our model at 10~$\mu$m (see fig.~[13] of Gillen et al.~2014). Given that we have calculated irradiation of the disc's inner edge in a very crude way, our results are close enough to the observations that emission from the disc's inner edge appears as a serious candidate for producing the observed excess. We note that, as the system is seen with an inclination $i=85^{\circ}$ (Gillen et al. 2014) which, although large, is such that $90^{\circ}-i$ in radians exceeds $ H/r$, accordingly the inner rim of the disc would not be obscured by the disc itself. \subsubsection*{Application toV1481~Ori:} V1481~Ori is a spectroscopic binary with a period of about 4.4 days in the Orion nebula cluster. It is associated with a strong infrared excess which suggests the presence of an accreting circumbinary disc (Messina et al. 2016 and references therein). Recent spectroscopic and photometric observations performed by Messina et al. (2016) in the $V$--band at $\sim 0.7$~$\mu$m and in the $I$--band have shown that the luminosity ratio of the stellar components varies in a way that can be explained by the presence of a hotspot on the secondary, which is speculated to be produced by accretion from a circumstellar disc. These observations also show that the minimum value of the luminosity ratio is 30\% larger than what would be expected in the case of no accretion. Messina et al. (2016) conclude that there has to be an additional source of emission which is visible at all phases. \noindent Here, we suggest that this excess of energy may be produced by the inner edge of the circumbinary disc. Given the short wavelength at which this excess occurs, it would have to be produced by shocks localised over a small fraction of the surface at the inner edge, as illustrated on the upper right plot of figure~(\ref{fig3}). The peak of the excess on this plot is at about 1~$\mu$m, so that there is significant emission in the bands in which Messina et al. (2016) have made the observations. This model requires a rather large accretion rate in the circumbinary disc which, as pointed out above, is suggested by the observations. | 16 | 9 | 1609.08465 |
1609 | 1609.08605_arXiv.txt | The critical Lyman--Werner flux required for direct collapse blackholes (DCBH) formation, or $\Jc$, depends on the shape of the irradiating spectral energy distribution (SED). The SEDs employed thus far have been representative of {{realistic}} single stellar populations. We study the effect of binary stellar populations on the formation of DCBH, as a result of their contribution to the Lyman--Werner radiation field. Although binary populations with ages $>$ 10 Myr yield a larger LW photon output, we find that the corresponding values of $\Jc$ can be up to 100 times higher than single stellar populations. We attribute this to the shape of the binary SEDs as they produce a sub--critical rate of H$^-$ photodetaching 0.76 eV photons as compared to single stellar populations, reaffirming the role that H$^-$ plays in DCBH formation. This further corroborates the idea that DCBH formation is better understood in terms of a critical region in the H$_2$--H$^-$ photo--destruction rate parameter space, rather than a single value of LW flux. | Direct collapse black holes (DCBH) have gathered much attention recently \citep{Dijkstra2014a,Ferrara14a,Agarwal12,Agarwal14,Habouzit16a} as a plausible solution to the problem of forming billion solar mass black holes very early in cosmic history as is required to explain the existence of very luminous quasars at redshifts $z>6$. Pristine gas in an atomic cooling halo exposed to a critical level of Lyman--Werner (LW) radiation can rid itself of molecular hydrogen (cooling threshold $\sim 200$ K), thereby collapsing isothermally in the presence of atomic hydrogen (cooling threshold $\sim 8000$ K). This leads to a Jeans mass threshold of $10^6 \msun$ at $n\sim 10^3 \rm \ cm^{-3}$, thereby allowing the entire gas mass in the halo \footnote{An atomic cooling halo, i.e. $\rm T_{vir}=10^4 \ K$ corresponds to a $\rm M_{DM} \approx 10^7 \msun$ at $z\approx10$. If we assume that the baryon fraction in this halo is the same as the cosmological mean value, i.e. $f_b \approx 0.16$, then the baryonic mass of such a halo will be at least $10^6 \msun$} to undergo runaway collapse eventually forming a $10^{4-5} \msun$ black hole in one go \citep{Omukai:2001p128}. The collapse must withstand fragmentation into Population III (Pop III) stars, which requires the gas to get rid of its angular momentum via bars--within--bars instabilities \citep{Begelman:2006p3700}, low--spin disks \citep[e.g.][]{Bromm:2003p22,Koushiappas:2004p871,Regan08,Lodato:2006p375} or high inflow rates in turbulent medium \citep{Volonteri:2005p793,Latif:2013p3629,Schleicher:2013p3661,Borm13}. In order for this mechanism to work, initially there must be a LW radiation field strong enough to delay Pop III star formation in a minihalo, $\rm 2000<T_{vir}\le10^4\ K$, till it reaches the atomic cooling limit of $\rm T_{vir} \ge 10^4\ K$ \citep{Machacek:2001p150,OShea:2008p41} . At this point, the flux of LW radiation illuminating the halo from nearby external stellar source(s) must be higher than a critical value $\Jc$ (conventionally written in units of $10^{-21} \rm\ erg/s/cm^2/sr/Hz$) to facilitate isothermal collapse of the pristine gas at 8000 K into a DCBH {\cite[e.g. recent simulations by ][]{2014ApJ...795..137R,2014MNRAS.445L.109I,2015MNRAS.446.2380B}}. Many previous studies of DCBH formation have adopted highly simplified prescriptions for the spectrum of this external radiation field, approximating the spectrum of a source dominated by Pop III stars as a $\rm T = 10^5$~K black body, and of a source dominated by Population II (Pop II) stars as a $\rm T=10^4$ K black body \citep{Omukai:2001p128,Shang:2010p33,WolcottGreen:2012p3854}. However, recent studies have emphasised the need for using more realistic spectral energy distributions (SED) for these sources as the value of $\Jc$ depends on the shape of the irradiating source's SED \citep[, A16 hereafter]{Sugimura:2014p3946, Agarwal15a, Agarwal15b}. These studies employed single stellar populations to represent the SEDs of Pop II stars, generating them using publicly available single stellar synthesis codes such as {\sc{Starburst99}} \citep{Leitherer:1999p112}, {\sc{Yggdrasil}} \citep{Zackrisson11} and \citet{Bruzual:2003p3256} model. However, in reality it is likely that a significant number of the stars will be part of binary systems. Stellar populations with significant binary fractions have higher hydrogen ionising photon yields than single stellar populations \citep[e.g.][]{2016MNRAS.456..485S,2016MNRAS.459.3614M}, and so it is plausible that accounting for their existence will lead to significant differences in the value of J$_{\rm crit}$ that we derive. | We study the LW flux requirement for DCBH formation from galaxies that have a stellar population that includes a significant binary fraction. We show that despite their high LW output, binary populations are in fact inefficient at causing DCBH in their vicinity when compared to single stellar populations, contrary to what one would naively expect. This can be attributed to the SEDs of binary populations that are systematically bluer than those of populations composed only of single stars, meaning that the light from them is much less effective at causing H$^{-}$ photodetachment. The lower H$^{-}$ photodetachment rates mean that higher H$_{2}$ photodissociation rates are needed in order to bring about DCBH formation, and so the required values of $\Jc$ are larger. Consistent with A16, we a find a distribution in the values of the $\Jc$ produced by binary populations, albeit narrower ($\Jc \sim 300-3000$) than the one produced by single stellar populations ($\Jc \sim 0.1- 3000$). Furthermore the need for older single stellar populations becomes clear as they produce the lowest values of $\Jc$ in both cases, due to a higher k$_{\rm de}$. This pushes the idea further that the formation of DCBHs must be understood in terms of the k$_{\rm de}$--k$_{\rm di}$ parameter space (Eq.~\ref{eq.ratecurve}), and not in terms of a single flux value. | 16 | 9 | 1609.08605 |
1609 | 1609.06332_arXiv.txt | {Since the mid-1980s the shock-in-jet model has been the preferred paradigm to explain radio-band flaring in blazar jets. We describe our radiative transfer model incorporating relativistically-propagating shocks, and illustrate how the 4.8, 8, and 14.5 GHz linear polarization and total flux density data from the University of Michigan monitoring program, in combination with the model, constrain jet flow conditions and shock attributes. Results from strong {\it Fermi}-era flares in 4 blazars with widely-ranging properties are presented. Additionally, to investigate jet evolution on decadal times scales we analyze 3 outbursts in OT~081 spanning nearly 3 decades and find intrinsic changes attributable to flow changes at a common spatial location, or, alternatively, to a change in the jet segment viewed. The model’s success in reproducing these data supports a scenario in which relativistic shocks compress a plasma with an embedded passive, initially-turbulent magnetic field, with additional ordered magnetic field components, one of which may be helical.} \keyword{blazars; shocks; linear polarization; centimeter-band} \conferencetitle{Blazars through Sharp Multi-Frequency Eyes} \begin{document} | The shock-in-jet model has been the preferred paradigm for explaining flaring in blazar jets since the 1980s \cite{MG85, HAA85}. In the radio-band the evidence supporting this scenario is the spectral evolution of the linear polarization during total flux density outbursts, consistent with the expected changes as a propagating shock compresses an initially-turbulent magnetic field; this compression increases the particle density and the magnetic field energy density, and hence the emissivity, and partially orders the magnetic field. The signatures of the passage of a shock in the light curves are a flare in total flux density (S) with a temporally-associated increase in the fractional linear polarization (P\%) and a systematic swing in the electric vector position angle (denoted by EVPA or $\chi$). An example long-term light curve which illustrates the shock signatures — temporally well-separated outbursts in S (an envelope over blended, individual flares); outbursts in polarized flux (ranging from a few percent to $\sim$15\%) and ordered changes in the EVPA on timescales of a few years — is shown in Figure 1. Prompted by renewed interest in models to explain blazar variability with the discovery that $\gamma$-ray flaring from blazars is common \cite{ACK15} we recently revisited the shock-in-jet model. In the new work we allow the shocks to be orientated at any direction to the flow \cite{HAA11} while the original modeling \cite{HAA85} assumed that all shocks were oriented transversely to the flow direction, and we consider the sensitivity of the model fitting to the inclusion of an ordered component of the magnetic field \cite{HAA15}. By comparing time segments of the centimeter-band total flux density and linear polarization data at three centimeter-band frequencies from the University of Michigan blazar variability program (UMRAO) for four $\gamma$-ray-flaring blazars with radiative transfer simulations incorporating propagating shocks, we have identified the jet flow properties and shock attributes \cite{AHA14, AHAJ14}. We summarize those results here, and we use this method to identify long-term changes spanning nearly 3 decades in the flow of a newly-modeled source exhibiting more modest $\gamma$-ray flaring, OT~081, during the first 4 years of {\it Fermi} (see http://www.bu.edu/blazars/VLBA\_GLAST/1749.html). \begin{figure}[H] \centering \includegraphics[width=3.8in]{Fig1.jpg} \caption{ From bottom to top: monthly-averaged total flux density, linear polarization, and electric vector position angle observations for OT~081 (1749+096) from the UMRAO variability program. The three centimeter-band frequencies are symbol and color coded as denoted at the top left.} \end{figure} | We have shown that the spectral evolution of outbursts in total flux density and linear polarization at centimeter-band has successfully been simulated in blazars of both the BL Lac and QSO classes. Our analysis assumes a shock-in-jet model in which the radiation is produced within a particle dominated flow with an embedded, passive magnetic field which is primarily turbulent in character but in which ordered components also contribute to the energy density. A key point in the method is that the spectral variability in the linear polarization is the crucial constraint in the modeling, and that it is a particularly stringent constraint when the jet is oriented along or near to the line of sight to the observer. In contrast, the total flux density is nearly insensitive to changes in the key parameters. Using the modeling as a tool, we have identified both the jet flow and shock properties during selected outbursts in $\gamma$-ray-bright blazars. While simple analytic approaches can restrict the range of permitted values, detailed modeling such as we present here, is required to limit these ranges especially when extreme flow conditions may exist, such as in 0716+714 and in OT 081. An additional advantage of the radiative transfer method is that it allows us to separate out a variety of processes. These include relativistic effects such as Doppler boosts, and the signature on the emission of independently permitting a variety of magnetic field geometries: random, axial, and helical. Further, the long-term continuity of the UMRAO data, up to 4 decades for many sources, provides a method whereby intrinsic long term changes in blazar jets can be identified. This goal cannot be achieved using inhomogeneous data sets analyzed employing heterogeneous procedures. A knowledge of these parameters and their time-dependent behavior impacts our understanding of blazar emission including the processes responsible for $\gamma$-ray flaring. OT~081 has recently increased in photon flux by a factor of 20 in the GeV band over the average value during the period of our modeling \cite{GON16}, and this blazar has recently now been detected at TeV energies \cite{MAG16}. Modeling such as we demonstrate here can elucidate intrinsic changes in the flow associated with the onset of these high states in the High Energy and Very High Energy bands. We are continuing to refine the shock-in-jet model described here to improve the agreement between the archival UMRAO data and the simulations. Including a modest contribution from an ordered helical magnetic field in OT~081 has already demonstrated that future inclusion may improve the fits in other outbursts modeled, and the importance of this additional magnetic field component, which the simulations suggest contributes only modestly to the overall energy density of the system, needs to be explored more fully. Additionally we would like to allow for a range of shock obliquities since our current models do not reproduce the details of the observed EVPA spectral evolution. Our modeling assumes that the shocks are non-interacting and moving at constant velocity in rectilinear motion, while there is evidence from VLBI measurements that all of these assumptions are violated and that must also affect to some degree the interpretation of the variability in the light curves. Nevertheless, the modeling has already proven to be an effective tool in investigating the intrinsic properties of blazar jets and in identifying extreme conditions. \vspace{6pt} | 16 | 9 | 1609.06332 |
1609 | 1609.00178_arXiv.txt | The planetary nebula (PN) stage is the ultimate fate of stars with mass 1 to 8 solar masses (M$_\odot$). The origin of their complex morphologies is poorly understood\citeart{balick02}, although several mechanisms involving binary interaction have been proposed\citeart{demarco09}$^,$\citeart{garciaarredondo04}. In close binary systems, the orbital separation is short enough for the primary star to overfill its Roche lobe as it expands during the Asymptotic Giant Branch (AGB) phase. The excess material ends up forming a common-envelope (CE) surrounding both stars. Drag forces would then result in the envelope being ejected into a bipolar PN whose equator is coincident with the orbital plane of the system. Systems in which both stars have ejected their envelopes and evolve towards the white dwarf (WD) stage are called double-degenerates. Here we report that Henize~2--428 has a double-degenerate core with a combined mass unambiguously above the Chandrasekhar limit of 1.4 M$_\odot$. According to its short orbital period (4.2 hours) and total mass (1.76 M$_\odot$), the system should merge in 700 million years, triggering a Type Ia supernova (SN Ia) event. This finding supports the double-degenerate, super-Chandrasekhar evolutionary channel for the formation of SNe Ia\citeart{howell06}. | 16 | 9 | 1609.00178 |
||
1609 | 1609.09090_arXiv.txt | We present results on the clustering properties of galaxies as a function of both stellar mass and specific star formation rate (sSFR) using data from the PRIMUS and DEEP2 galaxy redshift surveys spanning $0.2 < z < 1.2$. We use spectroscopic redshifts of over 100,000 galaxies covering an area of 7.2 deg$^2$ over five separate fields on the sky, from which we calculate cosmic variance errors. We find that the galaxy clustering amplitude is as strong of a function of sSFR as of stellar mass, and that at a given sSFR, it does not significantly depend on stellar mass within the range probed here. We further find that within the star-forming population and at a given stellar mass, galaxies above the main sequence of star formation with higher sSFR are less clustered than galaxies below the main sequence with lower sSFR. We also find that within the quiescent population, galaxies with higher sSFR are less clustered than galaxies with lower sSFR, at a given stellar mass. We show that the galaxy clustering amplitude smoothly increases with both increasing stellar mass and decreasing sSFR, implying that galaxies likely evolve {\it across} the main sequence, not only along it, before galaxies eventually become quiescent. These results imply that the stellar mass to halo mass relation, which connects galaxies to dark matter halos, likely depends on sSFR. | \label{sec:intro} Galaxies are thought to form in the centers of dark matter halos, regions of the Universe that have collapsed under their own gravity. The observed clustering of galaxies matches well the predicted clustering of dark matter halos from $\Lambda$CDM cosmological numerical simulations, using various prescriptions for assigning galaxies to halos. However, it is not yet clear exactly how to map observed galaxies to dark matter halos, as it is not yet known exactly how galaxies form and evolve within these halos across cosmic time and how the dark matter halo influences the galaxy and vice versa. Earlier galaxy clustering papers often quantified in particular the luminosity-dependence of clustering, generally finding that the brightest galaxies are more clustered than fainter galaxies, with a sharp rise in the clustering amplitude above $L^*$ \citep[e.g.,][]{Alimi88, Benoist96, Norberg01}. \nocite{Alimi88} Similar results were found to hold at higher redshift as well, to $z\sim1$, when the Universe was less than half its current age \citep[e.g.,][]{Coil06, Pollo06, Meneux09}. As the observed bimodality in the optical colors of galaxies became increrasingly apparent \citep[e.g.,][]{Strateva01,Baldry04}, many authors turned towards measuring the luminosity-dependence of blue, star-forming and red, quiescent galaxies separately \citep[e.g.,][]{Norberg02, Hogg03, Coil04, Zehavi05, Meneux06}. These papers showed that at a given luminosity, red galaxies are more clustered than blue, and that {\it within} each of these two broad galaxy populations, the brightest galaxies are typically more clustered than fainter galaxies. Here again these results were found to hold out to $z\sim1$. However, it was also discovered at low redshift that within the red, quiescent galaxy population, low luminosity galaxies are highly clustered, likely reflecting that they tend to be satellite galaxies in massive dark matter halos hosting galaxy clusters \citep{Berlind05}. Some authors choose to split the galaxy population by morphology or spectral type instead of color, finding similar results, that galaxies with early-type, elliptical morphologies or early-type spectra are more clustered than late-type, spiral galaxies \citep[e.g.,][]{Loveday95, Madgwick03, Li06, delaTorre11}. Moving beyond considering the galaxy population as having only two general types, \citet{Coil08} used the DEEP2 galaxy redshift survey to split the $z\sim1$ galaxy population into finer bins in color, showing that the clustering amplitude rises {\it within} the blue, star-forming population alone, as the color becomes increasingly red. They did not find any clustering difference within the red, quiescent population when split by optical color. \citet{Zehavi11} found using SDSS at $z\sim0$ that clustering depends on color both within the blue, star-forming population and the red, quiescent population. Using the PRIMUS galaxy redshift survey at $z\sim0.7$, \citet{Skibba14} found again that clustering depends on color within the red, quiescent population (though not within the blue, star-forming population). These results began to more fully flesh out how galaxy clustering depends on the star formation properties of galaxies, beyond a simple division into star-forming or quiescent, and pointed to how galaxies must evolve with time in terms of their color (from very blue to very red). More recently, observers and theorists have moved from mapping the galaxy population in color-magnitude space to star formation rate (SFR) or specific SFR (sSFR, defined as the SFR per unit stellar mass) versus stellar mass space \citep[e.g.,][ and references therein]{Noeske07, Speagle14}. The latter quantities are more useful parameters as they are tied to physical processes occuring within galaxies (converting gas into stars, the growth of a galaxy) and are less impacted by dust obscuration. They are also easier quantities for theorists to model in cosmological simulations than color and magnitude. As a result, more recently there has been a lot of work quantifying the stellar mass-dependence of galaxy clustering \citep[e.g.,][]{Li06,Meneux08,Wake11,Leauthaud12, Marulli13}. These papers typically find that the clustering amplitude is a strong positive function of stellar mass above $M^*$ and is less dependent at lower stellar masses. This has led to many papers quantifying the stellar mass to halo mass relation and its evolution with cosmic time \citep[e.g.,][]{Behroozi10, Moster10, Leauthaud11, Durkalec15, Skibba15}. While there has been substantial work on the stellar mass-dependence of galaxy clustering, there have been few papers on the SFR or sSFR-dependence, either at low or high redshift. In a pair of related papers, \citet{Hearin14} and \citet{Watson15} show that the clustering properties of SDSS galaxies divided into star-forming or quiescent at a given stellar mass are very similar whether the galaxy subsamples are defined using either optical colors or sSFR. Essentially, as long as the observed bimodality in the galaxy population is used, whether the color or sSFR is used to define the bimodality does not matter in terms of the relative clustering of blue, star-forming galaxies to red, quiescent galaxies, perhaps not surprisingly. \citet{Li08} use SDSS to compare low and high sSFR samples within the star-forming population, and find that on very small scales (less than 100 kpc) the clustering amplitude is higher for galaxies with higher sSFR. This is likely due to galaxy-galaxy tidal interactions. \citet{Heinis09} use GALEX imaging of SDSS to investigate both the $NUV-r$ and \ssfr dependence of clustering, finding that the clustering amplitude increases with decreasing \ssfr or redder color, where they split the star-forming population into two bins and compare with the quiescent population. Other papers that have divided the fully galaxy population more finely into multiple bins in either SFR or sSFR have typically used only angular clustering measurements, where spectroscopic redshifts are lacking for individual galaxies \citep{Sobral10, Lin12, Dolley14, Kim15}. These papers, which span $z\sim0.2-2.0$, generally find that galaxy subsamples with higher SFR or lower sSFR have higher clustering amplitudes. \citet{Sobral10} measure the angular clustering of H$\alpha$ emitters at $z\sim0.8$ and find that clustering amplitude increases steadily with H$\alpha$ luminosity (which is a proxy for SFR), even at a fixed K-band luminosity (which is a proxy for stellar mass). \citet{Dolley14} measure the angular clustering of star-forming galaxies over a wide area of 8 deg$^2$, selecting galaxy subsamples based on {\it IRAC/MIPS} 24$\micron$ flux. They find that galaxies with higher 24$\micron$ flux (which is a proxy for SFR) have higher clustering amplitudes, though they do not investigate whether this difference may be accounted for by differences in the mean stellar mass of the samples. \citet{Kim15} measure the angular clustering of galaxies at $z\sim1$ in the UKIDDS DXS survey as a function of stellar mass and sSFR. They find a steady increase in the clustering amplitude with decreasing sSFR, above a given stellar mass threshold. \citet{Mostek13} use the DEEP2 galaxy redshift survey at $z\sim1$ to measure the stellar mass, SFR, and sSFR dependence of galaxy clustering, using multiple bins in each physical parameter. They find that within the star-forming population, clustering amplitude increases with increasing SFR and decreasing sSFR, though they find no SFR-depdence for quiescent galaxies. They investigate whether the SFR-depdence that is observed could be due to stellar mass and conclude that much, though not all, of the trend could be due to the known correlation between SFR and stellar mass (the star-forming ``main sequence''). They also investigate small-scale clustering properties and find a clustering excess for higher sSFR samples both within the star-forming and quiescent populations, which they attribute to galaxy-galaxy interactions. \citet{Mostek13} also find that star-forming galaxies above the ``main sequence'' of star formation are less clustered than those below, within a given stellar mass range, which points to the possibility of using clustering measurements to track the evolution of galaxies in the SFR-stellar mass plane. However, the DEEP2 sample is not large enough to further divide the galaxy population into multiple bins in SFR and stellar mass. Here we use data from the PRIMUS and DEEP2 galaxy redshift surveys to study the dependence of galaxy clustering on stellar mass and sSFR using a sample of over 100,000 spectroscopic redshifts at $0.2 < z < 1.2$. Our sample spans a total of five fields, which we use to quantify errors due to cosmic variance. We use deep multi-wavelength imaging in our fields to estimate stellar masses and sSFRs, from which we create multiple galaxy subsamples using cuts in both parameters. We measure cross-correlation functions of these galaxy subsamples with all galaxies in our survey at these redshifts, to better trace the underlying cosmic web and reduce our uncertainties. This paper is organized as follows. In \S\ref{sec:data} we present the relevant spectroscopic datasets used here and describe our methodology for deriving stellar masses and sSFRs. In \S\ref{sec:samples} we describe the various galaxy subsamples used in our clustering analysis. The methods used to perform the clustering analysis are presented in \S\ref{sec:methods}, and our results are given in \S\ref{sec:results}. We discuss our results in \S\ref{sec:discussion} and conclude in \S\ref{sec:conclusions}. Throughout the paper we assume a standard $\Lambda$CDM model with $\Omega_m=0.3$, $\Omega_\Lambda=0.7$, and $H_{0}=72$~km s$^{-1}$~Mpc$^{-1}$. | 16 | 9 | 1609.09090 |
|
1609 | 1609.05870_arXiv.txt | Following previous works on generalized Abelian Proca theory, also called vector Galileon, we investigate the massive extension of an SU(2) gauge theory, i.e., the generalized SU(2) Proca model, which could be dubbed non-Abelian vector Galileon. This particular symmetry group permits fruitful applications in cosmology such as inflation driven by gauge fields. Our approach consists in building, in an exhaustive way, all the Lagrangians containing up to six contracted Lorentz indices. For this purpose, and after identifying by group theoretical considerations all the independent Lagrangians which can be written at these orders, we consider the only linear combinations propagating three degrees of freedom and having healthy dynamics for their longitudinal mode, i.e., whose pure Stückelberg contribution turns into the SU(2) multi-Galileon dynamics. Finally, and after having considered the curved space-time expansion of these Lagrangians, we discuss the form of the theory at all subsequent orders. | In the search for well-motivated theories that describe the primordial universe, several attempts have been made to obtain inflationary descriptions from particle physics (the Standard Model, Supersymmetry, Grand Unified Theories, etc.; see, e.g., Refs.~\cite{Bezrukov:2012sa,Bezrukov:2007ep,Mazumdar:2010sa,Lyth:1998xn,Hertzberg:2014sza}), or from quantum theories of gravity such as Supergravity, String Theory, and Loop Quantum Gravity (see, e.g., Refs.~\cite{Ferrara:2013rsa,Olive:1989nu,Baumann:2014nda,Ashtekar:2009mm,Barrau:2010nd}). This top-down approach has been very fruitful, providing new ways to understand the structure of the high energy theories necessary to reproduce the observable properties of the Universe, ranging from the Cosmic Microwave Background Radiation (CMB) to the Large-Scale Structure (LSS). However, little is known from the observational point of view for many of these theories (those whose characteristic energy scale is much higher than the electroweak one), the CMB and LSS being, at present, the only situations in which they would have had observable consequences and would thus leave testable signatures. Since the power of the current and proposed accelerators is not going to increase as much as would be needed to directly test these theories in the foreseeable future, we need to devise another approach to the fundamental theory that describes nature. Such an approach already exists, and it boils down to the question of whether there is any choice in formulating the fundamental theory. This bottom-up approach consists in finding an action completely free of pathologies, the first of them being the Ostrogradski instability \cite{ostro} (the Hamiltonian could be unbounded from below), and satisfying a given set of assumptions, e.g., symmetry requirements. One then needs to define the material content of the universe (scalar fields, vector fields, ...), although, in principle, the construction itself and the stability requirements constrain some content and allow others so that the material content is, once the conditions are applied, somehow redefined. This very ambitious program is just beginning to be implemented, and interesting works have been carried out in which the extra material content (on top of gravity) is composed of one or many scalar fields. It was Horndeski \cite{Horndeski:1974wa} who found, for the first time, the most general action for a scalar field and gravity that produces second-order equations of motion. In general, if the Lagrangian is nondegenerate, having equations of motion of second order at most is a necessary requirement to avoid the Ostrogradsky instability \cite{Woodard:2006nt,Woodard:2015zca}. By pursuing this goal, an action is found that, however, still requires a Hamiltonian analysis in order to guarantee that the instability is not present. Horndeski's construction was rediscovered in the context of what is nowadays called Galileons \cite{Nicolis:2008in}. The Galileons are the scalar fields whose action, in flat spacetime, leads to equations of motion that involve only second-order derivatives. The idea has been extended by finding the so-called Generalized Galileons, by allowing for lower-order derivatives in the equations of motion \cite{Deffayet:2011gz,Deffayet:2013lga}. The background space-time geometry where these Generalized Galileons live can be promoted to a curved one by replacing the ordinary derivatives with covariant ones and adding some counterterms that involve nonminimal couplings to the curvature \cite{Deffayet:2009wt,Deffayet:2009mn}. The latter guarantees the equations of motion for both geometry and matter are still second order, so the Galileons, both Generalized and Covariantized, are found. This procedure is equivalent to that proposed by Horndeski for one scalar field \cite{Kobayashi:2011nu}, but it loses some interesting terms when more than one scalar field is present \cite{Kobayashi:2013ina}. The Galileon approach for scalar fields has found multiple applications in cosmology, ranging from inflation (see, e.g., Refs. \cite{Creminelli:2010ba,Kobayashi:2010cm,Mizuno:2010ag,Burrage:2010cu,Creminelli:2010qf,Kamada:2010qe,Libanov:2016kfc,Banerjee:2016hom,Hirano:2016gmv,Brandenberger:2016vhg,Nishi:2016wty}) to dark energy (see, e.g., Refs. \cite{Chow:2009fm,Silva:2009km,Kobayashi:2010wa,Gannouji:2010au,Tsujikawa:2010zza,DeFelice:2010pv,Ali:2010gr,Padilla:2010tj,DeFelice:2010nf,Mota:2010bs,Nesseris:2010pc,Gabadadze:2016llq,Neveu:2016gxp,Salvatelli:2016mgy,Shahalam:2016kkg,Minamitsuji:2016qyc,Saridakis:2016ahq,Biswas:2016bwq}). The original proposal was based on the requirement of second-order equations of motion for all the additional degrees of freedom to gravity, all of them therefore being dynamical so that the system is nondegenerate. The generalization to the so-called extended Horndeski theories also includes nonphysical degrees of freedom and thus considers degenerate theories \cite{Gleyzes:2014dya,Langlois:2015cwa,Langlois:2015skt,Motohashi:2016ftl,Crisostomi:2016czh}. Such a construction is by now well understood, and some cosmological applications have also been considered \cite{Harko:2016xip,Babichev:2016rlq,Sakstein:2016ggl,Kobayashi:2016xpl,Lagos:2016wyv,Crisostomi:2016czh,Frusciante:2016xoj,Qiu:2015aha,Akita:2015mho}. However, scalar fields are not the only possibilities as the matter content of the universe. Horndeski indeed wondered some fourty years ago what the action would be for an Abelian vector field in curved spacetime \cite{Horndeski:1976gi}. Working with curvature is a way to bypass the no-go theorem presented in Ref. \cite{Deffayet:2013tca}, which states that the only possible action for an Abelian vector field in flat spacetime that leads to second-order equations of motion is the Maxwell-type one. Relaxing the gauge invariance allows for a nontrivial action in flat spacetime, in this way generalizing the Proca action \cite{Heisenberg:2014rta,Tasinato:2014eka}. The construction of the resulting vector Galileon action has been well investigated and discussed, so there is already a consensus about the number and type of terms in the action, even in the covariantized version \cite{Allys:2015sht,Jimenez:2016isa,Allys:2016jaq}. Moreover, the analogous extended Horndeski theories have been built for a vector field \cite{Heisenberg:2016eld,Kimura:2016rzw}, and the corresponding cosmological applications have been explored \cite{Tasinato:2014eka,Tasinato:2014mia,Hull:2014bga,DeFelice:2016cri,DeFelice:2016yws,DeFelice:2016uil,Heisenberg:2016wtr}. Some cosmological applications of vector fields have been investigated, and interesting scenarios, such as the $fF^2$ model \cite{Watanabe:2009ct} and the vector curvaton \cite{Dimopoulos:2006ms,Dimopoulos:2009am}, have been devised. There is, however, an obstacle when dealing with vector fields in cosmology: they produce too much anisotropy, both at the background and at the perturbation levels, well above the observable limits, unless one implements some dilution mechanism or considers only the temporal component of the vector field (which is, however, usually nondynamical). In the $fF^2$ model, the potentially huge anisotropy is addressed by coupling the vector field to a scalar that dominates the energy density of the universe and, therefore, dilutes the anisotropy; in contrast, in the vector curvaton scenario, the anisotropy is diluted by the very rapid oscillations of the vector curvaton around the minimum of its potential. Another dilution mechanism is to consider many randomly oriented vector fields \cite{Golovnev:2008cf}; however, this requires a large number of them, indeed hundreds, so it is difficult to justify it from a particle physics point of view. There is, nevertheless, another possibility, the so-called ``cosmic triad'' \cite{Golovnev:2008cf,ArmendarizPicon:2004pm}, a situation in which three vector fields orthogonal to each other and of the same norm can give rise to a rich phenomenology while making the background and perturbations completely isotropic \cite{Rodriguez:2015xra}. A couple of very interesting models, gauge-flation \cite{Maleknejad:2011jw,Maleknejad:2011sq} and chromo-natural inflation \cite{Adshead:2012kp}, have implemented this idea by embedding it in a non-Abelian framework and exploiting the local isomorphism between the SO(3) and SU(2) groups of transformations. At first sight, the cosmic triad configuration looks very unnatural, but dynamical system studies have shown that it represents an attractor configuration \cite{Maleknejad:2011jr}. Unfortunately, although the background dynamics of these two models is successful, their perturbative dynamics makes them incompatible with the latest Planck observations \cite{Namba:2013kia,Adshead:2013nka}. Despite this failure, such models have shown the applicability that non-Abelian gauge fields can have in cosmological scenarios. Having in mind the above motivations, the purpose of this paper is to build the first-order terms of the generalized SU(2) Proca theory and to discuss the general form of the complete theory. For the most part, we focus on those Lagrangians containing up to six contracted Lorentz indices, which we obtain exhaustively. To ensure that we do not forget some terms, we first construct from group theoretical considerations all possible Lagrangians at these orders, before imposing the standard dynamical condition, i.e., that only three degrees of freedom propagate. Then, after identifying all the Lagrangians that imply the same dynamics, e.g., those related by a conserved current, we verify that the pure Stückelberg part of the Lagrangians is healthy, i.e., that it implies the SU(2) multi-Galileon dynamics. To this end, it is useful to derive all the equivalent formulations of the SU(2) adjoint multi-Galileon model, which we provide in the Appendix. Then, after computing the relevant curved space-time extension of our Lagrangians, we conclude about the status of the complete formulation of the theory, i.e., that containing the higher order terms we did not consider in this work. The layout of this paper is the following. In Section \ref{gnapc}, the generalized non-Abelian Proca theory is introduced, and some technical aspects needed for later sections are laid out; the procedure to build the theory is also described. In Section \ref{PartSinglet}, the building blocks of the Lagrangian are systematically obtained. Section \ref{cht} deals with the right number of propagating degrees of freedom and the consistency of the obtained Lagrangian with the scalar Galileon nature of its longitudinal part. The covariantization of the theory is performed in Section \ref{covproc} and the final model, together with a discussion and comparison with the Abelian case, is presented in Section \ref{FinalModel}. The appendix presents the construction of the multi-Galileon scalar Lagrangian in the 3-dimensional representation of SU(2) and its equivalent formulations. Throughout this paper, we have employed the mostly plus signature, i.e., $\eta_{\mu\nu} = \rm{diag}\,\left( -, +,+,+\right)$, and set $\hbar=c=1$. | \label{FinalModel} Let us summarize the results obtained for the generalized SU(2) Proca theory. First, we showed that any function of the vector field, Faraday tensor, and its Hodge dual (either in their Abelian or non-Abelian formulation) was possible, i.e., \begin{equation} \mathcal{L}_2 = f(A_{\mu}^{\a}, G_{\mu\nu}^\a, \tilde{G}_{\mu\nu} ^\a) = \tilde{f}(A_{\mu}^{\a}, F_{\mu\nu}^\a, \tilde{F}_{\mu\nu} ^\a). \end{equation} Such a general $\mathcal{L}_2$ term involving only gauge-invariant quantities for the derivatives is also present in the Abelian case; we will not discuss it any further since it appears similarly (and for the same reasons) in both the Abelian and non-Abelian theories. Before presenting the other terms contained in the non-Abelian action, let us pursue the summary of what was found for its Abelian counterpart, as worked out in Refs.~\cite{Heisenberg:2014rta,Allys:2015sht,Jimenez:2016isa,Allys:2016jaq}; as usual, we denote $\mathcal{L}_{n+2}$ the Lagrangians containing $n\geq 1$ first-order derivatives of the vector field. First, the relation between the more general scalar and vector theories, i.e., the Galileon and generalized Proca models, provide, in this case, a deeper understanding through the use of the Stückelberg trick to go from one sector to another (i.e., switching between $\partial_\mu \pi$ and $A_\mu$). In the scalar Galileon theory, only one term exists in the Lagrangians $\mathcal{L}_3$ to $\mathcal{L}_5$, each of which generates a contribution to the vector sector by the Stückelberg trick, i.e., those with a prefactor $f_i(X)$ in the conclusion of Ref.~\cite{Allys:2016jaq}. An additional freedom stems from the fact that a given scalar Lagrangian can give different vector Lagrangians when permuting the second-order derivatives before introducing the vector field: although $\partial_\mu\partial_\nu \pi = \partial_\nu\partial_\mu \pi$, this symmetry is absent in the pure vector case since $\partial_\mu A_\nu \not= \partial_\nu A_\mu$. This property led to one additional contribution to the vector sector of each $\mathcal{L}_4$ to $\mathcal{L}_6$. These contributions appear with the prefactor $g_i(X)$ in Ref.~\cite{Allys:2016jaq}; they vanish in the pure scalar sector. Coming back to the non-Abelian situation, and in addition to $\mathcal{L}_2$, we derived those relevant Lagrangians implying up to 6 contracted Lorentz indices and being nontrivial in flat spacetime. Contrary to the Abelian case, we found no such Lagrangian for $n=1$. For $n=2$, there are three possible terms; i.e., $\mathcal{L}_4$ contains \begin{equation} \left\{ \begin{array}{l} \mathcal{L}_4^1 = \delta^{\mu_1 \mu_2}_{\nu_1 \nu_2} A^{\lambda}_{\b} A^{\b}_{\lambda} \left(\nabla_{\mu_1} A^{\nu_1}_{\a} \right) \left(\nabla_{\mu_2} A^{\nu_2 \a} \right) + \frac14 A^{\lambda}_{\b} A ^{\b}_{\lambda} A^{\mu}_{\a} A^{\a}_{\mu} R \\ ~~~~~~~ + 2 \delta^{\mu_1 \mu_2}_{\nu_1 \nu_2} A^{\lambda}_{\a} A_{\lambda\b} \left(\nabla_{\mu_1} A^{\nu_1\a} \right) \left(\nabla_{\mu_2} A^{\nu_2 \b} \right) + \frac12 A^{\lambda}_{\b} A _{\lambda\a} A^{\mu\b} A^{\a}_{\mu} R , \\ \mathcal{L}_4^2 = \delta^{\mu_1 \mu_2}_{\nu_1 \nu_2} A^{\lambda}_{\a} A_{\lambda\b} \left(\nabla_{\mu_1} A^{\nu_1\a} \right) \left(\nabla_{\mu_2} A^{\nu_2 \b} \right) + \frac{1}{4} A^{\lambda}_{\b} A _{\lambda\a} A^{\mu\b} A^{\a}_{\mu} R \\ ~~~~~~~ + \delta^{\mu_1 \mu_2}_{\nu_1\nu_2} A_{\mu_1}^{\a} A_{\mu_2}^{\b} \left(\nabla^{\nu_1}A^{\alpha}_{\a} \right) \left(\nabla^{\nu_2} A_{\alpha \b} \right) -\frac12 A^{\mu\a} A^{\nu \b} A^{\rho}_{\a} A^{\sigma}_{\b} R_{\mu\nu\rho \sigma},\\ \mathcal{L}_4^3 = \tilde{G}_{\mu\sigma}^{\b}A^\mu_{\a} A_{\alpha\b} S^{\alpha\sigma \a},\\ \end{array} \right. \end{equation} the first two terms giving, once developed, the following forms: \begin{equation} \left\{ \begin{array}{l} \mathcal{L}_4^1 = (A_{\b} \cdot A^{\b}) \left[ \left(\nabla\cdot A_\a \right)\left(\nabla\cdot A^\a \right) - (\nabla_\mu A^\nu_{\a})(\nabla^\mu A_\nu^{\a}) +\frac14 A_{\a} \cdot A^{\a} R \right] \\ \hspace{1cm} + 2 (A_{\a} \cdot A_{\b}) \left[\left(\nabla\cdot A^\a \right)\left(\nabla\cdot A^\b \right) - (\nabla_\mu A^{\nu \a})(\nabla^\mu A_\nu^{\b}) +\frac12 A^{\a} \cdot A^{\b} R\right], \\ \mathcal{L}_4^2 = (A_{\a} \cdot A_{\b}) \left[\left(\nabla\cdot A^\a \right)\left(\nabla\cdot A^\b \right) - (\nabla_\mu A^{\nu \a})(\nabla^\mu A_\nu^{\b}) +\frac14 A^{\a} \cdot A^{\b} R\right] \\ \hspace{1cm} + (A^{\mu \a} A^{\nu\b}) \left[\left(\nabla_\mu A^\alpha_\a \right)\left(\nabla_\nu A_{\alpha\b} \right) - \left(\nabla_\nu A^\alpha_\a \right)\left(\nabla_\mu A_{\alpha\b} \right) -\frac12 A^\rho_{\b} A^{\sigma \b} R_{\mu\nu\rho\sigma}\right], \end{array} \right. \end{equation} which are more easily compared with the equivalent results for the Abelian case. Finally, we also found four extra possibilities for the Lagrangians, implying a coupling with the curvature \begin{equation} \begin{array}{l} \mathcal{L}^\mathrm{curv}_{1} = G_{\mu\nu}A^{\mu\a}A^{\nu}_{\a},\\ \mathcal{L}^\mathrm{curv}_{2} = L_{\mu\nu\rho\sigma} F^{\mu\nu}_{\a} F_{\mu\nu}^{\a},\\ \mathcal{L}^\mathrm{curv}_{3} = L_{\mu\nu\rho\sigma} \epsilon_{\a\b\c} F^{\mu \nu \a} A^{\rho\b} A^{\sigma\c},\\ \mathcal{L}^\mathrm{curv}_{4} = L_{\mu\nu\rho\sigma} A^{\mu\a} A^{\nu \b} A^{\rho}_{\a}A^{\sigma}_{\b}, \end{array} \end{equation} thereby completing the full action at that order. Let us first consider the actions whose equations of motion involve only second-order derivatives for the scalar (not first-order ones), which is equivalent to having only two vector fields together with the relevant gradients in the action. The multi-Galileon SU(2) model in the adjoint representation has been considered in \cite{Padilla:2010ir}, where it was shown that building a Lagrangian is only possible at the order of $\mathcal{L}_4$ (not to mention the order $\mathcal{L}_2$ already discussed above). The equivalent formulations of this Lagrangian are detailed in Appendix~\ref{AppendixGalileon}. Following the previous considerations, no Lagrangian in the vector sector should appear at the order of $\mathcal{L}_3$ since there is no such associated Lagrangian for the multi-Galileon at that order; we explicitly confirmed this expectation. In addition, two Lagrangians should appear at the order of $\mathcal{L}_4$, one associated with the multi-Galileon dynamics and one associated with the commutation of second-order derivatives of the scalar field. In fact, three Lagrangians have been found, two of them giving the multi-Galileon dynamics in the scalar sector. We then interpret these two previous terms as contributions which are equivalent in the scalar case but not in the vector case. The fact that there are two nonvanishing Lagrangians in the scalar sector is also due to a commutation of the second-order derivatives of the scalar fields but in a current term, which implies that it is not possible to describe this commutation with a Lagrangian vanishing in the pure scalar sector. This additional term is specific to the non-Abelian case: the term in $\delta^{\mu_1 \mu_2}_{\nu_1\nu_2} A_{\mu_1}^{\a} A_{\mu_2}^{\b} \left(\nabla^{\nu_1}A^{\alpha}_{\a} \right) \left(\nabla^{\nu_2} A_{\alpha \b} \right)$ vanishes in the Abelian case, while $\mathcal{L}_4^1$ and $\mathcal{L}_4^2$ both reduce to $\mathcal{L}_4^{\text{Abelian}} = \delta^{\mu_1 \mu_2}_{\nu_1 \nu_2} A^{\lambda}A_{\lambda} \left(\nabla_{\mu_1} A^{\nu_1}\right) \left(\nabla_{\mu_2} A^{\nu_2} \right)$. To go further, let us first consider terms implying more derivatives, i.e., having $n\geq 3$. At the order of $\mathcal{L}_5$, and since there is no possible dynamics for the SU(2) adjoint multi-Galileon, we expect that no term having a nonvanishing pure scalar contribution is possible. This suggests that the only possible term is \begin{equation} \mathcal{L}_5 = \epsilon_{\a\b\c} \left(A^\a \cdot A^\d\right) \tilde{G}^{\alpha\mu}_\d\tilde{G}^{\beta}{}_\mu^\b S_{\alpha\beta}^\c, \end{equation} with the other SU(2) index contractions giving a vanishing result. At the order of $\mathcal{L}_6$, the only possibility seems to be the independent possible contractions of SU(2) indices on $\mathcal{L}_6^{\text{Abelian}}=\left(A\cdot A \right)\tilde{G}^{\alpha\beta}\tilde{G}^{\mu\nu} S_{\alpha\mu}S_{\beta\nu}$, since there is no possibility of having a term that does not vanish in the pure scalar sector. However, one should verify that there is no other term vanishing in the pure scalar sector, not included in $\mathcal{L}_2$, and whose dynamics is not described by the previous ones. This kind of terms would be specific to a non-Abelian theory, as is the second term of $\mathcal{L}_4^2$, and they would vanish for a vector field in a trivial group representation. Concerning the Lagrangians with more than two vector fields together with the relevant gradients, one has to pay attention to the fact that fully factorizing an $f\left(A_\mu^\a\right)$ as in the Abelian case is not guaranteed to lead to a valid procedure, although factorizing such an arbitrary function in front of any valid contribution also leads to another valid contribution. In addition, one could think that if there is no valid Lagrangian with only a few nongradient vector fields at a given derivative order, it is fairly probable that there is also no such valid Lagrangian at all at this order. For instance, we showed explicitly that terms at the order of $\mathcal{L}_3$ are not possible with up to $4$ vector fields, and this questions the possibility of having such a term even with a higher number of vector fields. An interesting point is that if a Lagrangian is allowed which does not vanish in the pure scalar sector, it corresponds to a possible term in the multi-Galileon action, which shows that both theories are closely related. To conclude, this discussion showed that even if the full action of the model has not been obtained yet, discussing the low order terms permits us to identify and understand the whole Lagrangian structure. The above discussion is not specific to the SU(2) case and therefore can be extended to other group representations. For a theory with a vector field transforming under any representation of any group, a systematic study of all possible terms in the action should be performed in parallel with the corresponding multi-Galileon theory. | 16 | 9 | 1609.05870 |
1609 | 1609.00387_arXiv.txt | Extended inverse Compton halos are generally anticipated around extragalactic sources of gamma rays with energies above 100~GeV. These result from inverse Compton scattered cosmic microwave background photons by a population of high-energy electron/positron pairs produced by the annihilation of the high-energy gamma rays on the infrared background. Despite the observed attenuation of the high-energy gamma rays, the halo emission has yet to be directly detected. Here, we demonstrate that in most cases these halos are expected to be highly anisotropic, distributing the up-scattered gamma rays along axes defined either by the radio jets of the sources or oriented perpendicular to a global magnetic field. We present a pedagogical derivation of the angular structure in the inverse Compton halo and provide an analytic formalism that facilitates the generation of mock images. We discuss exploiting this fact for the purpose of detecting gamma-ray halos in a set of companion papers. | \label{sec:I} The extragalactic gamma-ray sky above 100~GeV is dominated by the unresolved emission of a subset of active galactic nuclei (AGNs) \citep{50GevBkgnd}. Of these, the vast majority are blazars -- objects with relativistic jets pointed in our direction \citep[see, e.g., Table 5 of ][]{2LAC}. While the mechanisms by which these very-high energy gamma rays (VHEGRs) are produced remain unclear \citep{1993A&A...269...67M,1998MNRAS.301..451G,2007Ap&SS.309...95B}, their propagation through the cosmos has provided an invaluable means by which to probe the intervening universe \citep{Goul-Schr:67,Stec-deJa-Sala:92,deJa-Stec-Sala:94,Sala-Stec:98,Domi_etal:11,Gilm_etal:12,Vovk+12}. Extragalactic VHEGR sources are observed to be strongly biased towards low redshifts, with the number of known sources peaking at a redshift of 0.1-0.2 \citep[see, e.g., the redshift distribution of high-syncrotron-peak sources in ][]{2LAC,3LAC}. This is a natural consequence of the annihilation of VHEGR on the nearly-homogeneous infrared-ultraviolet extragalactic background light (EBL) that permeates the universe, generated by previous generations of stars and quasars \citep[][]{Goul-Schr:67}{, which can thus be probed by propagating VHEGRs \citep{2012Sci...338.1190A,2013APh....43..112D}}. The center of momentum energy of the VHEGRs and EBL photon exceeds the pair-creation threshold, i.e., $E_\gamma E_{\rm IR}\gtrsim 4 m_e^2 c^4$; and thus VHEGRs can annihilate as they propagate through the EBL. In practice, the mean free path, $\Dpp$, of VHEGRs to absorption on the EBL is both energy and redshift dependent, depending on the evolving density and spectrum of the EBL. Higher EBL densities at larger redshifts correspond to shorter $\Dpp$, which peaks in comoving units near $z\approx1$ due to the peak in the cosmological star formation rate around that time. As such, observations of the absorbed VHEGR spectra of nearby sources have resulted in direct measurements of the EBL \citep{2015ApJ...812...60B}. From these it is clear that even today the universe is effectively optically thick to VHEGRs, with \begin{equation} \Dpp(E_\gamma,z) = \Dppn(z)\frac{E_0}{E_\gamma} = 35 \left(\frac{1+z}{2}\right)^{-\zeta} \left(\frac{E_\gamma}{1~{\rm TeV}}\right)^{-1} ~{\rm Mpc}\,, \label{eq:Dpp} \end{equation} where $E_0$ is a fiducial energy and $\zeta=4.5$ for $z<1$ and $\zeta=0$ for $z\ge1$ \citep{Knei_etal:04, Nero-Semi:09}. More recently, the evolution of the electron-positron pairs has provided a means to probe the magnetization of the intervening cosmos. The homogeneity of the EBL coupled with the large $\Dpp$ places many of these pairs within intergalactic voids, where their propagation can be affected by the intergalactic magnetic field (IGMF). These pairs thus ostensibly permit the only measurement of the large scale IGMF located in the mean density regions far removed from galactic activity. The high energy of VHEGRs imply similarly high-energy pairs, which correspond to Lorentz factors of \begin{equation} \gamma \approx \frac{E_\gamma}{2 m_\e c^2} \approx 10^6 \frac{E_\gamma}{1~\TeV}\,. \end{equation} If nothing else happens, these pairs will cool on the cosmic microwave background (CMB), producing an inverse Compton cascade (ICC) of photons with typical energies of \begin{equation} E_{\rm IC} \approx 2 \gamma^2 E_{\rm CMB} = 2 \left(\frac{E_\gamma}{1~\TeV}\right)^2 \frac{E_{\rm CMB}}{1~\meV}~\GeV\,. \label{eq:ICE} \end{equation} That is, the ICC effectively reprocesses an initial TeV gamma ray into many GeV gamma rays. It is the non-observation of this ICC component in known VHEGR sources that have provided the strongest lower limits on the IGMF to date \citep[see, e.g.,][]{Nero-Semi:09}. In a number of extragalactic VHEGR sources the intrinsic gamma-ray spectrum can now be constructed after making weak assumptions either about the intrinsic spectrum or the absorption on the EBL, and thus the resulting ICC emission estimated. This is then limited directly by observations by the \Fermi gamma-ray space telescope, which has ruled out the presence of the ICC component with extraordinary confidence \citep[e.g.,][]{Nero-Semi:09}. This is natural if an IGMF is present -- within the IGMF the electron-positron pairs deflect away from the line of sight and therefore the up-scattered gamma rays are beamed away from us. Based upon this scenario typical estimates for the IGMF range from $10^{-17}$--$10^{-15}$~G, depending on assumptions on duty cycles \citep[see e.g.,][]{Nero-Semi:09,Nero-Vovk:10,Tave_etal:10a,Tave_etal:10b,Derm_etal:10, Tayl-Vovk-Nero:11,Taka_etal:11,Dola_etal:11,HESS:2014,Prokhorov:2016}. This argument for non-zero IGMFs is predicated on three key assumptions: \begin{enumerate} \item The intrinsic TeV emission is narrowly beamed. \item The intrinsic TeV emission spectrum can be reasonably approximated, usually by an exponentially cut-off power law. \item No other processes control the evolution of the electron-positron pairs. \end{enumerate} The first is well supported by the prevalence of blazars among VHEGR-bright AGN specifically, and gamma-ray bright AGN generally \citep{2LAC,3LAC,TeVCat:2008}, which immediately implies that VHEGR emission is localized near the axis of the radio jet. The second is reasonably well supported by the gamma-ray spectra of nearby VHEGR-bright AGN, the systematic softening of observed gamma-ray spectra with increasing redshift \citep{2012Sci...338.1190A}, and the underlying assumption that the intrinsic spectra only weakly evolve. The third remains unclear. Should any alternate cooling mechanism dominate inverse Compton cooling it would preempt the generation of ICCs directly. A recently suggested example would be cooling mediated by large-scale beam plasma instabilities driven by the bulk motion of the relativistic electron-positron pairs through the ionized intergalactic medium \citep{PaperI,Schl_etal:12,Schlickeiser:2013,Chang:2014}. While the nonlinear development of these instabilities is uncertain, there are a variety of lines of astronomical evidence that suggest a cooling mechanism with very similar properties is at work \citep{PaperII,PaperIII,PaperIV,PaperV,2015ApJ...811...19L}. Regardless of the origin of the additional cooling, however, should the ICCs be preempted the resulting gamma-ray spectra would necessarily be consistent with the current lack of a detection of an ICC in known extragalactic VHEGR sources, independent of an IGMF. The situation would be immediately clarified by the direct detection of the ICC component, laying squarely within the energy range probed the by large-area telescope (LAT) on \Fermi. Not only would doing so obviate the above assumptions (importantly including the third), it would also settle questions regarding the duration of VHEGR outbursts and thereby reduce the uncertainty on the IGMF lower limits substantially \citep{Derm_etal:10}. Even in the presence of an IGMF the ICCs are deflected away from the axis along which the original VHEGR emission is beamed. Thus the ICC component should be visible for observers who either do not see the VHEGR emission, or see only weak VHEGR emission. Efforts to directly detect the ICC component are fundamentally complicated by the large mean free paths of the VHEGRs, typically resulting in halos that extend over many degrees and therefore having a low surface brightness (see, e.g., Figure \ref{fig:size}). Therefore, all efforts to detect this emission to date have stacked multiple \Fermi gamma-ray images to increase the significance with which the halo emission can be separated from that due to the central source and background. These efforts are further complicated operationally by the uncertainties in the point spread function (PSF) of \Fermi and the spatially varying gamma-ray background \citep[e.g.,][]{Nero-Semi-Tiny-Tkac:11,FLAT-stack:2013}. As a result, this procedure has led to now disproven detections of an excess \citep{Ando:2010}. A more recent attempt that utilizes the most recent PSF, reported in \citet{Chen:2015}, nevertheless exhibits similar sensitivies to the uncertain instrument response. All such efforts have ignored the possibility of structure within the gamma-ray halo. However, such structure is a natural consequence of either the original beaming of the VHEGRs responsible for the generation of the pairs or the orientation of the IGMF where the pairs are created \citep[see, e.g.,][]{2015JCAP...09..065L}. Thus for a wide range of parameters for an IGMF we expect highly anisotropic ICC halo. Here we present semi-analytical computations of the halo structure, explicitly demonstrating the presence of the structure, identifying its origin, and creating the facility to generate mock images of VHEGR sources with realistic ICC halo structures. In principle, knowledge of the halo structure can aid substantially in efforts to directly detect the ICC component. We report on an explicit implementation of a method to do so in a companion paper \citep{BowTiesII}. In a companion letter, we apply this formalism to \Fermi-LAT data and discuss the consequences of this measurement for the IGMF \citep{BowTiesIII}. In Section \ref{sec:qualexp} we describe qualitatively the origin of the anisotropy and how it relates to the structure of the IGMF. General expressions describing the generation and evolution of the energetic pairs are presented in Section \ref{sec:geneqs}. Applications to cases of highly tangled and ordered fields are described in Sections \ref{sec:GHiso} and \ref{sec:GHweak} with typical applications shown. The construction of mock \Fermi images, including various components and instrumental effects is dicussed in Section \ref{sec:mocksgen}. The dependence of the mock ICC halos for a typical bright \Fermi source are explored in Section \ref{sec:mocksexpl}. Finally, concluding remarks are collected in Section \ref{sec:conc}. In order to streamline the paper, most of the demonstrations are left to the Appendix. \begin{figure} \includegraphics[scale=.45]{fig1.eps} \caption{Angular size of VHEGR mean free path as a function of redshift and energy.} \label{fig:size} \end{figure} | \label{sec:conc} The putative gamma-ray halos that surround bright VHEGR sources are generally highly structured. The reason for and degree of structure within the ICC halo depends most strongly on the strength and geometry of the IGMF and jet orientation. For small-scale, tangled IGMFs ($\lambda_B\ll3~\Mpc$) the structure arises solely from the jetted nature of the VHEGR emission. Oblique viewing angles ($\Theta\gg\theta_j$) produce clear bimodal gamma-ray features. In contrast, acute viewing angles ($\Theta\lesssim\theta_j$) result in foreshortened ICC halos, suppressing the anisotropy. For large-scale, uniform IGMFs ($\lambda_B\gg3~\Mpc$) the structure arises from the ordered gyration of the pairs, i.e., their limited phase-space evolution. This results in a pair of thin halo features whose size and strength are indicative of the magnetic field strength and viewing angle, respectively. Strong fields result in large deflection angles and therefore extended halos. Acute viewing angles produce nearly-symmetric bimodal halos; oblique viewing angles generate increasing asymmetry. For nearby VHEGR sources, i.e., $z<1$, the asymmetric halo structures typically remain visible after convolution with the LAT instrument response. This is a consequence of their typically large angular extent. However, for large-scale, uniform IGMFs with strength smaller than $10^{-17}~\G$ this is not true, generating halos that are fully contained within the effective PSF of the intrinsic source photons. As a result, for acute viewing angles the ICC halos arising in a sufficiently strong, large-scale, uniform IGMF are readily identifiable. Similarly, for oblique viewing angles a sufficiently strong, small-scale, tangled IGMF is clearly evident. These conclusions are quantitatively but not qualitatively dependent on the remaining source parameters: redshift, jet opening angle, where the event converted in the LAT (i.e., front- vs back-conversion). The structure is independent of the total source flux, background, and angular structure of the LAT. In \citet{BowTiesII} we present a statistical scheme to identify the presence of gamma-ray halos around \Fermi sources that exploits their nearly symmetric, bimodal structure. In \citet{BowTiesIII} we apply this to the existing sample of suitable \Fermi blazars, placing constraints on the geometry and strength of the IGMF. | 16 | 9 | 1609.00387 |
1609 | 1609.03271_arXiv.txt | Observations by the Atacama Large Millimetre/sub-millimetre Array of the 358 GHz continuum emission of the gravitationally lensed quasar host RX J0911.4+0551 have been analysed. They complement earlier Plateau de Bure Interferometer observations of the CO(7-6) emission. The good knowledge of the lensing potential obtained from Hubble Space Telescope observations of the quasar makes a joint analysis of the three emissions possible. It gives evidence for the quasar source to be concentric with the continuum source within 0.31 kpc and with the CO(7-6) source within 1.10 kpc. It also provides a measurement of the size of the continuum source, 0.76 $\pm$ 0.04 kpc FWHM, making RX J0911.4+0551 one of the few high redshift galaxies for which the dust and gas components are resolved with dimensions being measured. Both are found to be very compact, the former being smaller than the latter by a factor of $\sim$3.4$\pm$0.4. Moreover, new measurements of the CO ladder $-$ CO(10-9) and CO(11-10) $-$ are presented that confirm the extreme narrowness of the CO line width (107$\pm$20 km s$^{-1}$ on average). Their mere detection implies higher temperature and/or density than for typical quasar hosts at this redshift and suggests a possible contribution of the central AGN to gas and dust heating. The results are interpreted in terms of current understanding of galaxy evolution at the peak of star formation. They suggest that RX J0911.4+0551 is a young galaxy in an early stage of its evolution, having experienced no recent major mergers, star formation being concentrated in its centre. | Our knowledge of galaxy evolution at early cosmic times, in particular at the epoch when star formation and AGN activity reach their maximum, with redshift in the $\sim$2 to $\sim$4.5 range, owes much to the study of quasar hosts (for a review see \citealt{Carilli2013}, for recent new results see \citealt{Aravena2016,Glikman2016}). Four main actors of this evolution are accessible to observation: the AGN emission around the supermassive black hole in the centre, the gas reservoir from which stars are formed, its dust content $-$ the emission of which is a proxy of star formation $-$ and the stars themselves. The fifth main actor, dark matter, is not directly accessible to observation. Typically, each of these covers a specific frequency range in the rest frame: optical and X rays for the black hole, millimetre/sub-millimetre for the molecular gas, infrared for the dust and optical for the stars. These observations are reduced to a few quantities that summarize our knowledge of the observed galaxy, such as masses and luminosities of the above mentioned components, star formation rate, dynamical mass, starburstiness, etc. The molecule that is most commonly used as a tracer of the gas component is carbon monoxide; the emission of the molecular rotation levels probes different regions depending on the excitation energy. Other tracers, such as carbon and nitrogen fine structure lines, have also been detected successfully. It is only recently that some gas and dust components of high redshift quasar hosts and sub-millimetre galaxies could be spatially resolved, usually taking advantage of the important magnification provided by gravitational lensing \citep{Riechers2011a,Riechers2011b,Riechers2011c}. One of these is the host galaxy of RX J0911.4+0551 (hereafter referred to as RX J0911), which we studied earlier using Plateau de Bure Interferometer observations \citep{Tuananh2013,Tuananh2014,Hoai2013} and which is the subject of the present article. While many quasar hosts and sub-millimetre galaxies hosting an AGN have been detected at redshifts between $\sim$2 and $\sim$4.5, most of them have been observed in the continuum and only few in molecular or atomic lines. Table \ref{Table1} lists some typical cases chosen among the better resolved. In the case of gravitational lensing it is not sufficient to resolve the lensed images with good spatial resolution; in addition, a precise knowledge of the lensing potential is mandatory in order to resolve the source and measure its size. The resolved images, while giving often evidence for extended dust and gas components, the former being relatively more compact, benefit rarely from a sufficient spatial resolution for their morphology to be reconstructed in the source plane. From the study of high redshift galaxies, one learns that they are often the seat of mergers, in particular wet mergers that are identified by comparing the respective locations of the optical, gas and dust components. Examples are J 123707, SMM J02399 and BRI 1335. Typically, mergers are identified with dimensions in excess of a kiloparsec in the source plane. Mergers cause the gravitational field to strongly increase locally, in particular as the result of induced shocks and turbulences, triggering the local collapse of gas clouds and causing star bursts. These are seen as important sources of dust away from the central supermassive black hole. In cases where the gas and dust components are roughly concentric with the supermassive black hole, with dimensions not exceeding a kiloparsec in the source plane, a velocity gradient is often observed that reveals a disc-like rotating structure; in most cases these give no evidence for recent mergers but the better resolution observations may display a clumpy structure of the source, possibly revealing the effect of merging: such is the case of SDP.81, the best resolved of all cases. In summary, understanding the genesis of early galaxies requires observations made at different stages of their evolution, in order to reveal the relative roles played by each actor as a function of time. At each stage, multi-wavelength observations are mandatory in order to disentangle the respective morphologies of the gas and dust components and their locations with respect to the central supermassive black hole. The next section, Section 2, recalls what is known of RX J0911; Section 3 introduces the new observations and the reduction of the data, with Section 4 discussing the relative positions of the quasar and the gas and dust components and Section 5 discussing their sizes. Section 6 locates the continuum flux measurement on the SED and Section 7 discusses the line data. The article closes with a summary of the main results and their interpretation in the framework of our present knowledge of galaxy evolution at high redshifts. \begin{table*} \centering \caption{Some typical quasar hosts and sub-millimetre galaxies at redshifts in the 2.5 to 4.5 range for which both the gas and dust components have been resolved and measured. References are given for the line emission data, the continuum emission data and the lensing mechanism. In the last column, mergers are indicated as M and others as NM; in both cases, the figure gives the scale of the emission in kpc.} \label{Table1} \begin{tabular}{|c|c|c|c|c|c|} \hline Name & $z$ & Ref. line & Ref. continuum & Ref. lens & Mergers \\ \hline J123707+6214 & 2.5 & CO(1-0) \& (5-4)[20] & 1.4 GHz, [20,21] & unlensed & M, 20 \\ \hline Cloverleaf & 2.56 & CO(7-6) [6] & 122 microns [7] & [6] & NM, 0.8 \\ \hline SPT0538-50 & 2.78 & CO(1-0) \&(3-2) [8] & 860 microns [9] & [8,9] & M, 1.6 \\ \hline RX J0911 & 2.8 & CO(7-6) [1,2,4] & [5] & [3,4] & NM, 0.8 \\ \hline SMMJ02399-0136 & 2.81 & CO(1-0) [10] & 122 microns [7] & weakly lensed [11] & M, 25 \\ \hline SDP.81 & 3.04 & CO(5-4) \& (8-7) [12,14] & 236 and 290 GHz [12,13] & [13,14] & M, 8\\ \hline APM 08279+5255 & 3.91 & CO(1-0) [15] & 2.6 mm [15] & [15] & NM, 0.5 \\ \hline PSS J2322+1944 & 4.12 & CO(2-1) [16] & 1.4 GHz [17] & [16,17] & NM, 2 \\ \hline BRI 1335-0417 & 4.41 & CO(2-1) [18] & 1.4 GHz [19] & unlensed & M, 5 \\ \hline \end{tabular} 1. \citet{Weiss2012}; 2. \citet{Tuananh2013}; 3. \citet{Hoai2013}; 4. \citet{Tuananh2014}; 5. ALMA archive, this work; \\6. \citet{Venturini2003}; 7. \citet{Ferkinhoff2015}; 8. \citet{Spilker2015}; 9. \citet{Hezaveh2013}; 10. \citet{Ivison2010};\\ 11. \citet{Richard2009}; 12. \citet{ALMA2015}; 13. \citet{Rybak2015a}; 14. \citet{Rybak2015b};\\ 15. \citet{Riechers2009a}; 16. \citet{Riechers2008a}; 17. \citet{Carilli2001} \& \citet{Carilli2003}; 18. \citet{Riechers2008b}; 19. \citet{Momjian2007}; 20. \citet{Riechers2011e}; 21. \citet{Morrison2010}.\\ \end{table*} | \subsection{Summary} The new observations of the quasar host RX J0911 that have been presented here are an important complement to what was already known of this galaxy. It occupies an eccentric situation in the family of quasar hosts at $z\sim$3, with CO luminosity (related to the gas mass), FIR luminosity (related to the dust mass), line width (related to the dynamical mass) and X ray luminosity (related to the black hole mass) typically five times smaller than usual for such galaxies (Figure \ref{fig3}, \citet{Carilli2013}). The main contribution of the present work is evidence for the gas and dust components to be compact and concentric with the central black hole, the former being \mbox{$\sim$3.4$\pm$0.4} times more extended than the latter. The availability of HST images, showing also the lens galaxy, has made it possible to define the lensing mechanism with particularly good precision in the image plane, $\pm$29 mas in $x$ and $\pm$11 mas in $y$. The concentricity of the three sources, quasar, gas and dust, measured to better than 0.31 kpc for the quasar versus \mbox{358 GHz} continuum emissions and to better than 1.1 kpc for the quasar versus CO(7-6) emissions, argues against any important and recent merger contribution. The compactness of the dust and gas sources, with respective FWHM values of 0.76$\pm$0.08 kpc and 2.6$\pm$0.3 kpc, adds to the argument. Another important contribution is the detection of high J excitations of the CO molecule. It is remarkable that they confirm the extreme narrowness of the CO line (\mbox{107$\pm$20 km s$^{-1}$} on average). The measurements of the CO ladder displayed in Figure \ref{fig11} as ratios to $S_{1-0}$ extend beyond the typical ladder predicted by a single component Large Velocity Gradient (LVG) approximation, suggesting that the temperature and/or density of the gas are/is on the high side. The sensitivity and spatial resolution with which the CO(11-10) and CO(10-9) emissions have been observed are insufficient to resolve the gas component and measure its size as could be done for CO(7-6), but their mere detection suggests a possibly significant contribution of the central AGN to gas and dust heating. In the remaining of this section, we explore the consequences of the present results in the framework of our present knowledge of galaxy evolution at high redshifts. \subsection{Comparison with high redshift galaxies for which the sizes of both the gas and dust components have been measured} Of the galaxies listed in Table \ref{Table1}, five are interpreted as hosting major wet mergers (J123707+6214, SPT0538-50, SMMJ02399-0136, SDP.81 and BRI 1335-0417). One, \mbox{PSS J2322+1944}, a $z$=4.12 quasar host with a star formation rate of $\sim$680 solar masses per year, displays a complex image pattern interpreted as resulting from the lensing of a source significantly offset from the central quasar and hosting interaction possibly caused by a major merger. This leaves only two galaxies with gas and dust distributions comparable with those of RX J0911: the Cloverleaf ($z$=2.56) and APM 08279+5255 ($z$=3.91). The CO emission of the Cloverleaf, with line widths of 300 to 400 \mbox{km s$^{-1}$} FWHM, is interpreted by \citet{Venturini2003} as originating from a rotating disc-like structure with a characteristic radius of 0.8 kpc, concentric with the quasar. \citet{Bradford2009} measure the CO ladder up to J$_\textrm{up}$=9 with no sign of decrease and suggest that the Cloverleaf host is undergoing a massive starburst, but that it has additional energy input into the ISM via hard X-rays originating in the accretion zone. \citet{Riechers2011d} and \citet{Weiss2003} measure CO emission with \mbox{$L'_\textrm{CO}$$\sim18$$\times$$10^9$ K km s$^{-1}$ pc$^2$}. The FIR luminosity is 22$\times10^{11}$ solar luminosities and the dust emission is confined to the same central region as the gas emission. Unfortunately, the lens parameters are difficult to evaluate with precision. \citet{Granato1996} measure an X-ray luminosity of \mbox{3.5$\times$10$^{46}$ erg s$^{-1}$} giving a black hole mass of 6.8$\times$$10^8$ solar masses. APM 08279+5255, a radio-quiet quasar host, has CO line widths in the 400 to 500 km s$^{-1}$ range and is lensed into a quad in a similar configuration to RX J0911. \citet{Riechers2009a} give a very detailed analysis of the multi-wavelength observations made of this galaxy and argue in favour of a modest lensing magnification, $\sim$4. The size of the CO component is $\sim$1.1 kpc FWHM, the CO and FIR luminosities are respectively \mbox{26$\times$10$^9$ K km s$^{-1}$ pc$^2$} and 240$\times$10$^{11}$ solar luminosities; together with a black hole mass of $\sim$230$\times$10$^8$ solar masses, this makes APM 08279+5255 a dust- and gas-rich galaxy with a very massive, active black hole in its centre. The gas in the central region is accordingly both dense and warm, the contribution of the starburst to the heating of the gas and dust being rather modest: ~10\% as estimated by \citet{Riechers2009a} and 35\% as estimated by \citet{Ferkinhoff2010}. Evidence is obtained for the central black hole growing faster than the stars in the early phase of galaxy formation. In addition to the galaxies listed in Table \ref{Table1}, there exist several other high redshift galaxies for which the sizes of the dust and gas components have been measured and found compact. A complete review is beyond the scope of the present article but two examples are worth quoting as an illustration: \mbox{MACSJ0032-arc} and \mbox{SDSS J1148}. \mbox{MACSJ0032-arc} \mbox{\citep{Dessauges-Zavadsky2016}} is a $z$=3.63 main sequence star forming galaxy lensed into an arc structure by a complex set of lenses with a magnification of order 60. The star formation region, with a rate in excess of 3000 solar masses per year, is concentrated in the centre of the galaxy, with most of the gas and dust residing within in a diameter of $\sim$1.2 kpc. It is surrounded by two UV-bright star-forming regions, possibly revealing a merger. Most of the star forming rate originates from thermal dust emission. CO emission, with line widths in excess of 300 km s$^{-1}$ FWHM, remains high up to J=6 but higher J levels have not been observed. SDSS J114816.64+525150.3 is a $z$=6.42 quasar host with a black hole of $\sim$30$\times$10$^8$ solar masses \citep{Willott2003} in its centre. CO emission, with line widths of \mbox{$\sim$300 km s$^{-1}$} FWHM, and dust emission, with a far infrared luminosity of $\sim$120$\times$10$^{11}$ solar luminosities, are concentric and confined within radii of $\sim$2.5 kpc and 0.75 kpc respectively \citep{Riechers2009b}. In summary, compactness of the star formation region about the galactic centre is a property shared by galaxies covering a very broad range of scales, with black hole masses up to two orders of magnitude above that of RX J0911. Its main virtue is to exclude major mergers and multiple or eccentric star bursts. \subsection{Comparison with high redshift galaxies displaying narrow CO lines} RX J0911 has the narrowest CO line of all high $z$ galaxies known to us. Table \ref{Table5} lists such galaxies having a CO line width not exceeding 200 km s$^{-1}$ \citep{Carilli2013}. The CO line width is expected to result from several possible causes such as rotation of the gas around the centre of the galaxy, multiple sources associated with different starbursts in a same galaxy, merging galaxies, etc. Indeed, most ultra-luminous infrared galaxies and many sub-millimetre galaxies are known to be the seats of wet mergers and/or multiple starbursts. In the case of RX J0911, the evidence for ellipticity of the gas component (3.3 standard deviations away from a circular source hypothesis) and for velocity gradient along the major axis (25 km s$^{-1}$ kpc$^{-1}$, 4.5 standard deviations from zero) make it unlikely that the gas reservoir be a rotating disc seen close to face on. Moreover, the limits obtained in the present work on the size of the gas and dust components (respectively 2.6$\pm$0.3 and 0.76$\pm$0.08 kpc FWHM) and their concentricity with respect to the quasar make multiple star bursts and/or wet mergers equally unlikely. \begin{table*} \centering \caption{Galaxies having redshift in excess of 2 and a CO line width not exceeding 200 km s$^{-1}$. The type (column 3) is abbreviated as QSO for quasars, SMM for sub-millimetre galaxies and LBG for Lyman-break galaxies. The FWHM of the CO line (column 5) is averaged over all observed lines. ${L_\textrm{FIR}}$, ${L'_\textrm{CO}}$ and $\textrm{SFR}$ and M$_\textrm{BH}$ (columns 6 to 8) are respectively the far-infrared luminosity in units of $10^{11}$ solar luminosities, the CO luminosity in units of $10^9$ K km s$^{-1}$ pc$^2$ and the star formation rate in units of solar masses per year. Column 9 gives references (a to p) and comments (1 to 3). References stand for: a) \citet{Simpson2012}; b) \citet{Walter2011}; c) \citet{Weiss2005}; d) \citet{Downes2003}; e) \citet{Harris2010}; f) \citet{Baker2004}; g) \citet{Riechers2010}; h) \citet{Riechers2011d}; i) \citet{Ao2008}; j) \citet{Vandenbout2004}; k) \citet{Weiss2005}; l) \citet{Coppin2007}; m) \citet{Lestrade2010} and \citet{Lestrade2011}; n) \citet{Coppin2010}; o) \citet{Combes2012}; p) \citet{Wang2010}. Comments stand for: 1) estimates a gas component size of 700 pc from indirect luminosity arguments; 2) assuming a magnification of 100; 3) the source gives evidence for a merger of two galaxies, one of which only displays a narrow CO line.} \label{Table5} \begin{tabular}{|c|c|c|c|c|c|c|c|c|} \hline Source & $z$ & Type & J$_\textrm{up}$ & FWHM & $L_\textrm{FIR}$ & $L'_\textrm{CO}$ & SFR & R/C \\ \hline IRAS F10214+4724 & 2.29 & QSO & 1,3,4,6,7 & 215$\pm$7 & 35 & 5.8 & - & hijk \\ \hline J0908-0034 & 2.55 & QSO & 3 & 125$\pm$25 & 31 & 7.7 & - & a \\ \hline SMM J14011+0252 & 2.57 & QSO & 1,2,3,7 & 192$\pm$9 & 2.6 & 4 & - & bcde;1 \\ \hline MS1512-cB58 & 2.73 & LBG & 3 & 174$\pm$43 & 0.8 & 0.6 & 24 & fg \\ \hline RX J0911.4+0551 & 2.80 & QSO & 1,7,10,11 & 107$\pm$20 & 15 & 4 & 230 & - \\ \hline J213512.73-010143 & 3.07 & LBG & 3 & 190$\pm$24 & 3.4 & 3 & 60 & lg \\ \hline MM18423+5938 & 3.93 & SMM & 1,2,4,6,7 & 188$\pm$10 & 48 & 2.7 & 830 & m;2 \\ \hline J033229.4-275619 & 4.76 & QSO & 2 & 160$\pm$65 & 60 & 20 & 1000 & n \\ \hline J091828.6+514223 & 5.24 & SMM & 2,5,6,7 & 145$\pm$15 & 100 & 20 & 1600 & o;3 \\ \hline J 1044-0125 & 5.78 & QSO & 6 & 160$\pm$50 & 52 & 8 & 530 & p \\ \hline \end{tabular} \end{table*} To the extent that the CO line width is directly related to the star velocity dispersion ($\sigma^*$), it is expected to be strongly correlated with the mass of the central black hole (M$_\textrm{BH}$). Such a tight correlation is indeed found for local quasars \citep[for a recent review, see][]{Sheinis2016}. However, this is no longer the case for high redshift quasars \citep{Wang2010} as illustrated in Figure \ref{fig11} (middle panel). Most of the high-$z$ CO-detected quasars are above the local M$_\textrm{BH}$-$\sigma^*$ relation with offsets exceeding one order of magnitude in black hole mass \citep{Shields2006}. Only a small part of these offsets can be blamed on the inclination of the disc on the sky plane. This reflects the observation of large mass ratios between the black hole and the bulge, an effect of the early and more rapid formation of the central supermassive black hole with respect to the bulge. In the case of RX J0911, the local relation predicts a black hole mass nearly 300 times too small or a CO line width nearly 5 times too large, depending on what one starts from. In this context, it is interesting to compare RX J0911 with the high redshift galaxies listed in Table \ref{Table5} and illustrated in Figure \ref{fig11} (right panel). RX J0911 is the only galaxy for which a measurement of the size of the gas and/or dust component is available but a common feature observed in most of these galaxies is the absence of an extended massive gas reservoir; in most cases, their properties are qualitatively similar to those of the lower-$z$ sub-millimetre galaxies studied by \citet{Greve2005}. As many high redshift luminous galaxies display evidence for mergers and/or multiple starbursts, these are likely to contribute a significant part of the CO line width; as a consequence, the observation of a narrow line width is evidence against such activity, irrespective of the inclination of the gas disc with respect to the sky plane. In the present case of RX J0911, the evidence for a velocity gradient along the elongation of the gas component suggests that most, if not all, of the CO line width is likely to be associated with gas rotation. In the cases where the CO ladder has been measured, RX J0911 is the only galaxy listed in Table \ref{Table5} for which high J lines have been detected. Earlier observations have suggested that in the others the main source of gas heating is from star formation rather than from the central AGN, but one must underline that before the recent availability of band 7 detection at ALMA, searching for such high excitations was simply out of reach. In the case of RX J0911, the strong concentration of the star formation region around the central black hole makes it difficult to disentangle the respective contributions of stars and AGN to gas heating and a significant contribution of the latter cannot be excluded. \subsection{Comparison with high redshift galaxies displaying high level excitation of the CO molecules} Table \ref{Table6} lists high redshift galaxies for which CO excitations of J$_\textrm{up}$ of 9 or above have been detected. Other galaxies, such as J1148 \citep{Riechers2013} and BR 1202 \mbox{\citep{Salome2012}}, which remain at a high level of excitation up to J$_\textrm{up}$=8, may be of the same category but higher J excitations have not been searched for. CO ladders are commonly interpreted in the framework of the Large Velocity Gradient approximation \mbox{\citep{Greve2014}}, based on the existence of important turbulence allowing photons to escape in spite of optical thickness. Temperatures of $\sim$40 K are obtained for typical high redshift galaxies \mbox{\citep{Carilli2013, Weiss2007a}} assuming molecular hydrogen densities of $\sim$10$^3$ cm$^{-3}$. Higher J excitations can be accounted for by a temperature typically 20 K larger with molecular hydrogen densities of $\sim$10$^4$ cm$^{-3}$. However, different excitations probe different temperatures and therefore different regions: a two component description is likely to be more appropriate in cases where the central AGN contributes significant heating. High J excitations have also been used as probes of the gas heating mechanism in local starbursts and/or mergers, such as NGC 6240 and Arp 193 \mbox{\citep{Papadopoulos2014}} and found to peak at moderate values of J$_{\textrm{up}}$. On the contrary, starburst galaxies M82 \citep{Panuzzo2010} and NGC 253 \citep{Rosenberg2014} display a CO ladder that stays high up to J$_{\textrm{up}}$=13, requiring an additional heating source, most likely mechanical, associated with shocks and/or turbulences. Both sub-millimetre galaxies listed in Table \ref{Table6}, \mbox{J213511} and \mbox{HLSW-01}, have CO ladders peaking at J$_\textrm{up}$ $\sim$ 6 to 7. The four other galaxies (mostly QSOs), including RX J0911, display excitations that remain high up to the larger values of J$_\textrm{up}$. The Cloverleaf and APM 08279 have been discussed in Section 8.2. In the latter case heating from the central AGN is held responsible for the higher J excitation. In the case of the Cloverleaf, both sources seem to contribute, however with dominance from the starburst \mbox{\citep{Riechers2011f}}. SPT 21323 is a $z$=4.77 lensed hyper-luminous infrared galaxy with a star formation rate of 1120$\pm$200 solar masses per year. The excitation remains high at J$_\textrm{up}$=12, the very short depletion timescale (34$\pm$13 Myr) indicates that this source is an extreme starburst with moderate gas content, but very high star formation efficiency. Here again, as in Section 8.2, high J$_\textrm{up}$ CO excitation is a property shared by galaxies covering a very broad range of scales, with black hole masses up to two orders of magnitude above that of RX J0911. It reveals the importance of central heating, either directly from the AGN or from an intense star burst surrounding it. \begin{table} \centering \caption{High redshift galaxies for which CO excitations of J$_\textrm{up}$ of 9 or above have been detected. The type (column 3) is abbreviated as QSO for quasars, SMM for submillimetre galaxies and ULIRG for Ultrahigh luminosity infrared galaxies. L$_\textrm{FIR}$ (column 5) is the far-infrared luminosity in units of 10$^{11}$ solar luminosities. Column 6 gives references (1 to 7): 1) \citet{Danielson2011}; 2) \citet{Bradford2009}; 3) \citet{Scott2011}; 4) \citet{Bradford2011}; 5) \citet{Downes1999}; 6) \citet{Weiss2007b}; 7) \citet{Bethermin2016}} \label{Table6} \begin{tabular}{|c|c|c|c|c|c|} \hline Source & $z$ & Type & J$_\textrm{up}$ & $L_\textrm{FIR}$ & Ref. \\ \hline J213511 & 2.32 & SMM & 1,3,4,5,6,7,8,9 & 23 & 1 \\ \hline Cloverleaf & 2.56 & QSO & 1,3,4,5,6,7,8,9 & 22 & 2 \\ \hline RX J0911 & 2.80 & QSO & 1,7,10,11 & 15 & - \\ \hline HLSW-01 & 2.96 & SMM & 1,3,5,7,9,10 & 143 & 3 \\ \hline APM08279 & 3.91 & QSO & 1,2,4,6,8,9,10,11 & 240 & 4,5,6\\ \hline SPT 21323 & 4.77 & ULIRG & 2,5,12 & 68 & 7\\ \hline \end{tabular} \end{table} \begin{figure*} \centering \includegraphics[height=5.cm,trim=0.cm 0.cm 0.cm 0.cm,clip]{fig11.eps} \includegraphics[height=5.cm,trim=0.cm 0.cm 0.cm 0.cm,clip]{fig11_002.eps} \includegraphics[height=5.5cm,trim=0.cm 0.cm 0.75cm 0.cm,clip]{fig11_003.eps} \caption{Left: CO emission ladder normalised to CO(1-0) for rotational states as a function of the angular momentum $J$ of the initial state \citep[from][]{Carilli2013}. Measured fluxes are displayed for different kinds of galaxies separately as indicated in the insert. The results of the present work are shown as blue crosses. The blue line is drawn to guide the eye. Middle: Black hole masses (M$_\textrm{BH}$) vs bulge velocity dispersion ($\sigma$) \citep[from][]{Wang2010}. The dashed line and filled circles denote the local M$_\textrm{BH}$-$\sigma$ relationship \mbox{\citep{Tremaine2002}}. The open diamonds are for high redshift quasars with 1.4<$z$<5. The squares are for $z$$\sim$6 quasars \mbox{\citep{Wang2010}}. The red filled circle is for RX J0911. Right: CO line width (FWHM) vs. redshift for the galaxies listed in Table \ref{Table5}.} \label{fig11} \end{figure*} \subsection{Conclusion} According to current understanding, galaxies first formed from the gravitational collapse of gas and dark matter inhomogeneity, with rapid production of massive short-lived stars resulting in coeval and prompt generation of the central black hole and enrichment of the ISM in metallic species. Subsequent growth occurred via both mergers and additional gas accretion from the intergalactic medium, with star formation rates reaching a maximum at redshifts of order $\sim$3, when the constantly decreasing ratio of gas to star mass crossed unity. How does RX J0911 fit into such a picture? Its main characteristics are the compactness of the dust component, the narrow width of the CO emission lines, the extension of the CO ladder to high J values and overall scaled-down values of relevant masses and luminosities in comparison with quasar hosts at similar redshifts. The most natural interpretation is that of a young galaxy in an early stage of its evolution, having experienced no recent major mergers, star formation being concentrated in its centre. When compared with other high redshift star forming galaxies, the Cloverleaf may be that closest to resemble it, although scaled-up by a factor of order 4. Additional observations are necessary to deepen our understanding of the physics governing the evolution of RX J0911. In particular, the high spatial resolution available from ALMA long baselines should be exploited to map the emission of CO lines for at least two representative values of J$_\textrm{up}$, such as 5 and 10, as well as the emission of atomic lines, such as [C$_\textrm{II}$] and of high dipole moment molecules. | 16 | 9 | 1609.03271 |
1609 | 1609.06348_arXiv.txt | Recent analysis of strongly-lensed sources in the Hubble Frontier Fields indicates that the rest-frame UV luminosity function of galaxies at $z=$6--8 rises as a power law down to $\MUV=-15$, and possibly as faint as -12.5. We use predictions from a cosmological radiation hydrodynamic simulation to map these luminosities onto physical space, constraining the minimum dark matter halo mass and stellar mass that the Frontier Fields probe. While previously-published theoretical studies have suggested or assumed that early star formation was suppressed in halos less massive than $10^9$--$10^{11}\msun$, we find that recent observations demand vigorous star formation in halos at least as massive as (3.1, 5.6, 10.5)$\times10^9\msun$ at $z=(6,7,8)$. Likewise, we find that Frontier Fields observations probe down to stellar masses of (8.1, 18, 32)$\times10^6\msun$; that is, they are observing the likely progenitors of analogues to Local Group dwarfs such as Pegasus and M32. Our simulations yield somewhat different constraints than two complementary models that have been invoked in similar analyses, emphasizing the need for further observational constraints on the galaxy-halo connection. | \label{sec:intro} A key question regarding the way in which dark matter halos grow galaxies is the minimum mass of a dark matter halo $\Mmin$ that can both retain its gas and condense it efficiently onto a galaxy. Locally, this question is central to both the ``missing satellites" and the ``too big to fail" problems~\citep{kly99,boy11}. In the context of the reionization epoch ($z\geq6$), it arises because of the possible role of faint galaxies in driving the growth of the nascent ultraviolet ionizing background (UVB): the steep observed slope of the UV luminosity function's (LF) faint end gives rise to a luminosity density that diverges if integrated to arbitrarily faint luminosities. Hence faint galaxies could well have driven reionization and dominated the UVB, but only if the limiting luminosity out to which the LF continues as a power law is quite faint~\citep{rob13,fink12,kuh12b}. As a minimum luminosity would imply a minimum halo mass, measurements of the former can be invoked as a constraint on the latter. This, in turn, would constrain a number of physical processes that can regulate gas cooling, star formation, and feedback in low-mass halos. For example, ``Jeans Filtering" prevents gas from being accreted by dark matter halos whose virial temperature is less than the (appropriately time-averaged) temperature of the ambient intergalactic medium (IGM;~\citealt{efs92,qui96,gne00,oka08}). While the effect is expected to play a role even in the absence of reionization~\citep{nao09}, idealized simulations predict that the latent heat from photoionization completely halts gas accretion in halos with circular velocities $V_\mathrm{circ}$ below $30\kms$ but is negligible for $V_\mathrm{circ}>75\kms$~\citep{tho96}. Similarly, three-dimensional simulations indicate that it suppresses the gas reservoirs of halos less massive than $3\times10^8\msun$ ($V_\mathrm{circ}<26\kms$) at $z=6$~\citep{oka08,fin12}. This idea has motivated extended reionization models in which star formation is assumed not to occur in reionized regions in halos less massive than $1\mbox{--}2\times10^9\msun$~\citep{ili07,alv12,mes08}. Building further on this idea,~\citet{bou10} proposed that gas in halos less massive than $10^{11}\msun$ does not accrete efficiently onto the central galaxy owing to photoionization feedback from hot stars at $z>6$. They showed that this assumption naturally allows a simple model to reproduce measurements of galaxy downsizing at $z\leq2$. Note that these theoretical studies invoke similar physics but assume threshold masses that vary by two orders of magnitude, clearly motivating the need for observational constraints. An additional effect that is related to Jeans suppression is photoevaporation: During the reionization epoch, ionization fronts likely evaporated gas that was bound to minihalos ($< 10^{7\mbox{--}8}\msun$;\citealt{sha04}), suppressing further star formation below this mass range. Even in the absence of a UVB, halos whose virial temperature is less than $10^4$K cannot cool and condense their gas through collisional excitation of neutral hydrogen; they are dependent on molecular hydrogen formation cooling, which is relatively inefficient. These effects have been modeled in high-resolution, ab-initio numerical simulations, leading to predictions that the reionization-epoch LF flattens for luminosities $\MUV < -12$~\citep{wis14}. Another source of suppression is galactic outflows, which are closely associated with vigorous star formation~\citep{vei05}. A variety of theoretical models indicate that outflows are more efficient at removing gas from galaxies and suppressing their growth when they live in low-mass halos~\citep{dek86,hec02,mur05,mura15,chr16}. They may even introduce a characteristic scale below which suppression is particularly efficient~\citep{dek86,mura15}. Finally, recent models in which the local star formation rate density is computed from the local density of molecular hydrogen (as opposed to the total gas density) predict that galaxy growth is suppressed in dark matter halos less massive than $10^{10}\msun$~\citep{kru12,kuh12}. If true, then the $z=6$ UV LF is expected to turn over at an absolute magnitude of $\approx-15$~\citep{jaa13}. Given the importance of the high-redshift UV LF as a probe of activity in low-mass halos, multiple techniques for interpreting it have been developed. One method involves assuming that local group dwarf galaxies are representative of the high-redshift population and using spatially-resolved star formation histories to infer the abundance and activity in their progenitor population.~\citet{wei14b} used this approach to argue that the intrinsic $z>5$ LF grows without a turnover to at least $\MUV=-5$. However, a subsequent analysis found that it may flatten for $\MUV>-13$ (\citealt{boy15}; see also~\citealt{boy14}), in better agreement with predictions from ab-initio simulations~\citep[for example,][]{wis14}. More directly, a number of groups have used imaging from the Hubble Frontier Fields to trace the UV LF out to unprecedented depths~\citep{ate15,ish15,ish16,lap16,liv16,bou16}. The deepest measurements are reported by~\citet{liv16}, who use a wavelet decomposition approach to remove foreground light from galaxies associated with the lensing clusters. This allows them to identify many more faint systems than had previously been detected. They find that the UV LF is inconsistent with a turnover at $M<-12.5$ at $z=6$, with weaker constraints at higher redshifts. What do the~\citet{liv16} measurements imply for galaxy growth in low-mass halos? Qualitatively, the finding that the slopes of the dark matter halo mass function at low masses and the UV LF at faint luminosities are similar ($\approx -2$ in both cases) implies that galaxy luminosity varies nearly linearly with the host halo's mass down to the faintest detected systems. Moreover, evidence for a minimum halo mass below which gas accretion and star formation are inefficient has not yet been detected. Instead, the~\citet{liv16} data place an upper limit on $\Mmin$. This limit constrains the efficiency of physical effects that limit star formation in low-mass halos. Additionally, it informs simplified reionization models that ``paint" luminosities directly onto dark matter halos: they are disfavored if they invoke significant suppression where it is not observed. A simple way to derive this limit is to impose a minimum halo mass cutoff on a galaxy formation model and ask how large that cutoff may be made before it introduces conflict with the observed LF. While similar comparisons have been undertaken before~\citep{mun11,cas16}, the arrival of significantly deeper measurements motivates us to revisit the problem. Additionally, galaxy formation models are now able to address a broader range of observables simultaneously than before. In particular,~\citet{fin16} discussed a numerical hydrodynamic + continuum radiation transport model that reproduces both the observed UV LF and the abundance of low-ionization metal absorbers~\citep{fin16}. Moreover, it predicts an integrated optical depth to Thomson scattering of 0.059, in excellent agreement with the most recent constraints from the cosmic microwave background~\citep{pla16a,pla16b}. In short, this model yields favorable agreement with observations of galaxies, absorbers, and a spatially-inhomogeneous reionization history simultaneously, opening up the possibility of understanding how these observables relate to one another. It is therefore a particularly well-tested framework for interpreting the observed LF. In this work, we use this simulation to interpret the constraints reported by~\citet{liv16} as an upper limit on $\Mmin$ at $z>6$--8, and on the lowest stellar mass of the galaxies that have been observed at this epoch. In Section~\ref{sec:sims}, we review our simulations. In Section~\ref{sec:res}, we present our results. In Section~\ref{sec:disc}, we discuss their implications and compare to previous work. In Section~\ref{sec:sum}, we summarize. | \label{sec:disc} The purpose of this study is, in essence, to invoke the relation shown in Figure~\ref{fig:MUVMhalo} (and its analogues at higher redshifts) in order to constrain $\Mmin$ from the observed UV LF. Our analysis shows that current observations already exert pressure on some previous treatments for star formation. For example, we find that star formation must be relatively unsuppressed in halos down to $3.1\times10^9\msun$ at $z=6$, whereas the fiducial metallicity-based H$_2$ model of~\citet{kru12} predicts $\approx80\%$ suppression at this mass scale (their Figure 8). Our results are also in conflict with the accretion-floor model of~\citet{bou10}: if halos below $10^{11}\msun$ do not form stars, then galaxies must populate halos in a very different way than arises in hydrodynamic simulations (that is, Figure~\ref{fig:MUVMhalo}) in order to match the~\citet{liv16} measurements. Of course, the~\citet{bou10} model was forwarded as an interpretation of observations at $z\leq6$, hence a more convincing test would be to repeat our analysis using the observed UV LF at lower redshifts. For the present, we therefore limit ourselves to the conclusion that an accretion floor at $10^{11}\msun$ does not apply at $z>6$. This agrees with~\citet{beh13}, whose analysis of the stellar mass - halo mass relation likewise indicates robust star formation in $10^{10}\msun$ halos out to $z=8$. An obvious improvement over our study would be to \emph{test} the prediction presented in Figure~\ref{fig:MUVMhalo}. In order to motivate the need for such a test, we compare our analysis to two previous efforts.~\citet{mun11} constructed a semi-analytical model and used it to analyze the shallower observations that were available at the time ($\MUV \leq -18$). In their entirely complementary model, halos are assumed to condense their gas into a star-forming disk whenever there is a merger, after which the condensed gas forms into stars. They fitted simultaneously for the luminosity amplitude $L_{10}$ (that is, the luminosity $\log(L_{1500}/\erg\rm{\ s}^{-1} \Hz^{-1})$ within $10^{10}\msun$ halos), and for $\Mmin$ ($m_\mathrm{supp}$ in their notation), and found that $\Mmin\leq10^{9.5}\msun$ at $z=6$ (we will compare with their results only at $z=6$; similar results occur at $z=7$ and $z=8$). In other words, for an observed luminosity function that probes $100\times$ shallower than~\citet{liv16}, they derived roughly the same $\Mmin$ as we do. It is reasonable to assume that, were they to confront their model with the most recent constraints, the inferred $\Mmin$ would be much lower than ours. This discrepancy reveals a very significant theoretical uncertainty. The fact that both models match the observed LF's normalization despite this remarkable discrepancy owes to a cancellation between two effects. First, their derived normalization $L_{10}=27.2$ is nearly $10\times$ larger than our model, which predicts $L_{10}=26.3$ (Figure~\ref{fig:MUVMhalo}), corresponding to a much lower star formation efficiency. Second, their model predicts highly bursty star formation histories: For their fiducial model, the ``active fraction" of $10^{10}\msun$ halos is $<50\%$. Additionally, their model assumes that star formation in satellite halos stops whenever there is a major merger. This contrasts with our model, in which star formation is generally smooth~\citep{fin11} and satellites constitute a significant fraction of the observable population (Figure~\ref{fig:MUVMhalo}). Together, these differences increase the predicted characteristic luminosity at a given halo mass with respect to our model. More recently,~\citet{cas16} (see also~\citealt{yue14,yue16}) confronted a different semi-analytic model with Frontier Fields measurements that probed down to $\MUV=-15$ and inferred that the threshold circular velocity must be below $50\kms$, corresponding to a halo mass threshold of $\log(M_c/\msun)=9.3$. Given that their input LF is 2--3 magnitudes shallower than ours, the result that the inferred halo mass threshold is similar once again implies a discrepancy in the underlying physics. It is likely that, as before, tradeoffs between the unknown burstiness and star formation efficiency of high-redshift galaxies are to blame. The need for observations that can distinguish between these models is clear. For example, \emph{JWST} will enable measurements of the relationship between complementary observables such as UV luminosity, continuum slope, stellar mass, and H$\alpha$ luminosity, which will probe the underlying level of burstiness. In the nearer term, comparison with clustering measurements may already be able to distinguish between the models; however, this is beyond the scope of the current study. For the present, it is most noteworthy that, in all three analyses, observations forbid star formation to be suppressed in halos above $\log(M_c/\msun)=9.5$. The uncertainties inherent in using galaxy formation models to interpret the UV LF may be mitigated by invoking observations of long gamma-ray bursts (LGRBs), which are also believed to trace star formation.~~\citet{wei16} recently used a simple model that ties galaxy growth and LGRBs to the evolving dark matter halo mass function to derive $\Mmin$ from \emph{Swift} observations. At $1\sigma$ confidence, they constrain it to be less than $10^{10.5}\msun$ at $z<4$ and in the range of $10^{7.7}$--$10^{11.6}\msun$ at $4 < z < 5$. These results are consistent with ours, although they apply to lower redshifts. Finally, we note that there is room for progress in reducing both observational and theoretical uncertainties. On the observational side, the luminosity of the faintest \emph{Hubble Frontier Fields} sources remain uncertain owing to the challenge of measuring faint sources in lensed fields. Additionally, the unknown intrinsic sizes of the lensed sources introduces uncertainty into the observational incompleteness corrections that are needed to compute volume densities. The effect is particularly dramatic for the faintest luminosities~\citep{bou16}. On the theoretical side, resolution limitations may affect our analysis: with our cosmology and dynamic range, the minimum stellar mass to which~\citet{liv16} probe ($6\times10^6\msun$; Figure~\ref{fig:BICstar}) corresponds to $\approx52$ star particles, whereas we have previously argued that 64 are required for converged predictions of global galaxy properties such as mass and luminosity~\citep{fin06}. The host halos, by contrast, are quite well-resolved: the minimum halo mass of $3.1\times10^9\msun$ corresponds to 7100 dark matter particles, which is more than sufficient to resolve both the halo's mass and internal structure~\citep{tre10} as well as its gas accretion history~\citep{nao09}. Hence while we do not believe that resolution limitations are severe, at present they limit our ability to comment in more detail on the nature of the faintest currently-observed galaxies. | 16 | 9 | 1609.06348 |
1609 | 1609.08449_arXiv.txt | Be stars have generally been characterized by the emission lines in their spectra, and especially the time variability of those spectroscopic features. They are known to also exhibit photometric variability at multiple timescales, but have not been broadly compared and analyzed by that behavior. We have taken advantage of the advent of wide-field, long-baseline, and high-cadence photometric surveys that search for transiting exoplanets to perform a comprehensive analysis of brightness variations among a large number of known Be stars. The photometric data comes from the KELT transit survey, with a typical cadence of 30 minutes, baseline of up to ten years, photometric precision of about 1\%, and coverage of about 60\% of the sky. We analyze KELT light curves of 610 known Be stars in both the Northern and Southern hemispheres in an effort to study their variability. Consistent with other studies of Be star variability, we find most of the stars to be photometrically variable. We derive lower limits on the fraction of stars in our sample that exhibit features consistent with non-radial pulsations (25$\%$), outbursts (36$\%$), and long term trends in the circumstellar disk (37$\%$), and show how these are correlated with spectral sub-type. Other types of variability, such as those owing to binarity, are also explored. Simultaneous spectroscopy for some of these systems from the Be Star Spectral Database (BeSS) allow us to better understand the physical causes for the observed variability, especially in cases of outbursts and changes in the disk. | Classical Be stars are a rapidly rotating subset of non-giant B-type stars. Unlike normal B-type stars, Be stars exhibit line emission (hence the `e' in Be), which is attributed to a gaseous circumstellar disk in Keplerian motion. The circumstellar disk of a Be star is best described by the viscous decretion disk model \citep[e.g.][]{Lee1991,Carciofi2011}, where the disk is formed and subsequently sustained by mass ejected from the stellar surface in discrete events called `outbursts' (e.g. \citet{Kroll1997} or \citet{Kee2014} for a theoretical framework, or \citet{Rivinius1998} or \citet{Grundstrom2011} for observations). This ejected material then orbits the star, settling into a disk. However, if the flow of material from star to disk stops, the disk will dissipate through viscous forces, which facilitate the transfer of angular momentum to the outer disk (simultaneous with a loss of angular momentum in the inner disk), causing the disk to clear from the inside outward \citep{Haubois2012}, or alternatively via line-driven ablation \citep{Kee2016}, which also would clear the disk from the inside outward. The details of the physical mechanism which launches stellar material into orbit remains elusive, but very rapid (near critical) stellar rotation combined with non-radial pulsations (NRP) is theorized to play an important role in the mass transfer mechanism \citep{Rivinius2013}. The physical changes taking place within Be stars and their disks systems leave imprints in various observables that are accessible through spectroscopy, photometry, polarimetry, and interferometry, allowing the underlying mechanisms of variability to be studied through a variety of techniques. Observations of Be stars during both both variable and non-variable phases provide insight into the physical mechanisms taking place, helping work towards a more thorough understanding of the `Be phenomenon'. Here we present long time-baseline, high-precision photometric observations of large numbers (hundreds) of Galactic Be stars, substantially increasing our knowledge of the photometric variability of this population. Be stars are known to show variability in their brightness and spectral features across a large range of timescales from hours to decades. When variability in a Be star is observed, the associated timescales often give insight into the physical cause of these changes. Periodicity on shorter timescales of hours to days are typically attributed to stellar NRP. Outbursts and quasi-periodic oscillations are typically found on intermediate timescales of days to months, although outbursts occasionally have durations of years. The longest timescales typically involve changes in the disk, the most dramatic of which is the total disappearance (or reappearance) of the central star's circumstellar disk. It is not uncommon for these disks to appear and disappear over the course of years to decades \citep{McSwain2009}, and as such an object classified as a Be star must have shown emission at some point in time, but does not necessarily show emission (or posses a disk) in the current epoch. Other changes in the circumstellar environment and the disk can manifest themselves in a variety of observational ways. NRP are commonly observed in Be stars, with typical timescales between $\lesssim$0.1 day to 2 days. \citet{Cuypers1989} detect NRP in $\sim$82\% of a sample of 17 Be stars. In a sample of 57 Be stars, \citet{Gutierrez-Soto2007} detect short-term variability indicative of NRP in 74$\%$ of early-type Be stars, and in 31$\%$ of mid- and late-type Be stars. The photometric amplitudes associated with NRP in Be stars can be quite low, down to the sub-mmag level \citep{Emilio2010, Walker2005, Walker2005a, Saio2007}. This is further complicated by the fact that pulsational frequencies and amplitudes can shift, especially when the star also undergoes outbursts \citep{Hubert2007,Huat2009}. Although signatures of these pulsations can be very difficult to detect, all Be stars that have been analyzed with high-cadence, long-duration space-based photometry have been reported to be multiperiodic and to pulsate, with amplitudes decreasing with later spectral subtypes \citep{Rivinius2013}. It therefore seems that, as a class of objects, Be stars are pulsators. Periodic variability has been detected in Be stars with periods longer than what can be explained by single NRP modes. Cyclical variability between 60 - 100 days was detected in the Be star $\delta$ Scorpii \citep{Jones2013}. \citet{Sterken1996} find periodic and quasi-periodic oscillations (QPO) in brightness in 4 Be stars with periods ranging between 4 and 93 days. In a similar analysis, \citet{Mennickent1994} detect QPO in two stars (27 CMa and 28 CMa) with periods between 10 and 20 days. \citet{Hubert1998} make use of Hipparcos photometry \citep{VanLeeuwen1997} and find QPO with a period of 11.546 days in the Be star MX Pup. One possible explanation or this type of periodic variability is the beating of two or more NRP modes with closely spaced frequencies. This is an important topic of study, since it appears that the beating of multiple NRP modes can trigger outbursts \citep{Rivinius2001, Rivinius2013}. There are, however, a multitude of other possible explanations for this observed periodicity. Outbursts are generally understood as discrete events where material is transferred from the stellar surface to the inner region of the circumstellar disk, where it is then governed by gravity and viscosity. In visible photometric observations, outbursts are typically characterized by a `sudden' change in flux of the system, followed by a more gradual (relative to the initial change) decay back to baseline. Disk emission and polarization rise steeply during an outburst event \citep{Rivinius2013}, signifying an increasing density in the innermost regions of the disk, but the net change in brightness depends on the inclination angle of the system and can be positive or negative. If viewed edge-on or very nearly edge-on, the disk will obscure the central star and the system will appear dimmer as the disk grows, but at low to moderate inclination angles (e.g. $\sim$pole-on) the system will appear brighter with disk growth \citep{Sigut2013}. Models described in \citet{Sigut2013} where a disk grows over the course of a year and then dissipates over the course of two years show changes in V-band magnitude of a few tenths of a magnitude throughout this process, although this can vary depending on orientation, gravitational darkening due to rapid rotation, spectral sub-type, disk density, scale height, and the details of disk growth and evolution. \citet{Haubois2012} predict V-band brightening of up to 0.4 mag ($i$ = 0$^{\circ}$), and dimming of 0.2 mag ($i$ = 90$^{\circ}$) during disk growth/dissipation phases using the viscous decretion disk model. At an inclination angle of $i\sim$70$^{\circ}$, the net change in optical flux during an outburst is predicted to be nearly zero, because the additional emission and absorption introduced by the disk effectively cancel each other out. Although outbursts are commonly seen in classical Be stars, their frequency, duration, and amplitude vary greatly from star to star, and a given Be star can show large variation in its outbursts over time. It is therefore necessary to amass a large number of observed outbursts for as many systems as possible, in order to better understand their systematic behavior and possible correlations with the underlying stellar properties. Some Be stars retain their disks over many years or decades, and in these cases it is not uncommon for disks to posses density waves, which can then travel around the disk on timescales hundreds of times longer than that of Keplarian motion at a given radius from the central star, typically at a period on the order of $\sim$10 years \citep{Okazaki1991,Papaloizou1992}. In spectroscopic observations, these global oscillation modes in the disk cause variations in the ratio of the violet-to-red (V/R) peaks typical of Be star emission lines \citep{Rivinius2013}. When the high density material is approaching the observer, the violet (V) peak is enhanced, and when the high density part is moving away from the observer the red (R) peak is enhanced. Density waves can also produce photometric variations as a line-of-sight effect, which largely depends on the inclination angle of the system. Classical Be stars are a very heterogeneous class of objects, and a given Be star may show all, some, or none of the aforementioned types of variability over some observing baseline. To better understand Be stars as a class of objects, it is necessary to observe a large number of Be stars for a long time, with as many techniques as possible. This strategy will give a more complete and probabilistic picture of the characteristics of Be stars as a population. This is a necessary step, since results from a small number of well-understood Be stars can not accurately be extrapolated to the entire population. Any theoretical models describing Be stars and their disks must accommodate the severe inhomogeneity seen amongst Be stars as a population, and can not be specifically tailored to just a few well understood cases. In this paper, we explore the variability of over 500 known Galactic Be stars, primarily through the use of time-series photometric data. In \S \ref{sec:data} we introduce the data products that make up the core of our analysis. \S \ref{sec:analysis} is a discussion of the various methods used to analyze the data, describing how each type of signal is recovered from a KELT light curve. In \S \ref{sec:results} we present our results. General patterns and trends are explored, and specific interesting cases are highlighted. In some cases, archival time-series spectroscopic data are included in our analysis, providing a more comprehensive view of such systems. In \S \ref{sec:conclusion}, we summarize the ensemble properties of our sample and discuss some highlights. Finally, the appendix includes plots for every variable object in our sample, as well as a brief discussion of each eclipsing system that we found. | \label{sec:conclusion} \begin{table*} \centering \caption{Fractions showing variability} \label{tbl:variable_fractions} \begin{tabular}{|l|l|l|l|l|l|} \hline \hline \textit{Variable Type} & \textit{Description of Variability} & \textit{All} & \textit{Early} & \textit{Mid + Late} & \textit{Unclassified}\\ \hline Outburst Variation (ObV) & One or more outburst & 36$\%$ (\(\nicefrac{168}{470}\))& 51$\%$ (\(\nicefrac{135}{265}\)) & 12$\%$ (\(\nicefrac{16}{139}\)) & 26$\%$ (\(\nicefrac{17}{66}\)) \\ \hline Semi-Regular Outbursts (SRO) & Outbursts occurring at regular intervals & 21$\%$ (\(\nicefrac{35}{168}\)) & 22$\%$ (\(\nicefrac{29}{135}\)) & 19$\%$ (\(\nicefrac{3}{16}\)) & 18$\%$ (\(\nicefrac{3}{17}\)) \\ \hline Long-Term Variation (LTV) & Variability on timescales of years & 37$\%$ (\(\nicefrac{81}{217}\)) & 45$\%$ (\(\nicefrac{54}{121}\)) & 21$\%$ (\(\nicefrac{15}{73}\)) & 52$\%$ (\(\nicefrac{12}{23}\))\\ \hline Non-Radial Pulsations (NRP) & Periodic variation, $P$ $\leq$ 2 days & 25$\%$ (\(\nicefrac{125}{510}\))& 28$\%$ (\(\nicefrac{80}{287}\))& 21$\%$ (\(\nicefrac{32}{155}\)) & 19$\%$ (\(\nicefrac{13}{68}\)) \\ \hline Intermediate Periodicity (IP) & Periodic variation, 2 $<$ $P$ $\leq$ 200 days & 39$\%$ (\(\nicefrac{201}{510}\)) & 52$\%$ (\(\nicefrac{150}{287}\)) & 21$\%$ (\(\nicefrac{32}{155}\)) & 28$\%$ (\(\nicefrac{19}{68}\)) \\ \hline Eclipsing Binaries (EB) & Eclipsing systems & 2.7$\%$ (\(\nicefrac{14}{510}\)) & 2.8$\%$ (\(\nicefrac{8}{287}\)) & 1.9$\%$ (\(\nicefrac{3}{155}\)) & 4.4$\%$ (\(\nicefrac{3}{68}\)) \\ \hline \hline \end{tabular} \begin{flushleft} \footnotesize \footnotesize Fraction of stars showing each type of variability, according to their spectral type. The category `all' includes early-, mid-, and late-type stars, as well as those unclassified in BeSS. The fraction of systems with SROs is calculated from the subset of stars showing at least one outburst, and EBs from all non-saturated objects. See Section~\ref{sec:results} for an explanation of how the other fractions were calculated. \end{flushleft} \end{table*} \vspace{2mm} \begin{figure}[!ht] \centering\epsfig{file=figure11,clip=,width=0.99\linewidth} \caption{Histogram showing the distribution of all periodic signals found in objects within the BeSS-KELT sample on a logarithmic scale spanning between 0.1 - 200 days, displayed in the same manner as in Figure~\ref{fig:Outburst_rates_early}. The vertical dotted line at 2 days marks the cutoff between short periods and intermediate periods. The short-period variables are left of the dotted line, and are best explained by NRP modes in the star. All types of periodic variability (NRP, IP, SRO, and EB) are included in this histogram. A single star may have multiple periods at different timescales and thus may appear in up to three different bins, although the majority of periodically variable stars have a single dominant frequency.} \label{fig:hist_all_periodic} \end{figure} The fractions of stars in this sample showing each type of variability mentioned in Table~\ref{tbl:variable_class} are summarized in Table~\ref{tbl:variable_fractions}. There are many cases where a star exhibits both non-periodic (e.g. outbursts) and periodic variability (e.g. pulsations), and it is important to note that these categories are not mutually exclusive. A histogram showing the distribution of all recovered periods is shown in Figure~\ref{fig:hist_all_periodic}. This histogram includes all types of periodic variability (NRP, IP, SRO, and EB variables), and is complicated by the fact that a single star can exhibit more than one period in its light curve. Because of the diurnal sampling of the KELT survey, periodic signals very close to one day are poorly sampled. The dearth of detected periods very near to one day is a result of this systematic effect. From analyzing the KELT light curves of this sample of Be stars, we arrive at a few important conclusions. Consistent with other studies \citep[e.g.][]{Cuypers1989,Gutierrez-Soto2008,McSwain2009,Chojnowski2015}, we find that Be stars are a highly variable class of objects, with a greater degree of variability seen in earlier spectral types. About 1/5 of Be stars with clear photometric outbursts have them occurring at semi-regular intervals. Intermediate periodicity (longward of 2 days) is a common occurrence, and is seen in 41$\%$ of our sample. By combining KELT data with BeSS spectra, we provide evidence that photometric outbursts correspond to disk creation or disk building events, and that global disk oscillations manifesting in V/R variability can also modulate the brightness of a Be star + disk system. This work is unique in its large sample size of Galactic Be stars and long temporal coverage, both of which will continue to grow as new KELT data is collected and reduced. Future work will involve increasing our statistics with larger sample sizes and baselines, as well as more detailed investigations of particularly interesting systems and a more thorough treatment of the types of variability discussed here. | 16 | 9 | 1609.08449 |
1609 | 1609.09215_arXiv.txt | {Be/X-ray binaries represent the main group of high-mass X-ray binaries. The determination of the astrophysical parameters of the counterparts of these high-energy sources is important for the study of X-ray binary populations in our Galaxy. X-ray observations suggest that SAX\,J2239.3+6116 is a Be/X-ray binary. However, little is known about the astrophysical parameters of its massive companion. } {The main goal of this work is to perform a detailed study of the optical variability of the Be/X-ray binary SAX\,J2239.3+6116. } {We obtained multi-colour $BVRI$ photometry and polarimetry and 4000-7000 \AA\ spectroscopy. The 4000--5000 \AA\ spectra allowed us to determine the spectral type and projected rotational velocity of the optical companion; the 6000-7000 \AA\ spectra, together with the photometric magnitudes, were used to derive the colour excess $E(B-V)$, estimate the distance, and to study the variability of the H$\alpha$ line. } {The optical counterpart to SAX J2239.3+6116 is a $V=14.8$ B0Ve star located at a distance of $\sim$4.9 kpc. The interstellar reddening in the direction of the source is $E(B-V)=1.70\pm0.03$ mag. The monitoring of the H$\alpha$ line reveals a slow long-term decline of its equivalent width since 2001. The line profile is characterized by a stable double-peak profile with no indication of large-scale distortions. We measured intrinsic optical polarization for the first time. Although somewhat higher than predicted by the models, the optical polarization is consistent with electron scattering in the circumstellar disk. } {We attribute the long-term decrease in the intensity of the H$\alpha$ line to the dissipation of the circumstellar disk of the Be star. The longer variability timescales observed in SAX\,J2239.3+6116 compared to other Be/X-ray binaries may be explained by the wide orbit of the system.} | \sax\ was discovered by the {\em BeppoSAX} wide-field camera as a transient X-ray source during observations of the supernova remnant Cas A. It was first detected on 4 March 1997 and then again on 8 May 1999 \citep{intzand00}. The peak flux then was $3.3 \times 10^{-10}$ erg s$^{-1}$ cm$^{-2}$ in the energy range 2--10 keV and $1.0 \times 10^{-9}$ erg s$^{-1}$ cm$^{-2}$ in the energy range 2--26 keV. The X-ray spectral continuum was satisfactorily fitted with a single absorbed power law with $N_H=1\times 10^{22}$ cm$^{-2}$ and $\Gamma=1.1\pm0.1$. The source was not detected on 13--15 December 1999 with an upper limit on the 2--10 keV X-ray flux of $6 \times 10^{-11}$ erg s$^{-1}$ cm$^{-2}$ \citep{intzand00}. A search for detections with other X-ray instruments resulted in a detection with {\em BeppoSAX}/MECS on 24--25 November 1998 with a flux of $5 \times 10^{-13}$ erg s$^{-1}$ cm$^{-2}$ (2--10 keV) and another detection with {\em CGRO}/BATSE in March 1997 with a flux $1.4 \times 10^{-9}$ erg s$^{-1}$ cm$^{-2}$ (20--100 keV). The {\em RXTE}/ASM light curve showed increases in the X-ray intensity at regular interval times of $262\pm5$ days \citep{intzand00}. If this periodicity is interpreted as the orbital period of the system, then \sax\ has the longest orbital period of all the known BeXB in our Galaxy. Subsequent {\em RXTE}/PCA and {\em BeppoSAX}/MECS-LECS observations during a predicted outburst in July 2001 revealed X-ray pulsations with a pulse period of $1247.2\pm0.7$ s \citep{intzand01}. The X-ray flux during the observations that detected pulsations was $4 \times 10^{-12}$ erg s$^{-1}$ cm$^{-2}$ (2--10 keV). Optical observations were carried out on 2--3 December 1999 with the 2.1 m telescope of the Kitt Peak National Observatory. A B-type star showing \ha\ in emission with an equivalent width of --6.7 \AA\ was discovered $0.3\arcmin$ away from the best-fit X-ray position. This V=15.1 mag star was proposed to be the optical counterpart to \sax\ \citep{intzand00}. One more optical campaign was reported by \citet{riquelme12} on 2--5 July 2001 with the following photometric magnitudes: $U=16.19$, $B=16.06$, $V=14.55$, $R=13.60$, and $I=12.74,$ and an \ha\ equivalent width of --11.0 \AA. In this work, we performed the first detailed study of the optical variability of \sax. We present photometric observations covering the period 2007-2016, spectra in the region of the \ha\ line from 2001-2016, and for the first time polarimetric data from 2013-2016. Although photometric and spectroscopic data prior to 2015 have been presented in the context of a global study of the long-term optical variability of BeXBs by \citet{reig15} and \citet{reig16}, respectively, we include them here for the sake of completeness. Our polarimetric observations are the first dedicated observations of the source using this technique. | We have performed optical photometric, spectroscopic, and polarimetric observations of the optical counterpart to \sax. We derived a spectral type B0Ve from the ratios of various metallic lines and we estimated the rotation velocity of the underlying B star in 200 km s$^{-1}$ from the width of hydrogen and He I lines. We estimated the distance to be $\sim$ 4.9 kpc from the photometric magnitudes and colours and the strength of various diffuse interstellar bands. We report, for the first time, intrinsic optical linear polarization from the circumstellar disk of the Be star companion at a level of $4\%$. The long-term optical variability of this system is characterized by the slow dissipation of the circumstellar disk around the Be star companion. The observational consequence of this decline is the decrease of the optical brightness, the strength of the \ha\ line, and the polarization degree. We argue that the long variability timescales observed in this system are due to the relatively weak gravitational pull exerted by the neutron as a consequence of its wide orbit. | 16 | 9 | 1609.09215 |
1609 | 1609.09165_arXiv.txt | The thermonuclear rate of the $^{50}$Fe($p$,$\gamma$)$^{51}$Co reaction in the Type I X-ray bursts (XRBs) temperature range has been reevaluated based on a recent precise mass measurement at CSRe lanzhou, where the proton separation energy $S_p$=142$\pm$77 keV has been determined firstly for the $^{51}$Co nucleus. Comparing to the previous theoretical predictions, the experimental $S_p$ value has much smaller uncertainty. Based on the nuclear shell model and mirror nuclear structure information, we have calculated two sets of thermonuclear rates for the $^{50}$Fe($p$,$\gamma$)$^{51}$Co reaction by utilizing the experimental $S_p$ value. It shows that the statistical-model calculations are not ideally applicable for this reaction primarily because of the low density of low-lying excited states in $^{51}$Co. In this work, we recommend that a set of new reaction rate based on the mirror structure of $^{51}$Cr should be incorporated in the future astrophysical network calculations. | Type I X-ray bursts (XRBs) arise from thermonuclear runaways on the accreted envelopes of neutron stars in close binary systems~\cite{bib:woo76,bib:jos77}. During the thermonuclear runaway, an accreted envelope enriched in H and He may be transformed to matter strongly enriched in heavier species (up to A$\sim$100~\cite{bib:sch01,bib:elo09}) via the rapid proton capture process (rp-process)~\cite{bib:wal81,bib:sch98,bib:woo04}. Please see, e.g., Refs.~\cite{bib:lew93,bib:str06,bib:par13} for reviews on the XRBs. The rp-process is largely characterized by localized $({\rm p},\gamma)$-$(\gamma,{\rm p})$ equilibrium within particular isotonic chains near the proton drip-line. In such an equilibrium situation the abundance distribution within an isotonic chain depends exponentially on nuclear mass differences as the abundance ratio between two neighboring isotones is proportional to $\exp[S_p/kT]$ ($S_p$: proton separation energy, $T$: temperature of the stellar environment). In particular, those isotonic chains with sufficiently small $S_p$ values (relative to XRB temperatures - at 1 GK, $kT$$\approx$100 keV) need to be known with a precision of at least 50$\sim$100 keV~\cite{bib:sch98,bib:par09}. In order to compare model predictions with observations of the light curves~\cite{bib:sch06}, reliable nuclear-physics inputs, e.g., precise $S_p$ values and nuclear structure information, are needed for those nuclei along the rp-process path. Recently, precise mass measurements of nuclei along the rp-process path have become available. These measurements were made at the HIRFL-CSR (Cooler-Storage Ring at the Heavy Ion Research Facility in Lanzhou)~\cite{bib:xia02} in an IMS (Isochronous Mass Spectrometry) mode. The proton separation energy of $^{51}$Co has been experimentally determined to be $S_p^\mathrm{IMP}$=142$\pm$77 keV for the first time~\cite{bib:shu14}. Although the estimated values in the previous Atomic Mass Evaluations (i.e., $S_p$=240$\pm$210 keV in AME85~\cite{bib:aud85}, 290$\pm$160 keV in AME93~\cite{bib:aud93}, 90$\pm$160 keV in both AME95~\cite{bib:aud95} and AME03~\cite{bib:aud03}) agree with this experimental value within 1~$\sigma$ uncertainty, the experimental value is significantly more precise. Previously, the impact of $Q$-value (i.e., $S_p$ value) for the $^{50}$Fe($p$,$\gamma$)$^{51}$Co reaction was studied~\cite{bib:par09} based on the old $S_p$($^{51}$Co) value of AME03. In the XRB `short' model, it shows that the uncertainty of $Q$-value has very large impact on the final yields of $^{51}$Cr and $^{52}$Fe, whose yields can be significantly affected by factors of 4.9 and 2.0, respectively, by changing $Q$ to $Q+\Delta Q$. Therefore, a precise $S_p$ ($^{51}$Co) value is very important for constraining the final XRB yields. In this work, we have derived the thermonuclear rates of $^{50}$Fe($p$,$\gamma$)$^{51}$Co based on the new experimental $S_p$ ($^{51}$Co) value, with which the resonant and direct capture (DC) rates are recalculated. This precise $S_p$ value allows the uncertainty in the rate of the $^{50}$Fe($p$,$\gamma$)$^{51}$Co reaction to be dramatically reduced (e.g., see Ref.~\cite{bib:hjj14}), and hence the XRB yields can be well constrained too. | The thermonuclear rate (including direct-capture (DC) and resonant contribution) of the $^{50}$Fe($p$,$\gamma$)$^{51}$Co reaction has been recalculated by utilizing the recent precise proton separation energy of $S_p$($^{51}$Co)=142$\pm$77 keV measured at the HIRFL-CSR facility in Lanzhou, China. Here, the resonant rates have been calculated in two ways: one is to revise the previous shell-model results with this new $S_p$ value (i.e., \emph{Shell} rate), another is to rely on the mirror nuclear structure of $^{51}$Cr (i.e., \emph{Mirror} rate). Our new rates deviate significantly from those available in the literature. We conclude that statistical model calculations are not ideally applicable for this reaction primarily because of the low density of low-lying excited states in $^{51}$Co. Thus, we recommend that the present new \emph{Mirror} rate should be incorporated in the future astrophysical network calculations, since it is based on more solid experimental background. The astrophysical impact of our new rates in Type I x-ray burst calculations is now under progress, which is beyond the scope of this work. \end{multicols} \vspace{2mm} \centerline{\rule{80mm}{0.1pt}} \vspace{2mm} \begin{multicols}{2} | 16 | 9 | 1609.09165 |
1609 | 1609.02864_arXiv.txt | {Hybrid $\delta$ Scuti-$\gamma$ Doradus pulsating stars show acoustic ($p$) oscillation modes typical of $\delta$ Scuti variable stars, and gravity ($g$) pulsation modes characteristic of $\gamma$ Doradus variable stars simultaneously excited. Observations from space missions like MOST, CoRoT, and \emph{Kepler} have revealed a large number of hybrid $\delta$ Scuti-$\gamma$ Doradus pulsators, thus paving the way for a exciting new channel for asteroseismic studies.} {We perform a detailed asteroseismological modeling of five hybrid $\delta$ Scuti-$\gamma$ Doradus stars.}{We employ a grid-based modeling approach to sound the internal structure of the target stars by employing a huge grid of stellar models from the zero-age main sequence to the terminal-age main sequence, varying parameters like stellar mass, effective temperature, metallicity and core overshooting. We compute their adiabatic radial ($\ell= 0$) and non-radial ($\ell= 1, 2, 3$) $p$ and $g$ mode periods. We employ two model-fitting procedures to searching for the models that best reproduce the observed pulsation spectra of each target star, that is, the asteroseismological models.}{We derive the fundamental parameters and the evolutionary status of five hybrid $\delta$ Scuti-$\gamma$ Doradus variable stars recently observed with the CoRoT and \emph{Kepler} space missions: CoRoT 105733033, CoRoT 100866999, KIC 11145123, KIC 9244992, and HD 49434. The asteroseismological model for each star results from different criteria of model selection, in which we take full advantage of the richness of periods that characterizes the pulsation spectra of this kind of stars.}{} | \label{introduction} At the present time, pulsating stars constitute one of the most powerful tools for sounding the stellar interiors and to derive a wealth of information about the physical structure and evolutionary status of stars through asteroseismology \citep{2010aste.book.....A, 2010csp..book.....B,2015pust.book.....C}. Nowadays, a huge number of variable stars are routinely discovered and scrutinized by space missions like MOST \citep{2003PASP..115.1023W}, CoRoT \citep{2009IAUS..253...71B} and \emph{Kepler} \citep{2010ApJ...713L..79K}, which include long-term monitoring with high-temporal resolution and high-photometric sensibility for hundreds of thousands stars. Among the most intensively studied classes of variable stars in recent years we found the $\delta$ Scuti (Sct) and $\gamma$ Doradus (Dor), which are $\sim 1.2-2.2 M_{\sun}$ stars with spectral types between A and F, undergoing quiescent core H burning at (or near of) the Main Sequence (MS) ($ 6500\ {\rm K} \lesssim T_{\rm eff} \lesssim 8500$ K). They exhibit multiperiodic brightness variations due to global radial and non-radial pulsation modes. The $\delta$ Sct stars, discovered over a century ago \citep{1900ApJ....12..254C}, display high-frequency variations with typical periods in the range $\sim 0.008\ {\rm d} - 0.42$ d and amplitudes from milli-magnitudes up to almost one magnitude in blue bands. They are likely produced by nonradial $p$ modes of low radial order $n$ and low harmonic degree ($\ell= 1-3$), although the largest amplitude variations probably are induced by the radial fundamental mode ($n= 0, \ell= 0$) and/or low-overtone radial modes ($n= 1,2,3, \cdots, \ell= 0$). The fact that $\delta$ Sct stars pulsate in nonradial $p$ modes and radial modes implies that they are potentially useful for probing the stellar envelope. The $\delta$ Sct variables are Population I stars\footnote{There exist a Population II counterpart to $\delta$ Sct variables, the so called SX Phoenicis (Phe), which are usually observed in low-metallicity globular clusters \citep[see, e.g.,][]{2014RMxAA..50..307A}.} of spectral type between A0 and F5, lying on the extension of the Cepheid instability strip towards low luminosities, at effective temperatures between 7000 K and 8500 K, stellar masses in the interval $1.5-2.2 M_{\sun}$, and luminosities in the range $5 \lesssim L/L_{\sun} \lesssim 80$ \citep{2015pust.book.....C}. The projected rotation velocities are in the range $[0,150]$ km s$^{-1}$, although they can reach values up to $\sim 250$ km s$^{-1}$. The pulsations are thought to be driven by the $\kappa$ mechanism \citep{1980tsp..book.....C,1989nos..book.....U} operating in the partial ionization zone of He II \citep{1971A&A....14...24C,2004A&A...414L..17D,2005A&A...434.1055G}. Solar-like oscillations stochastically driven have also been predicted to occur in $\delta$ Sct stars \citep{2002A&A...395..563S}. Notably, these expectations have been confirmed in one object \citep{2011Natur.477..570A}. Among $\delta$ Sct stars, frequently a distinction is made between the so called \emph{high-amplitude $\delta$ Sct stars} (HADS), whose amplitudes in the V band exceed 0.3 mag, and their much more abundant \emph{low-amplitude $\delta$ Sct stars} (LADS) counterparts \citep[see][]{2008PASJ...60..551L}. The $\gamma$ Dor variables \citep{1999PASP..111..840K}, were recognized as a new class of pulsating stars about 20 years ago \citep{1994MNRAS.270..905B}. They are generally cooler than $\delta$ Sct stars, with $T_{\rm eff}$ between 6700 K and 7400 K (spectral types between A7 and F5) and masses in the range $1.5-1.8 M_{\sun}$ \citep{2015pust.book.....C}. The $\gamma$ Dor stars pulsate in low-degree, high-order $g$ modes driven by a flux modulation mechanism (``convective blocking'') induced by the outer convective zone \citep{2000ApJ...542L..57G, 2004A&A...414L..17D, 2005A&A...434.1055G}. The low-frequency variations shown by these stars have periods typically between $\sim 0.3$ d and $\sim 3$ d and amplitudes below $\sim 0.1$ magnitudes. The presence of $g$ modes in $\gamma$ Dor stars offers the chance of probing into the core regions. In addition, since high-order $g$ modes are excited ($n \gg 1$), it is possible to use the asymptotic theory \citep{1980ApJS...43..469T} and the departures from uniform period spacing (by mode trapping) to explore the possible chemical inhomogeneities in the structure of the convective cores \citep{Miglio2008}. Stochastic excitation of solar like oscillations has been also predicted in $\gamma$ Dor stars \citep{2007A&A...464..659P}, but no positive detection has been reported yet. The instability strips of $\delta$ Sct and $\gamma$ Dor stars partially overlaps in the Hertzprung-Russell (HR) diagram \citep[see, for instance, Fig. 4 of][]{2013A&A...556A..52T}, strongly suggesting the existence of $\delta$ Sct-$\gamma$ Dor \emph{hybrid} stars, that is, stars showing high-frequency $p$-mode pulsations typical of $\delta$ Sct stars simultaneously with low-frequency $g$-mode oscillations characteristic of $\gamma$ Dor stars \citep{2004A&A...414L..17D, 2010ApJ...713L.192G}. The first example of a star pulsating intrinsically with both $\delta$ Sct and $\gamma$ Dor frequencies was detected from the ground \citep{2005AJ....129.2026H}. Other examples are HD 49434 \citep{2008A&A...489.1213U} and HD 8801 \citep{2009MNRAS.398.1339H}. A large sample of \emph{Kepler} and CoRoT stars yielded the first hints that hybrid behavior might be common in A-F type stars \citep{2010ApJ...713L.192G, 2010arXiv1007.3176H}. A follow up study with a large ($> 750$ stars) sample of $\delta$ Sct and $\gamma$ Dor candidates by \citet{2011A&A...534A.125U} revealed that out of 471 stars showing $\delta$ Sct or $\gamma$ Dor pulsations, 36\% (171 stars) are hybrid $\delta$ Sct-$\gamma$ Dor stars. Very recent studies \citep[e.g.,][]{2015AJ....149...68B} analyzing larger samples of $\delta$ Sct or $\gamma$ Dor candidates strongly suggest that hybrid $\delta$ Sct-$\gamma$ Dor stars are very common. \citet{2015MNRAS.452.3073B} studied the frequency distributions of $\delta$ Sct stars observed by the \emph{Kepler} telescope in short-cadence mode and found low frequencies (typical of $\gamma$ Dor stars) in \emph{all} the analyzed $\delta$ Sct stars. This finding renders somewhat meaningless the concept of $\delta$ Sct-$\gamma$ Dor hybrids. Apart from these important investigations of large samples of stars, there are published studied on several individual hybrid $\delta$ Sct-$\gamma$ Dor stars observed from space missions. Among them, we mention HD 114839 \citep{2006CoAst.148...28K} and BD+18-4914 \citep{2006CoAst.148...34R}, both detected by the MOST satellite. On the other hand, hybrid $\delta$ Sct-$\gamma$ Dor stars discovered with CoRoT observations are CoRoT 102699796 \citep{2011MNRAS.416.1535R}, CoRoT 105733033 \citep{2012A&A...540A.117C}, CoRoT 100866999 \citep{2013A&A...556A..87C}, and HD 49434 \citet{2015MNRAS.447.2970B}. Finally, among hybrid stars discovered with the \emph{Kepler} mission, we mention KIC 6761539 \citep{2012AN....333.1077H}, KIC 11145123 \citep{2014MNRAS.444..102K}, KIC 8569819 \citep{2015MNRAS.446.1223K}, KIC 9244992 \citep{2015MNRAS.447.3264S}, KIC 9533489 \citep{2015A&A...581A..77B}, KIC 10080943 \citep{2015MNRAS.454.1792K}. A few attempts of asteroseismic modeling of $\delta$ Sct stars have proven to be a very difficult task \citep{2001AJ....122.2042C,2003A&A...411..503C,2008A&A...478..855L, 2013MNRAS.432.2284M}. In part, this is due to that generally there are many combinations of the stellar structure parameters ($T_{\rm eff}, M_{\star}, Y, Z$, overshooting, etc) that lead to very different seismic solutions but that reproduces with virtually the same degree of precision the set of observed frequencies. The situation is potentially much more favorable in the case of hybrid $\delta$ Sct-$\gamma$ Dor stars, because the simultaneous presence of both $g$ and $p$ nonradial modes (in addition to radial pulsations) excited, that allows to place strong constraints on the whole structure, thus eliminating most of the degeneration of solutions. As such, hybrid stars have a formidable asteroseismological potential and are very attractive targets for modeling. Among the above mentioned hybrid objects, detailed seismological modeling has so far been performed only for a few $\delta$ Sct-$\gamma$ Dor hybrids, namely KIC 11145123 \citep{2014MNRAS.444..102K}, KIC 9244992 \citep{2015MNRAS.447.3264S}, and the binary system KIC 10080943 \citep{2016arXiv160507958S}. In this study, we present a seismic modeling of five $\delta$ Sct-$\gamma$ Dor hybrids including those two studied by \citet{2014MNRAS.444..102K} and \citet{2015MNRAS.447.3264S}. Our approach consists in the comparison of the observed pulsation periods with the theoretical adiabatic pulsation periods (and period spacings in the case of high-order $g$ modes) computed on a huge set of stellar models representative of A-F MS stars with masses in the range $1.2-2.2 M_{\sun}$ generated with a state-of-the-art evolutionary code. This approach is frequently referred to as \emph{grid-based} or \emph{forward} modeling in the field of solar-like oscillations \citep[e.g.,][]{2011ApJ...730...63G,2014A&A...564A.105H} and has been the preferred asteroseismological approach in pulsating white dwarfs \citep{2008A&A...478..869C,2010A&ARv..18..471A,2012MNRAS.420.1462R}. Furthermore, this approach has been adopted for the study of $\delta$ Sct-$\gamma$ Dor hybrid stars by \citet{2016arXiv160507958S}. The characteristics of the target stars are determined by searching among the grid of models to get a ``best-fit model'' for a given observed set of periods of radial modes, and $p$ and $g$ nonradial modes. In particular, we make full use of the valuable property that some hybrid stars offer, that is, the value of the mean period spacing $\left(\overline{\Delta \Pi}\right)$ of $g$ modes. Specifically, we will perform an asteroseismic modeling of the hybrid $\delta$ Sct-$\gamma$ Dor stars CoRoT 105733033, \citep{2012A&A...540A.117C}, CoRoT 100866999, \citep{2013A&A...556A..87C}, KIC 11145123 \citep{2014MNRAS.444..102K}, KIC 9244992 \citep{2015MNRAS.447.3264S}, and HD 49434 \citet{2015MNRAS.447.2970B}. The use of $\overline{\Delta \Pi}$ allows to discard a large portion of the grid of models ---those models that do not reproduce the observed period spacing. Also we assume, as usual, that the largest amplitude mode in the $\delta$ Sct region of the pulsation spectrum is associated to the fundamental radial mode ($\ell= 0, n= 0$) or the first radial overtone modes ($\ell= 0, n= 1,2,3,4,\cdots$). This step further reduces the number of possible seismological models. Finally, we perform a period-to-period fit to the $p$ mode periods. We also carry out other possible model selections, for instance, by performing direct period-to-period fits to the complete set of observed periods (including individual periods of $g$ modes and radial modes). This research constitutes the first stage of an ongoing systematic asteroseismic modeling program of hybrid $\delta$ Sct-$\gamma$ Dor stars in the La Plata Observatory. The paper is organized as follows: in Sect. \ref{tools} we describe our evolutionary and pulsation numerical tools. The main ingredients of the model grid we use to assess the pulsation properties of hybrid $\delta$ Sct-$\gamma$ Dor stars are described in Sect. \ref{Modelling}. Sect. \ref{impact} is devoted to describe in the effects that core overshooting and metallicity have on the pulsation properties of $g$ and $p$ modes. In Sect. \ref{applications} we present our asteroseismic analysis of the target stars in detail. Finally, in Sect. \ref{conclusions} we summarize our main findings. | \label{conclusions} In this work, we have presented a detailed asteroseismic study of five hybrid $\delta$ Scuti-$\gamma$ Doradus pulsating stars, aiming at derive their fundamental stellar parameters. To this end we built a huge grid of stellar models, covering the evolution of low-mass stars from the ZAMS to the TAMS, varying the stellar mass, the metallicity and the amount of core overshooting (see Sect. \ref{Modelling}). We employed the observational data of the detected periods reported in \citet{2014MNRAS.444..102K} for KIC 11145123, \citet{2015MNRAS.447.3264S} for KIC 9244992, \citet{2015MNRAS.447.2970B} for HD 49434, \citet{2012A&A...540A.117C} for CoRoT 105733033, and \citet{2013A&A...556A..87C} for CoRoT 100866999. We were able to obtain the fundamental parameters of the target stars by performing two different procedures, which fully exploit the simultaneous presence of $p$ and $g$ modes (and also presumably radial modes) in this kind of pulsating stars. For \textbf{Procedure 1} we used three constraints to find the best-fit seismological models: 1) the mean period spacing of high-order $g$ modes; 2) the largest amplitude mode in the $\delta$ Sct period domain, to which we associate a radial mode; and 3) a period-to-period fit of the individual $p$ modes. Besides, in this case we explore the incidence of adding a period-to-period fit of $g$ modes in the selection of the best fit model. Finally in \textbf{Procedure 2} we used again the mean period spacing of $g$ modes and, in addition, a period-to-period fit between the frequencies detected in the $\delta$ Sct domain and those calculated for models with $p$-mode periods and also (simultaneously) radial-mode periods. It is worth mentioning that these procedures do not depend on the reported spectroscopic information of the target stars, e.g., the effective temperature. Below, we summarize the results obtained for each star: \begin{figure}[h!] \begin{center} \includegraphics[clip,width=9 cm]{HRmodelosfinales.eps} \caption{HR diagram showing the asteroseismological models found and the respective evolutionary tracks for each target star.} \label{figurerhmodelosfinales} \end{center} \end{figure} \begin{itemize} \item {\bf KIC 11145123}: Two different seismological models were obtained from the \textbf{Procedures 1} with and without including $\chi^g$ in $F_1$(see Table \ref{modeloskurtzprocedures}). This means that individual $g$-mode period fits play an important role in the modeling of this star. Comparing our models with those obtained in \citet{2014MNRAS.444..102K}, we note some differences in the mass and the metallicity, possibly due to the fact that we did not considered atomic diffusion in our simulations ---the Brunt-V\"ais\"al\"a frequency is modified by this physical process--- and also that we neglected rotation in our pulsation modeling. Nevertheless, our best-fit models are both in the TAMS overall contraction phase, in good agreement with those obtained in \citet{2014MNRAS.444..102K}, and also have a metallicity consistent with the hypothesis that this star could be a SX Phe star. Note that none of both models reproduce well the reported effective temperature for this star. \item {\bf KIC 9244992}: The characteristics of the model obtained for this star are shown in Table \ref{modelossaioprocedures}. In this case, we obtained the same best-fit model with and without including $\chi^g$ in \textbf{Procedures 1}. This means that including a $g$-mode period-to-period fit does not affect the selection of the model. Our $2.1M_{\sun}$ best-fit model is more massive that the one proposed in \citet{2015MNRAS.447.3264S} of $1.45M_{\sun}$. Also, our model does not have overshooting in the core and has higher metallicity ($Z= 0.02$). Thus, our results do not indicates that this star is actually a SX Phe star. In addition, the $\log g= 3.8$ obtained is in good agreement with the spectroscopic study cited in \citet{2015MNRAS.447.3264S}, but we note that our model has an higher effective temperature. Despite these differences, our best fit model for KIC 9244992 is also at the end of the MS stage. \item {\bf HD 49434}: We perform \textbf{Procedure 2} for this star since the mode classification is not conclusive and therefore the existence of radial modes in the $\delta$ Sct region cannot be discarded. We obtained one model with $Z= 0.01$ and $f= 0.01$ from \textbf{Procedure 1}, and another model with $Z= 0.015$ and $f= 0.03$ from \textbf{Procedure 2}. Both models have the same mass, $1.75 M_{\sun}$. One of them has $f= 0.03$ and is \textbf{near} the TAMS, and the other one has $f= 0.01$ and is before the evolutionary ``knee'' where the overall contraction phase begins, as can be observed in Fig. \ref{figurerhmodelosfinales}. One of the main characteristics of this star is that it is a rapid rotator, and so it not shows a clear gap between the $\delta$ Sct and $\gamma$ Dor pulsation spectra regions. It is possible, as it is mentioned in \citet{2015MNRAS.447.2970B}, that the absence of the gap is due precisely to rotational splitting of high-degree $p$ modes. If this is the case, it is necessary a correct mode identification since our methodology strongly depends on this. On the other hand it is worth mentioning that adding a period-to-period fit for the $g$-modes does not affect the selection of the best fit model since we obtained the same models including $\chi^g$ in \textbf{Procedure 1}. The selection of another model when we include the possibility of having radial modes in the $\delta$ Sct period domain shows and reinforces the need of having a correct mode classification. \item {\bf CoRoT 105733033}: One of the remarkable characteristics of this star is the richness of its pulsational spectra. As it was mentioned before, it is possible to observe a clear distinction between low- and high-frequency regions in this star, which may be the consequence of a quite low angular rotation \citep{2012A&A...540A.117C}. More spectroscopic data is required to confirm this hypothesis. So far, no asteroseismological model has been proposed for this star, and their physical characteristics are uncertain. In this paper, we present the first asteroseismic model (see Table \ref{modeloschapprocedures}). As it was mentioned, we performed three different procedures and we obtained two different models, both at the overall contraction phase (Fig. \ref{figurerhmodelosfinales}). Again, from \textbf{Procedure 1} when we include $\chi^g$ we obtained the same model (one with $1.75 M_{\sun}$) and another different model from \textbf{Procedure 2} (the one with $1.85 M_{\sun}$). This means that the best fit model selected persists when we consider a period-to-period fit of $g$-modes, but changes when we include the possibility of having radial modes among the frequencies detected in the $\delta$ Sct domain. Anyway, both models are at the overall contraction phase, and have the same overshooting parameter and similar effective temperatures and surface gravities. \item {\bf CoRoT 100866999}: We obtained one model from \textbf{Procedure 1} and a different model from \textbf{Procedure 2}. The one obtained with \textbf{Procedure 1} has $1.55 M_{\sun}$ and is located before the evolutionary ``knee'' of the MS. The other one, obtained with \textbf{Procedure 2}, has $2.10 M_{\sun}$ and is on the overall contraction phase. Both models have $Z= 0.02$. Comparing these mass values with the ones obtained in \citet{2013A&A...556A..87C} from the eclipsing curve fit, we can see that ours masses are close to those calculated for the primary star ($1.8 \pm 0.2 M_{\sun}$). \end{itemize} In summary, we have obtained for the first time reliable asteroseismological models representative of five $\delta$ Sct-$\gamma$ Dor hybrid stars by means of grid-based modeling. These asteroseismological models result from different criteria of model selection, in which we take full advantage of the richness of periods that characterizes the pulsation spectra of this kind of stars. For four out the five stars analyzed, we have obtained the same asteroseismological model from \textbf{Procedure 1} including or not a period-to-period fit of the $g$ modes. In the cases when it was possible to apply \textbf{Procedure 2}, we obtained a different model from this approach. It is worth of notice that the true seismic model for a given target star must reproduce not only observed frequencies and regularities in the frequency spectra, but also frequency ranges of observed oscillations as ranges of pulsationally unstable radial and nonradial modes. We considered only adiabatic oscillations in our approach, and a detailed stability analysis of oscillations, which is beyond the scope of the present work, will be addressed in a future paper. Clearly, more theoretical work in the frame of this issue (like nonadiabatic stability computations), and also in other topics, for instance, the inclusion of the effects of rotation on the pulsation periods and substantial improvement of mode identification, will help us to break the degeneracy of the asteroseismological solutions. | 16 | 9 | 1609.02864 |
1609 | 1609.09492_arXiv.txt | gPhoton is a new database product and software package that enables analysis of GALEX ultraviolet data at the photon level. The project's stand-alone, pure-Python calibration pipeline reproduces the functionality of the original mission pipeline to reduce raw spacecraft data to lists of time-tagged, sky-projected photons, which are then hosted in a publicly available database by the Mikulski Archive at Space Telescope (MAST). This database contains approximately 130 terabytes of data describing approximately 1.1 trillion sky-projected events with a timestamp resolution of five milliseconds. A handful of Python and command line modules serve as a front-end to interact with the database and to generate calibrated light curves and images from the photon-level data at user-defined temporal and spatial scales. The gPhoton software and source code are in active development and publicly available under a permissive license. We describe the motivation, design, and implementation of the calibration pipeline, database, and tools, with emphasis on divergence from prior work, as well as challenges created by the large data volume. We summarize the astrometric and photometric performance of gPhoton relative to the original mission pipeline. For a brief example of short time domain science capabilities enabled by gPhoton, we show new flares from the known M dwarf flare star CR Draconis. The gPhoton software has permanent object identifiers with the ASCL (ascl:1603.004) and DOI (doi:10.17909/T9CC7G). \edit1{This paper describes the software as of version v1.27.2.} | The Galaxy Evolution Explorer \citep{mar2005} was a NASA Small Explorer (SMEX) telescope that surveyed the sky in the ultraviolet over ten years between launch on 28 April 2003 and spacecraft termination on 28 June 2013. The spacecraft, instruments, data, and calibration are well described in previous publications \citep{mor2005,mor2007} and the mission`s online technical documentation.\footnote{\url{http://www.galex.caltech.edu/wiki/Public:Documentation}} We will restrict discussion to topics that are necessary for completeness, have not appeared elsewhere in the literature, or are of particular importance to the gPhoton project. GALEX carried two micro-channel plate detectors (MCP) with 1.25 degree fields-of-view (FoV), simultaneously exposed via a dichroic. \edit1{The detectors record signals from electrical cascades, referred to as ``events,'' which were produced by photons hitting the MCPs. Detector positions and time stamps of these events recorded by the spacecraft were then corrected for instrumental effects} and re-projected into celestial coordinates by a calibration pipeline on the ground. The detectors observed in two broad ultraviolet (UV) bands centered around $1528\,\rm{\AA}$ (Far Ultraviolet or ``FUV'') and $2271\,\rm{\AA}$ (Near Ultraviolet or ``NUV''). The FUV detector failed in May of 2009, but the NUV detector continued to operate until the end of the mission. The spacecraft could observe in either direct imaging or slitless spectroscopic (grism) modes. Observations were conducted while the spacecraft was on the night side of each orbit (an ``eclipse''), which lasted 1500-1800 seconds. To avoid detector burn-in or local gain sag effects caused by depletion of electrons in the multiplier plate, the telescope did not stare at a fixed location on the sky during an observation but continuously moved the boresight relative to the target position. Several boresight patterns, or ``modes,'' were used over the course of the mission, which impacted the nature of the corresponding observational data. In the most basic ``dither'' mode, the spacecraft boresight would trace out a tight spiral pattern with a radius of $\sim1'$. Dither mode was used most often for Deep or Medium Imaging Surveys (DIS, MIS) in which a full eclipse of $\sim1600$ seconds was spent observing a single region of the sky. In the All-sky Imaging Survey (AIS) mode, the spacecraft boresight would jump between multiple positions (or ``legs'') on the sky for short integrations of $\sim100$ seconds each. Between each leg, the detector was set to a non-observing, low voltage state. This resulted in one independent observation (or ``visit'') per leg. Another mode, called ``petal pattern,'' was used to distribute the flux from particularly bright targets across the detector. Petal pattern is in some ways similar to the AIS mode, but the legs were tightly clustered into the approximate area of a single FoV and the detector remained in its nominal high-voltage state in between. On 4 May 2010, the ``Coarse Sun Point'' (CSP) anomaly---\edit1{a reference} to the safe mode entered by the spacecraft at that time---resulted in image degradation of the NUV detector. The CSP anomaly precipitated severe streaking in the detector's Y-direction, likely due to a failed capacitor. Although the effect was largely corrected through subsequent calibration and on-board adjustments, observations taken between 4 May and 23 June 2010 have substantially worse point spread functions (PSF). Care should be used when comparing observations made before this time range to observations made after to discount bias due to either degraded PSF or uncorrected ``ghost'' photons.\footnote{\url{http://www.galex.caltech.edu/wiki/Public:Documentation/Chapter_8}} NASA support for the mission ended in February of 2011. At that time, ownership of the spacecraft was transferred to the California Institute of Technology for a phase called the ``Complete the All-sky UV Survey Extension'' (CAUSE), during which operating costs were solicited from individuals or institutions, and spacecraft engineering constraints related to field and source brightness were relaxed, making it possible to observe bright regions of the sky that were off limits during the primary mission.\footnote{\url{http://www.galex.caltech.edu/cause/index.html}} Spacecraft slew rate limits were also relaxed, permitting a high-coverage ``scan mode'' that swept across several degrees of sky in a single integration \edit1{\citep{olmedo2015deep}}. Ownership of the CAUSE-phase data resides with each of the primary investigators, and only a small fraction of it has been made available to the public through MAST at the time of writing. Although the new calibration capabilities described herein may be of particular value in using and interpreting CAUSE data generally, and scan mode observations of very bright or dense fields in particular, this paper \edit1{and the current gPhoton database only cover} the direct imaging data through the end of the NASA-supported mission, corresponding to General Release 7 (GR7) in the MAST archives. Through GR7, GALEX collected data over 34,389 direct image eclipses, covering $\sim76.9\%$ of the sky in at least one band. Future work may add gPhoton support for CAUSE phase, scan mode, or spectroscopic data collected throughout the mission. In Section \ref{motivation} we describe the motivation behind constructing the gPhoton database and software suite. In Section \ref{database} we describe the design and content of the $\sim 1.1$ trillion row database hosted at MAST. In Section \ref{softwaretools} we describe the primary modules for generating photon lists, light curves and images. In Section \ref{calibration} we present tests of the calibration precision with respect to astrometry \edit1{and} photometry in relation to the mission catalogs, and photometry in relation to a calibration standard. In Section \ref{implementation} we discuss implementation challenges and solutions. Finally, in Section \ref{scienceexamples}, we highlight an example science case enabled by gPhoton: stellar flares of CR Draconis. \edit1{This paper describes version 1.27.2 of gPhoton. The software is under active development, and users are encouraged to consult the online documentation to supplement the information presented herein.} | The gPhoton project extends the utility of the GALEX data set well beyond the scientific objectives of the original mission, most specifically towards the study of short time domain UV variability. Some of the techniques developed for gPhoton can be applied to other data sets produced by non-integrating detectors, particularly micro-channel plates. The fact that spatial analyses can be performed by making direct queries at the photon-level data, rather than artificially degrading the spatial resolution of the data by integrating and interpolating into pixelated images, offers potential advantages in terms of both the flexibility of the data archive and the computational overhead for some types of analysis. While not trivial, the corresponding data management and volume issues associated with storing and retrieving massive amounts of photon-level data are entirely solvable with appropriate use of existing, off-the-shelf database and storage technology. The behavior of the GALEX detector during very short timespans (which correspond to small spatial sampling of the detector) is not well characterized, and further work on improving the resolution of the detector flat fields, as well as correctly propagating flux uncertainties, will be required to derive the maximum utility from the photon-level data. The gPhoton project is also a trial in an emerging paradigm for data archiving, where the functioning machinery for generating higher level data from lower---the calibration pipeline---is incorporated into the data archive itself. Even when preparation of the higher level data for archiving is well documented and comprehensible to future researchers, the priorities, interests, and needs of those users may not be the same as the data creators or archivists. At present, the standard recourse in such cases is to go back to some minimally reduced version of the data and create new tools or procedures for reducing the data from scratch. This can be onerous, time consuming, or impossible depending on the type of data, the quality of the documentation, and the availability of members of the original project team to answer inevitable questions. Especially when the data record observations that are unique or would be difficult to reproduce---for example, of rare astrophysical events in wavelengths only detectable above the atmosphere---an inability to reanalyze the data diminishes the long term value of results. Incorporating a \emph{functioning} calibration pipeline into the archive significantly lowers the barrier for independent research groups to modify that machinery to produce new science that was not anticipated by the original project teams. | 16 | 9 | 1609.09492 |
1609 | 1609.05991_arXiv.txt | Interstellar Pickup ions (PUIs) play a significant part in mediating the solar wind (SW) interaction with the interstellar medium. In this paper, we examine the details of spatial variation of the PUI velocity distribution function (VDF) in the SW by solving the PUI transport equation. We assume the PUI distribution is isotropic resulting from strong pitch-angle scattering by wave-particle interaction. A three-dimensional model combining the MHD treatment of the background SW and neutrals with a kinetic treatment of PUIs throughout the heliosphere and the surrounding local interstellar medium (LISM) has been developed. The model generates PUI power law tails via second-order Fermi process. We analyze how PUIs transform across the heliospheric termination shock (TS) and obtain the PUI phase space distribution in the inner heliosheath including continuing velocity diffusion. Our simulated PUI spectra are compared with observations made by {\it New Horizons}, {\it Ulysses}, {\it Voyager 1, 2} and {\it Cassini}, and a satisfactory agreement is demonstrated. Some specific features in the observations, for example, a cutoff of PUI VDF at $v = V_{SW}$ and a $f \propto v^{-5}$ tail in the reference frame of the SW, are well represented by the model. | A neutral atom from the local interstellar medium (LISM) can be ionized while drifting into the heliosphere and turn into an interstellar pickup ion (PUI). The neutral atom can become ionized through charge exchange with an ion, photoionization by sun light or impact ionization by Solar Wind (SW) electrons. Charge exchange is the dominant mechanism of the three. In the SW frame, a newly created ion is ``picked up'' and starts to gyrate about the direction of the magnetic field. Subsequently they may be scattered into an isotropic distribution by either ambient preexisting or self-excited waves \citep{1976JGR....81.1247V,1987JGR....92.1067I,1999SSRv...89..413Z}. The resulting PUI velocity distribution function (VDF) in the reference frame of the SW is a spherical shell, at the lowest order. These PUIs are also convected outward by the expanding SW. As the PUIs travel outward through the heliosphere, the shell becomes filled by adiabatic cooling. Generally, one can easily identify PUIs by their VDFs which are distinctly different from that of SW ions. In view of the paucity of measurements of keV-ions in the outer heliosphere, the transport of PUIs is not yet fully understood. Especially, the production of the suprathermal (high-energy non-Maxwellian) tails on the VDFs is a subject of much debate \citep{2003JGRA..108.1266C,2011A&A...533A..92F}. It was suggested that during their propagation through the outer heliosphere, PUIs experience pitch-angle scattering and stochastic acceleration by interactions with different kinds of SW turbulences \citep{1997A&A...320..659C}. PUI kinetic transport theory describes the evolving PUI distribution in phase space. The transport equation, an equation of ``motion'' for the distribution function, is the basis for almost all work on PUI transport. An early paper by \cite{1976JGR....81.1247V} calculated an isotropic PUI VDF without energy diffusion but including the effects of adiabatic deceleration. The VDF appears as a thick shell centered around the SW beam. \cite{1987JGR....92.1067I} investigated how PUI distribution varies with radial distance and phase space speed. He has shown that the effects of energy diffusion can be important and can lead to substantial particle acceleration. \cite{1995A&A...304..609C} studied the influence of different representations for the energy diffusion coefficient which would result from different radial variations of relative fluctuation amplitudes of MHD turbulence in the outer heliosphere. These authors have calculated the energy spectra of PUIs upstream of the heliospheric termination shock (TS) on the basis of realistic PUI production rates. They showed that second-order Fermi acceleration by means of Alfvenic turbulence produces the suprathermal tail. \cite{2011A&A...533A..92F} replaced the adiabatic cooling with the ``magnetic cooling'' process resulting from the conservation of the first and second adiabatic invariants. They have demonstrated that small second-order Fermi acceleration and magnetic cooling generates a PUI VDF with $f\propto v^{-5}$. \cite{2012JGRA..117.6104I} first compared simulated PUI densities and temperatures with the SWICS measurements on board {\it Ulysses}. The model results matched well with the observations at about $5.2$ AU during both quiet periods and the disturbed period during the Halloween 2003 storm. Assuming that the PUI distribution functions are $\kappa$ distributions, \cite{2014JGRA..119.7998F} proceeded from the phase space transport equation to a pressure equation for the $\kappa$ parameter $\kappa = \kappa(r)$ as function of heliocentric distance $r$. They obtained the range of possible radial variations of $\kappa$ from relatively high values in the inner heliosphere to values between 1.5 and about 2 further out, depending on diffusion coefficient. The purpose of the present paper is to place PUI transport in the context of the global picture of the SW-LISM interaction. The treatment is based on the more conventional view where wave-particle interactions ensure rapid isotropization of the distribution. A parallel can be drawn with the approach of \citet{2012ApJ...757...74G}, who used a grid-based model for the isotropic PUI distribution function and for the waves generated by the PUI ring anisotropization process, but neglected PUI momentum diffusion. Their model was applied to the supersonic SW upstream of the TS. Here we investigate the detailed spatial distribution of PUIs in both the supersonic SW and the heliosheath. We take into account the effects of convection with the SW, adiabatic cooling, second-order Fermi process and ionization. Similar to \citet{2012ApJ...757...74G}, the model combines the MHD treatment of the background SW and neutral atoms with a kinetic treatment of PUIs in the isotropic approximation. The rest of this paper proceeds as follows. Section 2 explains how we simulate the SW-LISM interactions. The PUI transport model is introduced in Section 3. Section 4 presents the simulated PUI phase space density and energy spectra. Finally, Section 5 mentions the weaknesses of the presented model and proposes several improvements to the current methods. | Our model introduced several improvements over the existing models. \cite{2004AdSpR..34...99C} have investigated the spatial variation of PUI spectra, but only the upwind part of the heliosheath was considered. \cite{2006A&A...445..693M} have performed a multi-species simulation, but the magnetic field was ignored and their model was two-dimensional. The model presented here computes PUI distributions on a three dimensional grid. \cite{2006JGRA..111.7101U} considered the SW outside $1$ AU as a combination of three co-moving species, SW protons, electrons, and PUIs, but they only computed the global structure of the SW from the coronal base to $100$ AU without the TS. Our model's external boundary is well in the LISM covering supersonic SW region, the inner heliosheath and the outer heliosheath. Several weaknesses of the present model are now pointed out. Firstly, the two-population assumption is admittedly questionable. The \cite{1996GeoRL..23.2871S} paper reported a range of field strength variations, rather than a bimodal distribution. The observed distribution functions are averaged over longer times than the averaging interval in \cite{1996GeoRL..23.2871S}, and therefore are a product of a superposition of high and low values of $\eta^2$. The simulated spectra shown in Figures 1 and 5 should be compared with monthly or even yearly averages of the data. Secondly, our model assumes the PUI VDF is isotropic, which is not always the case. For example, ion angular data from {\it Voyager 1} observations during 2002.58 to 2003.10, $85.3$ to $87.3$ AU showed large beamlike anisotropies \citep{2005Sci...309.2020D}. Thirdly, our model is time independent. Incorporating time-dependent SW boundary conditions may improve the results and produce better agreement with the observations. Finally, charge exchange of the PUIs on interstellar H atoms was ignored. This process replaces one pickup ion with another, drawn from a different velocity distribution. While, in principle, including the loss of PUIs would be straightforward, the production term requires numerical integration over the VDF of PUIs in each computational cell, which is a costly procedure. PUI charge exchange may be important in the inner heliosheath, causing an energy redistribution in their VDF. In spite of these limitations, our model does provide insights into the interpretation of the PUI data and may be used to predict PUI distribution at all locations inside the heliosphere. These distributions show the details that are directly comparable with those seen in spacecraft data. We have obtained the rapid drops in the spectra that appear to be required to match the observations. The model also features power-law tails in the energy, which are commonly observed in space. A velocity diffusion origin of these tails appears to be a valid interpretation. The compressed SW and PUIs behind the TS create energetic neutral atoms (ENAs) via charge exchange. ENAs with energies high enough to overcome the outward flow speed can be directed back at Earth. Future work will use these PUI results to calculate ENA fluxes at $1$ AU. We plan to compare the simulated ENA fluxes with the {\it IBEX} distributed ENA sky maps. This will bring us closer to explaining why the distributed ENA flux spectrum does not show a knee and why is it close to a power law \citep{2011ApJ...731...56S}. | 16 | 9 | 1609.05991 |
1609 | 1609.02922_arXiv.txt | We re-examine the classifications of supernovae (SNe) presented in the Lick Observatory Supernova Search (LOSS) volume-limited sample with a focus on the stripped-envelope SNe. The LOSS volume-limited sample, presented by \citet{2011MNRAS.412.1419L} and \citet{2011MNRAS.412.1441L}, was calibrated to provide meaningful measurements of SN rates in the local universe; the results presented therein continue to be used for comparisons to theoretical and modeling efforts. Many of the objects from the LOSS sample were originally classified based upon only a small subset of the data now available, however, and recent studies have both updated some subtype distinctions and improved our ability to perform robust classifications, especially for stripped-envelope SNe. We re-examine the spectroscopic classifications of all events in the LOSS volume-limited sample (180 SNe and SN impostors) and update them if necessary. We discuss the populations of rare objects in our sample including broad-lined Type Ic SNe, Ca-rich SNe, SN~1987A-like events (we identify SN~2005io as SN~1987A-like here for the first time), and peculiar subtypes. The relative fractions of Type Ia SNe, Type II SNe, and stripped-envelope SNe in the local universe are not affected, but those of some subtypes are. Most significantly, after discussing the often unclear boundary between SNe~Ib and Ic when only noisy spectra are available, we find a higher SN Ib fraction and a lower SN Ic fraction than calculated by \citet{2011MNRAS.412.1441L}: spectroscopically normal SNe~Ib occur in the local universe $1.7 \pm 0.9$ times more often than do normal SNe~Ic. | \label{sec:intro} The Lick Observatory Supernova Search (LOSS) has been a long-running project at the University of California, Berkeley, using the Katzman Automatic Imaging Telescope at Lick Observatory \citep[KAIT; e.g.,][]{2000AIPC..522..103L,2001ASPC..246..121F,2003fthp.conf..171F,2005ASPC..332...33F}, with many spectroscopic follow-up observations obtained with the 3\,m Shane telescope at Lick and the 10\,m telescopes at Keck Observatory. LOSS/KAIT has been discovering and observing SNe since first light in 1996; these data have contributed to several PhD theses and formed the foundation of many research projects on SNe. A detailed examination of the relative rates of nearby SNe was one of those projects, and was published as a series of papers in 2011 \citep{2011MNRAS.412.1419L,2011MNRAS.412.1473L,2011MNRAS.412.1441L,2011MNRAS.412.1508M,2011MNRAS.412.1522S}. The second of these, \citet[][L11 hereafter]{2011MNRAS.412.1441L}, presents a sample of 180 events that occurred within 80\,Mpc (for Type Ia SNe) or 60\,Mpc (for core-collapse SNe), all of which were spectroscopically classified \citep[the classes of SNe are differentiated primarily via spectroscopy; e.g.,][]{1997ARAA..35..309F}. Most SN classifications from this time period were performed via visual inspection and comparisons with spectra of a few SNe of well-understood types and subtypes. Over time we have found that a small fraction of the objects in L11 deserve reclassification; in some cases this is because the original classifications were made using only a subset of the now-available data on the objects, while in other cases our more modern classification methods are less prone to errors than the methods used at the time of classification. Independent of data quality or cadence, there is a history of debate in the literature over the exact distinction (if any) between SNe~Ib and SNe~Ic and whether transitional events showing weak helium lines exist \citep[e.g.,][]{1990AJ....100.1575F,1990RPPh...53.1467W,1994ApJ...436L.135W,1996ApJ...462..462C,2001AJ....121.1648M,2006PASP..118..791B}. The results of recent efforts by \citet{2014arXiv1405.1437L}, \citet{2014AJ....147...99M}, and \citet{2016ApJ...827...90L} argue that the distinction between SNe~Ib and SNe~Ic is useful, and they offer a clearly defined scheme for discriminating between them alongside updated software tools to perform those classifications in a repeatable manner. \citet{2014AJ....147...99M} identify as SNe~Ib all events with detections of both the \ion{He}{1}\,$\lambda$6678 and \ion{He}{1}\,$\lambda$7065 lines at phases between maximum light and $\sim$50\,days post-maximum, regardless of line strengths (the stronger \ion{He}{1} $\lambda$5876 line is also present, but overlaps with \ion{Na}{1}). They find that at least one good spectrum observed at these phases is necessary and sufficient to detect the helium lines, which are often absent at pre-maximum and nebular phases even for helium-rich events. Using this classification scheme, they find evidence for a transitional population of ``weak helium'' SNe~Ib \citep{2011MNRAS.416.3138V,2014AJ....147...99M,2016ApJ...827...90L}. Clarifying the distinction between SNe~Ib and SNe~Ic is important given the surprising ratio of population fractions for these subtypes found by LOSS \citep[SNe~Ic/SNe~Ib = $14.9^{+4.2}_{-3.8}\% / 7.1^{+3.1}_{-2.6}\%$;][]{2011MNRAS.412.1522S}, which has proven difficult to reproduce with stellar modeling efforts \citep[e.g.,][]{2009AA...502..611G,2010ApJ...725..940Y,2015PASA...32...15Y}, though see also \citet{2013AA...558L...1G}. \citet{2014AJ....147...99M} show that a subset of the objects originally labeled SNe~Ic in their sample in fact do qualify for the SN Ib label according to the definition above, and so they relabel these events as SNe~Ib (see their discussion of all such cases, in their \S 4.2). For some of those SNe, the spectra that were used to classify them and thus announce their types were obtained before the helium lines became prominent; for some, applying proper telluric corrections made the \ion{He}{1}\,$\lambda$6678 or $\lambda$7065 lines more apparent; for others the spectra show clear helium but the exact division between SNe~Ib and SNe~Ic was under debate in the literature at the time of classification \citep[e.g., SN 1990U;][]{1990IAUC.5069....1F,1990IAUC.5111....2F,2001AJ....121.1648M}. We explore these issues within the LOSS sample and also find that some events with helium lines were systematically labeled as SNe~Ic --- we update the classifications for these events and recalculate the relative fractions of core-collapse events. In this article we re-examine the classifications of the 180 events in the volume-limited sample of L11 and we make public all spectra of them we have been able to locate. This work was performed in conjunction with \citet{2016arXiv160902921G,2016arXiv160902923G}, who re-examine correlations between SN rates and galaxy properties. Note that much of the spectroscopy discussed herein has already been described in the literature and made publicly available by, for example, \citet[SNe~Ia]{2012MNRAS.425.1789S}, \citet[SNe~II]{2014MNRAS.445..554F,2014MNRAS.442..844F}, \citet[SNe~IIb/Ib/Ic]{2001AJ....121.1648M}. We collect these spectra, light curves obtained by LOSS, as-yet unpublished spectra from our archives, and as-yet unpublished spectra contributed from other SN research groups' archives, and analyze the complete set. We present 151 newly published spectra of 71 SNe and 20 rereduced KAIT light curves. In \S\ref{sec:data} we describe these data, in \S\ref{sec:methods} we detail our methods for classification, in \S\ref{sec:updated} we present all updated classifications and discuss notable events within the sample, in \S\ref{sec:rates} we calculate updated core-collapse SN rates in the local universe, in \S\ref{sec:progenitor} we discuss the implications these updates have for our understanding of the progenitors of stripped-envelope SNe, and in \S\ref{sec:conclusion} we conclude. | 16 | 9 | 1609.02922 |
|
1609 | 1609.07500_arXiv.txt | \noindent Multi-color photometry is presented for a large sample of local ellipticals selected by morphology and isolation. The sample uses data from $GALEX$, SDSS, 2MASS and {\it Spitzer} to cover the filters $NUV$, $ugri$, $JHK$ and 3.6$\mu$m. Various two-color diagrams, using the half-light aperture defined in the 2MASS $J$ filter, are very coherent from color to color, meaning that galaxies defined to be red in one color are always red in other colors. Comparison to globular cluster colors demonstrates that ellipticals are {\it not} composed of a single age, single metallicity (e.g., [Fe/H]) stellar population, but require a multi-metallicity model using a chemical enrichment scenario. Such a model is sufficient to explain two-color diagrams and the color-magnitude relations for all colors {\it using only metallicity as a variable on a solely 12 Gyrs stellar population with no evidence of stars younger than 10 Gyrs}. The [Fe/H] values that match galaxy colors range from $-$0.5 to +0.4, much higher (and older) than population characteristics deduced from Lick/IDS line-strength system studies, indicating an inconsistency between galaxy colors and line indices values for reasons unknown. The $NUV$ colors have unusual behavior signaling the rise and fall of the UV upturn with elliptical luminosity. Models with BHB tracks can reproduce this behavior indicating the UV upturn is strictly a metallicity effect. | One of the most intriguing topics in extragalactic research is the stellar populations in galaxies. For the underlying stellar population of a galaxy's light reveals the star formation and chemical history of galaxies, as well as representing the dominant baryon component in most galaxies. It is through the stars in galaxies that we observe them at high redshift, and how we understand the large scale structure of the Universe and the evolution of galaxies from the formation epoch to the present. Understanding galaxy kinematics, $M/L$'s, IMF, morphology and formation scenarios all revolve around the properties and characteristics of a galaxy's stellar population. Ellipticals represent one of the most carefully studied type of galaxies with respect to stellar populations. They exhibit the simplest morphological and structure as well as internal kinematics and, thus, are well-modeled as a single stellar population rather than the kinematically distinct components found in disk galaxies. They typically occupy the highest masses (i.e., luminosities) and, therefore, are the clearest signposts at high redshift. Studies of galaxy evolution often focus on ellipticals owing to early indications that their spectrophotometric changes are the simplest to model and to trace reliability through cosmic time. The study of stellar populations in ellipticals has historically taken three different routes. The first is the use of optical and near-IR colors to interpret the integrated light of the underlying stellar population. The discovery of the color-magnitude relation (CMR, Sandage \& Visvanathan 1978), the separation of morphology types by color (Tojeiro \etal 2013) and different galaxy components by color (e.g., bulge versus disk, Head \etal 2014) were early explorations into the stellar populations and the meaning of color with respect to the star formation history of galaxies (Tinsley 1978). Technological improvements in the 1980's led to the obvious extension of multi-color work through a higher inspection of the spectroenergy distribution (SED) with study of various spectral indices related to different types of stars found in a stellar population with a range of stellar masses. This type of investigation reached a peak with the development of the Lick/IDS line-strength system (Worthey \etal 1994; Trager \etal 2000) where a set of specific spectral features were should to correlate with the two primary characteristics of a stellar population, its age and mean metallicity. Guided by SED models of the Lick/IDS line-strength system (see Graves \& Schiavon 2008), these spectral indices became the observable of choice to study nearby and distant galaxy stellar populations. Lastly, with the launch of HST, space imaging provides the best study of a stellar population by direct examination of their color-magnitude diagrams, although this is still limited to the nearby Universe. For a majority of stellar population studies, the Lick/IDS line-strength system is the method of choice. The properties of optical colors of ellipticals, particularly the various color-magnitude relations, have often been taken as evidence in favor of the monolithic scenario for galaxy formation, the production of the entire stellar population of a galaxy in one single burst at redshifts greater than 5. The lack of redshift evolution of the slope and scatter in optical CMRs (e.g., van Dokkum \etal 2000), and observed passive color evolution (Rakos \& Schombert 1995), is consistent with a high formation redshift ($z > 2$). However, the predicted star formation histories (SFHs) of ellipticals in the hierarchical merger paradigm (Kauffmann \& Charlot 1998; Khochfar \& Burkert 2003) are much more complicated. For example, the predicted SFHs of ellipticals in the merger scenario are expected to be quasi-monolithic, with an overwhelming majority of the stellar mass forming before a redshift of 1 (Kaviraj \etal 2005) and expected to be difficult to discriminate using optical colors given the well-known age-metallicity degeneracy problem (Worthey 1994). The use of colors for investigating stellar populations in galaxies is needed even in the spectroscopic era. For example, higher signal-to-noise is acquired for faint, distant galaxies using colors. Large areal surveys can be obtained by wide-field cameras using well chosen filter sets. In the 1990's, Rakos \& Schombert pioneered the use of narrow band filters selected to cover age/metallicity features around the 4000\AA\ break as an fast and efficient system to study cluster galaxies using imaging (see Rakos \& Schombert 1995). The results from those studies confirmed a passive evolution for the stellar populations in cluster ellipticals, but was in sharp disagreement with the results from spectroscopic surveys that found much younger ages and lower metallicities for the same objects (Trager \etal 2000, Graves \etal 2009, Conroy \etal 2014). Larger SDSS samples (Gallazzi \etal 2005, Graves \etal 2010) presented a wider and more diverse range of ages and metallicities, and more sophisticated analysis techniques (see Johansson \etal 2012; Conroy \etal 2014; Worthey \etal 2014) reinforced the trend of age and metallicity with luminosity and mass. The disagreement between color results and spectroscopic results was outlined in Schombert \& Rakos (2009), an analysis of the CMR in the cluster environment and the expected colors based on the measured ages and metallicities with the Lick/IDS line-strength system. Despite the clear disparity between the expected colors from the young, metal-poor stellar populations deduced from Lick/IDS indices, the system is still very much in use (McDermid \etal 2015) and the resulting age and metallicity estimates are the core to most theoretical scenarios for galaxy formation and evolution (see review by Naab 2013). Resolving the conflict between colors and line indices is a crucial stellar population problem. In addition, colors from the extreme ends of the UV to near-IR wavelengths provides particularly useful information to various astrophysical problems. For example, the near and far-IR colors are salient to studies of the old component in stellar populations and the baryonic mass to light ratio (Schombert \& McGaugh 2014) The near and far-UV colors investigate the so-called "UV upturn" problem (Bertola \etal 1982, Brown \etal 1997), the unusually high UV fluxes in ellipticals presumingly dominated by old stellar populations without current star formation. The past determination of UV colors has presented mixed results with positive (Donas \etal 2007) and negative (Jeong \etal 2009) CMR slopes, and the prediction that only bright ellipticals display this behavior (Yi \etal 2005). This paper presents a comprehensive analysis of elliptical colors using archival data from the near-UV ({\it GALEX NUV} to the far-IR ({\it Spitzer} 3.6$\mu$m). With a near-IR selected sample from the Revised Shapley-Ames and Uppsala Galaxy catalogs, we will explore the behavior of galaxy colors over a wavelength range provided by {\it GALEX}, SDSS, 2MASS and {\it Spitzer}. Comparison with previous color studies will examine the differences and various photometric relationships. The colors presented herein will anchor the zeropoint of elliptical colors for use by high redshift studies. The colors will also provide a window in the stellar populations of ellipticals by comparison with simple and multi-metallicity population models. The wide wavelength coverage offers an avenue to break the age-metallicity degeneracy that plagues optical colors (Worthey 1994). | The Lick/IDS indices are often used to deduce age and metallicity, typically, through the H$\beta$ versus [MgFe] diagram. The standard procedure is outlined in Graves \& Schiavon (2008) and is applied to a sample of local Universe ellipticals using stacked SDSS spectra in Graves \etal (2009). Their results can be summarized is that ellipticals (from $M_J = -21$ to $-25$) increase in $\alpha$/Fe with luminosity, increase in age (from 6 to 12 Gyrs) and increase in [Fe/H] (from $-$0.4 to nearly solar). They also present a CMR for the SDSS color $g-r$ that progresses from 0.720 at $M_r = -18.8$ ($M_J = -20.7$) to 0.795 at $M_r = -23.0$ ($M_J = -25.1$). Compared to the CMR in Table 3, this agrees well at the faint end (0.721 for our sample), but deviates significantly at the high luminosity end (0.829 for our sample). Other Lick/IDS indices work find similar results but the number of young ellipticals ($\tau < 6$ Gyrs) varies from study to study. For example, Kuntschner \etal (2010) find very few ellipticals with ages less than 10 Gyrs. Thomas \etal (2010) finds 10\% of their elliptical sample with ages around 3 Gyrs. Their Figure 7 displays almost constant age for most of the sample (approximately 8 Gyrs) with 10\% containing a much younger population. These galaxies have $u-r < 2.4$, whereas only 6\% of our sample is that blue and all at the lowest luminosities in agreement with the CMR. McDermid \etal (2015) also finds a vast majority of ellipticals are old ($\tau > 10$ Gyrs). Studies of high redshift ellipticals present a conflicting picture. The massive end of the galaxy mass spectrum grows considerably, presumably by mergers (Brammer \etal 2011), to become the local ellipticals observed by this study. However, those objects derive from a population beyond $z=2$ that displays a large range in color and SFRs (van Dokkum \etal 2011). Fumagalli \etal (2016) find that ellipticals at $z=2$ have ages that correspond to local ages of 8 Gyrs, in agreement with the Graves \etal ages. And other Lick/IDS indices studies propose that age is correlated with galaxy mass (Gallazzi \etal 2006), again in agreement with the observations of galaxy evolution from high redshift. With respect to metallicity, the Lick/IDS indices studies are more coherent. They all universally find increasing [Fe/H] with increasing luminosity, stellar mass and velocity dispersion. But the range of [Fe/H] values is significantly lower than predicted by colors. Most find [Fe/H] between $-$0.5 and solar for the range of luminosities investigated in this paper, with the metallicity range decreasing with later studies. While some studies find $<$Fe$>$ or [MgFe] values well above solar (e.g., Gallazzi \etal 2006), most find a majority of ellipticals with less than solar metallicities, which would overlap their colors with globular clusters. Most Lick/IDS indices studies are in consensus concerning the results from $\alpha$/Fe observations such that they find a trend of increasing $\alpha$/Fe with stellar mass. While interpretation of the $\alpha$/Fe trends can be complicated, the primary driver in $\alpha$/Fe is the duration of star formation. Long durations lead to low $\alpha$/Fe (due to the increasing contribution of Fe from SN Ia) and short bursts lead to high $\alpha$/Fe values. The trend of $\alpha$/Fe indicates, at the very least, that the age difference between low and high mass ellipticals may arises solely from a difference in the star formation duration (in the direction of younger ages for low mass ellipticals). The complication is that to achieve near solar $\alpha$/Fe values only requires a duration of 1 to 2 Gyrs, a difference in age that is difficult to discern in indices or colors. Thomas \etal (2010) propose a scenario where low mass ellipticals have a longer duration plus a rejuvenation phase several Gyrs after the initial SF era, again, difficult to result in the red colors of even low luminosity ellipticals. Lastly, many studies conclude that ellipticals host, to some degree, a young stellar component (e.g. Trager \etal 2000; Kaviraj \etal 2007; Kuntschner \etal 2010). The amount of recent SF (and how recent) varies considerably and it is unclear whether the Lick/IDS indices are more sensitive to recent SF than colors. This would explain the difference between index ages and color ages as they reflect differing effects from a young component. We do know that any SF event involving more than 5\% of the stellar mass of an elliptical within the last 500 Myrs would alter the colors in a fashion greater than the photometric errors quoted herein, particularly for the $NUV$ and $u$ filters. With respect to stellar mass in ellipticals, a majority of the Lick/IDS indices studies are in agreement that high mass ellipticals are all old (12 Gyrs) and high in metallicity (although not as metal-rich as deduced from their colors). Combined with red colors in all the filters (except $NUV$), it is assumed that high mass ellipticals formed from a monolithic scenario, an initial short of duration less than 0.5 Gyrs with a halt of SF by galactic winds. This is the canonical scenario where the range in [Fe/H] by galaxy mass is due solely to the depth of the gravitational well and the longer time required before the galactic winds gain enough energy to remove the remaining gas and halt SF. The formation scenarios proposed for low mass ellipticals vary from study to study. For example, McDermid \etal (2015) propose two phase elliptical star formation history starting with an early monolithic collapse phase producing metal-rich, $\alpha$/Fe enhanced ellipticals followed by an era accreting gas for an extended era of star formation. This produces a low mass population of elliptical that is younger and metal-poor (from the infalling gas) while the massive ellipticals quench early (for older ages). Chiosi \& Carraro (2002) also claim that line indices results can be modeled by a scenario where high mass ellipticals form by monolithic processes and low mass ellipticals are the result of irregular and intermittent episodes of star formation. Here the observed lower age indices are interpreted as the result of a more complex star formation history in low mass ellipticals (perhaps a series of bursts), particular longer durations of initial star formation, which would match the observed $\alpha$/Fe ratios. The difference in the evolution of low versus high mass ellipticals is supported by the Faber \etal (2007) observations that found rapid changes in the number of red, presumably, early-type galaxies since $z=1$. While it appears that most massive galaxies have already joined the red sequence at $z=1$ (Treu et al. 2005; Bundy et al. 2006; Cimatti et al. 2006), the low mass end of the red sequence seems to be subject to the continued arrival and rejuvenation of galaxies up to the present day (Treu \etal 2005; Schawinski \etal 2007). The key conflict, outlined by this paper, is that the results from the Lick/IDS line-strength studies can not be reconciled with the average colors of present-day ellipticals from $NUV$ to 3.6$\mu$m. And the claims that various formation and evolution scenarios reproduces observations ignores the colors of ellipticals. With respect to ages, we find agreement with results from Lick/IDS indices on the high mass end for old ($\tau = 12$ Gyrs) ages. However, the conjecture of ages from 4 to 6 Gyrs for the low mass end of the elliptical sequence is in direct contradiction with the observed colors. While this comparison is model dependent (i.e., the indices values and expected colors) it is important to note that both indices and colors are extracted from the same SSP models. And, while colors fade faster than indices for young populations, the model predicted colors are still significantly deviant from the indices extracted ages. This is shown in \S3.3 for all the colors in our sample. This type of color analysis has only be lightly explored once the Lick/IDS line-strength system was defined in the literature. For example, Chiosi \& Carraro (2002) consider the effect of secondary episodes of SF on $B-V$ colors and rule out any activity with 5 Gyrs and engaging more than 5\% of the galaxy mass. And Rakos \etal (2008) consider the impact of younger ages in cluster ellipticals using narrow band optical colors. The latter study outlined the incompatible with ages younger than 10 Gyrs and mean cluster elliptical colors. In addition to the difficulty of reconciling mean colors with ages, the CMR across our filter sets is also incompatible with even a very narrow range of ages outside 12 Gyrs. One can reproduce the slope of each individual CMR by adjusting age downward with a corresponding increase in metallicity. But the resulting age and metallicity values produce discrepant values in the other colors and their CMR's. In fact, as shown in \S5, all the CMR's across all colors are explained solely by changes in [Fe/H] and any introduction of younger age on the low luminosity end forces the slopes of other CMR's outside the photometric and fit errors. While Kaviraj \etal (2005) argues, through a $\Lambda$CDM hierarchical merger model, that the CMR is not a meaningful tool for testing monolithic versus merger scenarios, we find, in fact, that a comparison across a large range in wavelength does highly constrain the possible range in age and metallicity. The results of this study of the colors of ellipticals can be summarized as the following: \begin{itemize} \item{} Multi-color photometry is presented for a large sample of local ellipticals selected by morphology and isolation. The sample uses data from $GALEX$, SDSS, 2MASS and {\it Spitzer} to cover the filters $NUV$, $ugri$, $JHK$ and 3.6$\mu$m with various levels of completeness outlined in Table 1. \item{} Excellent agreement (typically less than 0.1 mags) is found between magnitudes measured herein and the colors and magnitudes from the RC3 and other published photometry. Metric colors defined by the total luminosity in the $J$ band are defined for all galaxies in the sample. Mean colors for various luminosity ranges are listed in Table 2. \item{} The various two-color diagrams are very coherent from color to color, meaning that galaxies defined to be red in one color are always red in other colors. The exception is the behavior of $NUV-u$, which is discussed separately in \S6. The colors of ellipticals are extensions of globular cluster colors with a detectable change in slope of each two-color relationship from the globulars. \item{} Various 12 Gyrs SSP models from the literature are tested against the two-color diagrams. All reproduce globular cluster colors with varying degrees of accuracy with respect to metallicity. Most importantly, the ellipticals have a shallower slope not predicted by the SSP models. The difference is greater for bluer colors. This is strong evidence that ellipticals are only roughly approximated by an SSP model and a composite population, with a range of metallicities, is required to explain the colors of ellipticals (see Rakos \& Schombert 2009). \item{} The simplest composite population is one of singular age and a range of metallicities outlined by a chemical enrichment model. We use a model with an age of 12 Gyrs and a chemical enrichment scenario outlined in Schombert \& McGaugh (2014). This first order composite model is sufficient to match all the colors in our sample except for the $NUV-u$ colors. Most importantly, only a 12 Gyrs aged population with varyingly metallicity is required to match ellipticals colors, and ages less than 8 Gyrs are ruled out based on colors unless extremely high metallicities are assumed. In addition, a mean [Fe/H] value can be assigned based on the average of five colors with the same level of accuracy as the globular calibration. \item{} Using the globular clusters to calibrate the color-metallicity relationship, we find the metallicities of ellipticals are expected to range from [Fe/H]=$-$0.5 to +0.4, much higher than the [Fe/H] values found by Lick/IDS indices values. And no evidence for less than 10 Gyrs ages, particularly younger ages for low mass ellipticals. In addition, comparison of Lick/IDS indices and colors plus composite population models are difficult to reconcile with young galaxy ages. While it is not the intent of this study to deduce age or metallicity values strictly from colors, we do find that the deduced ages and metallicity from line studies are in conflict with the expected colors. \item{} The CMR across all colors display the increasing color with luminosity (except $NUV-u$, see below). The measured slopes agree with slopes from numerous other studies on the CMR from the near-UV to the near-IR. Modeling the CMR using younger ages requires unrealistic metallicities at the low luminosity end and, more importantly, it is impossible to reproduce the CMR across all colors with younger age population even using metallicity as a variable. \item{} The $NUV$ colors have unusual behavior near $u$ with a inverse CMR and up-and-down signature in two-color diagrams. This indicates an decrease in the UV upturn to intermediate luminosity ellipticals, then a strengthening to higher masses. Models with BHB tracks can reproduce this behavior and matching the globular cluster colors indicating the UV upturn is a metallicity effect. \end{itemize} The tension between galaxy colors and Lick/IDS indices was first outlined in Schombert \& Rakos (2009) and is even more salient with this study in terms of improved accuracy and a wider wavelength coverage. On the other hand, the measured line indices values have also improved and there is no indication that the published values are anything less than increasingly accurate and multiple observations reinforce the reliability of those values. Improved models do not resolve the conflict and, as most of our conclusions about the formation and evolution of galaxies are based on these types of observations. It is critical that future indices work include colors (even if extracted from the spectra) as a reality check to the continuum measurements of the SED in galaxies. A combination of colors and indices are required to explore the effects of low luminosity components (e.g., a weak metal-poor component or a recent SF event). \noindent Acknowledgements: We thank the editorial staff and the referee, Guy Worthey, for their diligence and insightful comments. We are also grateful to Stacy McGaugh and Federico Lelli for encouragement. The software and funding for this project was supported by NASA's Applied Information Systems Research (AISR) and Astrophysics Data Analysis Program (ADAP) programs. Data used for this study was based on observations made with (1) the NASA Galaxy Evolution Explorer, GALEX is operated for NASA by the California Institute of Technology under NASA contract NAS5-98034, (2) SDSS where funding has been provided by the Alfred P. Sloan Foundation, the Participating Institutions, the National Science Foundation, the U.S. Department of Energy, the National Aeronautics and Space Administration, the Japanese Monbukagakusho, the Max Planck Society, and the Higher Education Funding Council for England, (3) the Two Micron All Sky Survey (2MASS), which is a joint project of the University of Massachusetts and the Infrared Processing and Analysis Center/California Institute of Technology, funded by the National Aeronautics and Space Administration and the National Science Foundation and (4) archival data obtained with the Spitzer Space Telescope, which is operated by the Jet Propulsion Laboratory, California Institute of Technology under a contract with NASA. In addition, this research has made use of the NASA/IPAC Extragalactic Database (NED) which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration. \pagebreak | 16 | 9 | 1609.07500 |
1609 | 1609.01110_arXiv.txt | In recent years, significant progress has been made in building new galaxy clusters samples, at low and high redshifts, from wide-area surveys, particularly exploiting the Sunyaev--Zel'dovich (SZ) effect. A large effort is underway to identify and characterize these new systems with optical/NIR and X-ray facilities, thus opening new avenues to constraint cosmological models using structure growth and geometrical tests. A census of galaxy clusters sets constraints on reionization mechanisms and epochs, which need to be reconciled with recent limits on the reionization optical depth from cosmic microwave background (CMB) experiments. Future advances in SZ effect measurements will include the possibility to (unambiguously) measure directly the kinematic SZ effect, to build an even larger catalogue of galaxy clusters able to study the high redshift universe, and to make (spatially-)resolved galaxy cluster maps with even spectral capability to (spectrally-)resolve the relativistic corrections of the SZ effect. | Galaxy clusters are among the most studied and interesting objects of our universe. Recent technological improvements have made possible a massive increase of the size of galaxy cluster catalogues and a key role has been played by the full exploitment of the Sunyaev--Zel'dovich (SZ) effect as galaxy clusters searcher. The SZ effect is a powerful tool to study our universe at low and high redshift, to investigate the adiabaticity of the universe expansion, the equation of state of dark energy, the dark matter distribution in the universe, and the astrophysics governing the biggest objects of our Universe. The SZ effect has been historically described by three main contributions: the thermal SZ effect arising from the thermal motion of ionised medium in the intra-cluster medium (ICM), the kinematic SZ effect arising from the peculiar motion of the galaxy clusters, and the relativistic corrections to the SZ effect which originate when one accounts for the relativistic temperatures characterising the ICM. \subsection{Thermal SZ effect} The thermal SZ effect \citep{sz70,sz72} is a spectral distortion of the cosmic microwave background (CMB) caused by inverse Compton scattering between the CMB photons and a hot electron gas present in the ICM in clusters of galaxies. A complete derivation of the effect can be found in several reviews \citep{ZS,reph95,birk99,carl02}. In the nonrelativistic limit, the spectral behavior of the distorted spectrum can be obtained by solving the Kompaneets equation \citep{komp1957}, which yields the occupation number $n(\nu)$ of the radiation energy levels: \begin{eqnarray} \frac{\partial n}{\partial t}=\frac{kT_{e}}{m_ec}\frac{\sigma_Tn_e}{x_{e}^2}\frac{\partial}{\partial x_{e}}\left[x_{e}^4\left(\frac{\partial n}{\partial x_{e}}+n+{n}^2\right)\right]\!, \end{eqnarray} where $n_e$ is the electron number density, $m_e$ and $T_e$ are the electron mass and gas temperature respectively, $\sigma_T$ is the Thomson cross-section and $x_{e}=h\nu/k_{B}T_{e}$ (not to be confused with the dimensionless frequency $x=h\nu/k_{B}T_{\rm CMB}$). Under the assumption that $T_{e}\gg T_{\rm CMB}$, the Kompaneets equation admits a simple analytical solution. We define the parameter $y$ as: \begin{eqnarray} y=n_{e}\sigma_{T}\frac{k_{B}T_{e}}{m_{e}c^{2}}ct, \end{eqnarray} which represents a dimensionless measurement of the time spent in the electron distribution and can be written in the form: \begin{eqnarray} y=\int\! n_{e}\sigma_{T}\frac{k_{B}T_{e}}{m_{e}c^{2}} dl=\tau\frac{k_{B}T_{e}}{m_{e}c^{2}}, \end{eqnarray} where the integration is calculated along the line of sight $dl$, $\tau$ is the optical depth and electron temperature $T_{e}$ has been assumed constant; $y$ is known as the Comptonization paramenter. At low optical depth and low electron temperature, for a Planckian incident radiation spectrum, we get a spectral variation of the occupation number of the form: \begin{eqnarray} \Delta n(x)=xy\frac{e^{x}}{(e^{x}-1)^{2}}\left(x\cdot \textnormal{coth}\left(\frac{x}{2}\right)-4\right) \end{eqnarray} and thus the specific intensity change due to thermal SZ effect is: \begin{eqnarray} \frac{\Delta I_{\rm TSZ}(x)}{I_{0}}=x^{4}y\frac{e^{x}}{(e^{x}-1)^{2}}\left(x\cdot \textnormal{coth}\left(\frac{x}{2}\right)-4\right)=y g(x),\label{eq:isz} \end{eqnarray} where $I_{0}=\frac{2h}{c^{2}}(\frac{k_{B}T_{\rm CMB}}{h})^{3}$. The spectral behavior of the thermal SZ effect is shown in Fig.~1(a). From the above equations, we can extract the main features of the thermal SZ effect in the nonrelativistic approximation: \begin{itemize} \item the SZ effect amplitude depends only on $y$: it linearly depends on the cluster electron temperature and on the optical depth $\tau$ (i.e. on the cluster pressure integrated along the line of sight); \item the spectral behavior is described by relatively simple analytical functions (i.e. $g(x)$); \item we have a zero point of the effect for $x=3.83$ (i.e. $\nu=217$\,GHz) with $\Delta I_{\rm TSZ}(x)\,{<}\,0$ for $x<3.83$ and $\Delta I_{\rm TSZ}(x)>0$ for $x>3.83$; \item the specific intensity is characterized by a minimum value at $x=2.26$ and a maximum at $x=6.51$; \item since the SZ effect arises solely from interaction of CMB photons with hot cluster gas and the CMB is a background that exists everywhere, the SZ effect does not suffer spherical dilution or cosmological dimming as an isotropic radiator does; this results in its redshift independence. This is valid also for its spectral behavior under the assumption that the CMB temperature scales with the redshift as \begin{eqnarray} T_{\rm CMB}(z)=T_{\rm CMB}(0)(1+z). \end{eqnarray} \end{itemize} \subsection{Kinematic SZ effect}\label{par:ksz} While the thermal SZ effect is due to random thermal motion of the ICM electrons with isotropic distribution, if the cluster has a finite peculiar velocity, one expects an additional kinematic effect due to the electrons motion with respect to the CMB, which causes a Doppler shift of the scattered photons. If one assumes that thermal and the kinematic SZ effect are separable, it is forward to extract the spectral dependence of the kinematic SZ: \begin{eqnarray} \frac{\Delta I_{\rm KSZ}(x)}{I_{0}}=x^{4}\frac{e^{x}}{(e^{x}-1)^{2}}\frac{v_{p}}{c}\tau= h(x)\frac{v_{p}}{c}\tau, \end{eqnarray} where $v_{p}$ is the peculiar cluster velocity along the line of sight. The net effect observed on the CMB spectrum is thus a change of the temperature of the Planckian spectrum, found to be higher for negative velocities (i.e. moving toward the observer) and vice versa (see Fig.~1). The kinematic SZ effect is a powerful method to determine peculiar cluster velocities along the line of sight. The spectral coincidence between the maximum intensity of the kinematic effect and the zero point of the thermal effect allows, in principle, to distinguish between them. The spectral behavior of the kinematic SZ effect, is anyway identical to that of the CMB anisotropies: this results in an effective difficulty to disentangle these two effects from their spectral behavior (see Fig.~1(b)). \begin{figure} \begin{tabular}{ccc} \includegraphics[width=5.3cm]{1630023f01a.pdf} \includegraphics[width=5.3cm]{1630023f01b.pdf} \includegraphics[width=5.3cm]{1630023f01c.pdf} \end{tabular} \caption{Left: thermal SZ effect for $y=10^{-4}$, $7 \times 10^{-5}$, $5 \times 10^{-5}$, $2 \times 10^{-5}$. Centre: thermal and kinematic SZ effect for $y=10^{-4}$ , $v_{p}/c=10^{-3}$ and $T_{e}=8.2$keV. The kinematic curve has been multiplied by a factor -10. Right: relativistic corrections to the SZ effect up to the $5^{th}$ order in $\Theta_{e}$ for $y=10^{-4}$, and $T_{e}=8.2$keV.} \label{fig:sz} \end{figure} \subsection{Relativistic corrections}\label{par:relSZ} ICM electrons with temperatures from $\simeq\!3$\,keV are characterized by near relativistic velocities. In order to perform a $more$ $exact$ calculation of the SZ effect, one needs to take into account relativistic corrections to it. Taking into account these effects drives to nonnegligible corrections of the order of a few percent of the thermal SZ effect itself. Different methods have been proposed in order to deal with this problem: a numerical approach consists in performing Monte Carlo simulations of interactions between electrons and photons in a fully relativistic regime. The analytical approach allows to explicit the relativistic correction in terms of powers of the electron cluster temperature. The inadequacy of the Kompaneets equation had already been considered early in 1979 \citep{wright} and in 1981 \citep{Fabbri81} when it was extended the formalism to the case of little number of scatterings. An explicit treatment for typical cluster temperatures has been presented by \cite{reph95b}, who has stressed, in agreement with \cite{itoh98} the shift of the zero point frequency $x_{0}$ of the thermal SZ effect: $x_{0}$ is pushed to higher values with increasing electron temperature $\Theta_{e}=\frac{k_{B}T_{e}}{m_{e}c^{2}}$. \cite{chal98} focused of the corrections in the Rayleigh--Jeans part of the spectrum while a second-order approximation which takes into account additional corrections related with the peculiar cluster velocity has been found by \cite{Saz99}. An analytical fitting formula for the relativistic corrected thermal SZ effect, $\frac{\Delta I_{{\rm T}_{R}{\rm SZ}}(x)}{I_{0}}$ up to the fifth-order in $\Theta_{e}$ has been reported by \cite{itoh98}{:} \begin{equation} \frac{\Delta I_{T_{R}SZ}(x)}{I_{0}}=yx^{4}\frac{e^{x}}{(e^{x}-1)^{2}}(Y_{0}+\Theta_{e}Y_{1}+\Theta_{e}^{2}Y_{2}+\Theta_{e}^{3}Y_{3}+ \Theta_{e}^{4}Y_{4})=yg_{rel}(x,T_{e}) \end{equation} where $Y_{n}$ are a set of functions depending on the dimensionless frequency only. The spectral behavior of the relativistic corrections to the thermal SZ effect is shown in Fig.~1(c), once subtracted by the uncorrected thermal SZ spectrum. \cite{noza98} have added to the above expression the terms originated by the effect of the peculiar cluster velocity, while \cite{itoh01} have extended the relativistic correction up to the seventh-order in power of $\Theta_{e}$ and \cite{shim04} have found an accurate expression of the cross over frequency up to the fourth-order in $\Theta_{e}$ considering also the dependence on $\tau$. In the relativistic corrected treatment the spectral behavior of the thermal SZ effect is thus dependent on $T_{e}$ and the one of the kinematic SZ effect depends on both $T_{e}$ and $v_{p}$. Multifrequency observations of the SZ effect are necessary to disentangle the different effects. In principle, one could use the kinematic SZ and the relativistic corrections of the SZ effect to infer the properties of the galaxy clusters. \clearpage | 16 | 9 | 1609.01110 |
|
1609 | 1609.06440_arXiv.txt | We have performed a multi-wavelength analysis of a mid-infrared (MIR) bubble N37 and its surrounding environment. The selected 15$' \times$15$'$ area around the bubble contains two molecular clouds (N37 cloud; V$_{lsr}\sim$37--43 km s$^{-1}$, and C25.29+0.31; V$_{lsr}\sim$43--48 km s$^{-1}$) along the line of sight. A total of seven OB stars are identified towards the bubble N37 using photometric criteria, and two of them are spectroscopically confirmed as O9V and B0V stars. Spectro-photometric distances of these two sources confirm their physical association with the bubble. The O9V star is appeared to be the primary ionizing source of the region, which is also in agreement with the desired Lyman continuum flux analysis estimated from the 20 cm data. The presence of the expanding H{\sc ii} region is revealed in the N37 cloud which could be responsible for the MIR bubble. Using the $^{13}$CO line data and photometric data, several cold molecular condensations as well as clusters of young stellar objects (YSOs) are identified in the N37 cloud, revealing ongoing star formation (SF) activities. However, the analysis of ages of YSOs and the dynamical age of the H{\sc ii} region do not support the origin of SF due to the influence of OB stars. The position-velocity analysis of $^{13}$CO data reveals that two molecular clouds are inter-connected by a bridge-like structure, favoring the onset of a cloud-cloud collision process. The SF activities (i.e. the formation of YSOs clusters and OB stars) in the N37 cloud are possibly influenced by the cloud-cloud collision. | Massive stars ($>$8 M$_\odot$) play a crucial role in the evolution of their host galaxies, but their exact formation and evolution mechanisms are still under debate \citep{zinnecker07,peters12,dale15,kuiper15}. It is not yet understood whether the formation of massive stars is only a scaled-up version of birth process of low mass stars, or is it a completely different process. One can find more details about the current theoretical scenarios of massive star formation in the recent reviews by \citet{zinnecker07} and \citet{tan14}. Recently, a collision between two molecular clouds followed by a strong shock compression of gas is considered as a probable formation mechanism of massive stars \citep{furukawa09,ohama10,fukui14,torii15}. \citet{habe92} numerically found that the head-on collision between two non-identical molecular clouds can trigger the formation of massive stars, and such process could also form a broken bubble-like structure. In a detailed study of RCW 120 star-forming region using the molecular line data, \citet{torii15} reported that the collision between two nearby molecular clouds has triggered the formation of an O star in RCW 120 in a short time scale. However, observational evidences for the formation of O stars via a collision between two molecular clouds are still very rare. Massive stars can significantly influence the surrounding interstellar medium (ISM) through their energetics such as ionizing radiation, stellar winds, and radiation pressure. They have an ability to help in accumulation of surrounding materials (i.e., positive feedback) and/or to disperse matter into the ISM. Furthermore, they can also affect the star formation positively and negatively \citep{deharveng10}. The positive feedback of massive stars can trigger the birth of a new generation of stars including young massive star(s). More details about the various processes of triggered star formation can be found in the review article by \citet{elmegreen98}. However, the feedback processes of massive stars are not yet well understood, and the direct observational proof of triggered star formation by massive stars is rare. But the influence of massive stars on their surroundings can be studied with several other observational signatures (like H {\sc ii} region, wind-blown or radiation driven Galactic bubble, etc.). Recently, {\it Spitzer} observations have revealed thousands of ring/shell/bubble-like structures in the 8 $\mu$m images \citep{churchwell06,churchwell07,simpson12}, and many of them often enclose the H\,{\sc ii} regions. Hence, the bubbles associated with H\,{\sc ii} regions are potential targets to probe the physical processes governing the interaction and feedback effect of massive stars on their surroundings. Additionally, these sites are often grouped with the infrared dark clouds (IRDCs) and young stellar clusters, which also allow to understand the formation and evolution of these stellar clusters. In this paper, we present a multi-wavelength study of such a mid-infrared (MIR) bubble, N37 \citep[$l=$ 25$^\circ$.292, $b=$ 0$^\circ$.293;][]{churchwell06}, which is associated with an H\,{\sc ii} region, G025.292+00.293 \citep{churchwell06,deharveng10,beaumont10}. The bubble N37 is classified as a broken or incomplete ring with an average radius and thickness of 1$\farcm$77 and 0$\farcm$49, respectively \citep{churchwell06}. The bubble is found in the direction of the H\,{\sc ii} region RCW 173 (Sh2-60) \citep[see Figure~9 in][]{marco11}. The velocity of the ionized gas \citep[$\sim$39.6~km\,s$^{-1}$;][]{hou14} is in agreement with the line-of-sight velocity of the molecular gas \citep[$\sim$41~km\,s$^{-1}$;][]{beaumont10,shirley13} towards the bubble N37, indicating the physical association of the ionized and molecular emissions. Presence of several IRDCs are also reported around the N37 bubble by \citet{peretto09}. \citet{marco11} analyzed the photometry and spectroscopy of stars in the direction of the H\,{\sc ii} region, RCW 173, and found that most of the stars in the field are reddened B-type stars. They also identified a star a805 (G025.2465+00.3011) having spectral type of O7II and suggested this as the main ionizing source in the area. Several kinematic distances (2.6, 3.1, 3.3, 12.3, and 12.6 kpc) to the region are listed in the literature \citep[e.g.][]{beaumont10,churchwell06,blitz82,watson10,deharveng10}. However, it has been pointed out by \citet{churchwell06} that the MIR bubbles located at the Galactic plane are likely to be veiled behind the foreground diffused emission if they are situated at a distance larger than $\sim$8 kpc. Hence, it is unlikely for the bubble N37 to be located at a distance of about 12 kpc. Therefore, in this work, we have adopted a distance of 3.0 kpc, the average value of all available near-kinematic distance estimates. We infer from the previous studies that the bubble is associated with an H\,{\sc ii} region and an IRDC together. However, the physical conditions inside and around the bubble N37 are not yet known, and the ionizing source(s) of the bubble is yet to be identified. Furthermore, the impact of the energetics of massive star(s) on its local environment is not yet explored. The detailed multi-wavelength study of the region will allow us to study the ongoing physical processes within and around the bubble N37. To study the physical environment and star formation mechanisms around the bubble, we employ multi-wavelength data covering from the optical, near-infrared (NIR) to radio wavelengths. The paper is presented in the following way. In Section~\ref{sec:observations}, we describe the details of the multi-wavelength data. We discuss the overall morphology of the region in Section~\ref{sec:morphology}. In Section~\ref{sec:result}, we present the main results of our analysis. The possible star formation scenarios based on the multi-wavelength outcomes are discussed in Section~\ref{sec:discussion}. Finally, we conclude in Section~\ref{sec:conclusions}. | \label{sec:conclusions} We performed a multi-wavelength analysis of the Galactic MIR bubble N37 and its surrounding environment. The aim of this study is to investigate the physical environment and star formation mechanisms around the bubble. The main conclusions of this study are the following. 1. In the selected $15\arcmin \times 15 \arcmin$ region around the MIR bubble N37, two molecular clouds (N37 molecular cloud and C25.29+0.31) are present along the line of sight. The molecular cloud associated with the bubble (i.e. N37 molecular cloud) is depicted in the velocity range from 37 to 43 km s$^{-1}$, while the C25.29+0.31 cloud is traced in the velocity range from 43 to 48 km s$^{-1}$. The N37 molecular cloud appears to be blue-shifted with respect to the C25.29+0.31 cloud. 2. Using photometric criteria, we find a total of seven OB stars within the N37 bubble, and spectroscopically confirmed two of these sources as O9V and B0V stars. The physical association of these sources with the N37 bubble is also confirmed by estimating their spectro-photometric distances. The O9V star is found as the primary ionizing source of the region. This result is in agreement with the Lyman continuum flux analysis using the 20 cm data. 3. Several molecular condensations surrounding the N37 bubble are identified in the {\it Herschel} column density map. The physical association of these condensations with the N37 bubble is inferred using the molecular gas distribution as traced in the integrated $^{13}$CO (J=1--0) map. Surface density analysis of the identified YSOs reveals that the YSOs are clustered toward these molecular condensations. 4. The mean ages of YSOs located in different parts of the region indicate that it is unlikely that these YSOs are triggered by energetics of the OB stars present within the bubble. This interpretation is supported with the knowledge of the dynamical age of the H {\sc ii} region. 5. The position-velocity analysis of $^{13}$CO data shows that two clouds (N37 molecular cloud and C25.29+0.31) are interconnected with a lower intensity emission known as broad bridge structure. The presence of such feature suggests the possibility of interaction between the N37 molecular cloud and the C25.29+0.31 cloud. 6. The position-velocity analysis of $^{13}$CO emission also reveals an inverted C-like structure, suggesting the signature of an expanding H\,{\sc ii} region. Based on the pressure calculations (P$_{HII}$, $P_{rad}$, and P$_{wind}$), the photoionized gas associated with the bubble is found as the primary contributor for the feedback mechanism in the N37 cloud. Possibly the expanding H\,{\sc ii} region is responsible for the origin of the MIR bubble N37. 7. The collision between two clouds (i.e. N37 molecular cloud and C25.29+0.31) might have changed the uniformity of the molecular cloud which is depicted by a slight change in the polarization position angles of background starlight. 8. The collision between the N37 molecular cloud and the C25.29+0.31 cloud might have triggered the formation of massive OB stars. This process might also have triggered the formation of YSOs clusters in the N37 molecular cloud. | 16 | 9 | 1609.06440 |
1609 | 1609.04359_arXiv.txt | {The methylidyne cation (CH$^+$) and hydroxyl (OH) are key molecules in the warm interstellar chemistry, but their formation and excitation mechanisms are not well understood. Their abundance and excitation are predicted to be enhanced by the presence of vibrationally excited H$_2$ or hot gas ($\sim$500$-$1000~K) in photodissociation regions with high incident FUV radiation field. The excitation may also originate in dense gas ($>10^5$~cm$^{-3}$) followed by nonreactive collisions with H$_2$, H, and electrons. Previous observations of the Orion Bar suggest that the rotationally excited CH$^+$ and OH correlate with the excited CO, a tracer of dense and warm gas, and formation pumping contributes to CH$^+$ excitation.} {Our goal is to examine the spatial distribution of the rotationally excited CH$^+$ and OH emission lines in the Orion Bar in order to establish their physical origin and main formation and excitation mechanisms.} {We present spatially sampled maps of the CH$^+$ J=3-2 transition at 119.8~$\muup$m and the OH $\Lambda$-doublet at 84~$\muup$m in the Orion Bar over an area of 110$\arcsec \times$110$\arcsec$ with \textit{Herschel} (PACS). We compare the spatial distribution of these molecules with those of their chemical precursors, C$^+$, O and H$_2$, and tracers of warm and dense gas (high-J CO). We assess the spatial variation of CH$^+$ J=2-1 velocity-resolved line profile at 1669~GHz with \textit{Herschel} HIFI spectrometer observations.} {The OH and especially CH$^+$ lines correlate well with the high-J CO emission and delineate the warm and dense molecular region at the edge of the Bar. While notably similar, the differences in the CH$^+$ and OH morphologies indicate that CH$^+$ formation and excitation are strongly related to the observed vibrationally excited H$_2$. This, together with the observed broad CH$^+$ line widths, indicates that formation pumping contributes to the excitation of this reactive molecular ion. Interestingly, the peak of the rotationally excited OH 84~$\muup$m emission coincides with a bright young object, proplyd 244-440, which shows that OH can be an excellent tracer of UV-irradiated dense gas.} {The spatial distribution of CH$^+$ and OH revealed in our maps is consistent with previous modeling studies. Both formation pumping and nonreactive collisions in a UV-irradiated dense gas are important CH$^+$ J=3-2 excitation processes. The excitation of the OH $\Lambda$-doublet at 84~$\muup$m is mainly sensitive to the temperature and density.} | \label{sect:intro} The methylidyne cation (CH$^+$) and hydroxyl (OH) have been observed in different environments from the local \citep[e.g.,][]{Storey1981, Nagy2013, Dawson2014} to the extragalactic ISM \citep[e.g.,][]{Schmelz1986, Baan1992, Darling2002, Spinoglio2012, Rangwala2014}. They are key molecules in the warm interstellar chemistry, and because they require ultraviolet (UV) radiation and high temperatures to form, they trace specific physical processes in the ISM \citep[e.g., ][]{Gerin2016}. Thus, it is expected that in addition to the physical conditions of the regions they are observed in, the spatial distribution of the rotationally excited far infrared (FIR) emission of these species might give clues on their formation and excitation processes. \textit{Herschel} allows, for the first time, to study the spatial distribution of these lines. In this paper, we present the first spatially resolved maps of these lines in the Orion Bar. Atoms and hydrogen molecules and radicals are the first chemical building blocks of the ISM. In warm gas CH$^+$ is believed to form mainly as a product of the $C^+ + H_2$ reaction. This formation route has a very high endothermicity of 0.374~eV (4300~K), and it has been suggested that this barrier could be overcome by reactive collisions with the vibrationally excited H$_2$ in strongly irradiated PDRs \citep[e.g.,][]{White1984, Lambert1986, Jones1986, Agundez2010, Naylor2010, Godard2013, Nagy2013, Zanchet2013}. In diffuse interstellar clouds with low far-ultraviolet (FUV) radiation field and very low density, shocks and turbulence \citep[e.g.,][]{Elitzur1978, PineaudesForets1986, Godard2009, Godard2012, Falgarone2010a, Falgarone2010b} have been proposed to overcome the high endothermicity. In addition, \citet{Lim1999} found that the observed abundances of CH$^+$ by \citet{Cernicharo1997} are explained through thermal reaction between C$^+$ and H$_2 \,(v=0)$ if the gas is hot enough ($>$1000~K). The reaction $C^+ + H_2 (\rm v)$ becomes exothermic when H$_2$ is in vibrationally excited states \citep{Godard2013, Zanchet2013}. On the other hand, CH$^+$ is highly reactive and easily destroyed by reactive collisions with H$_2$, H, and electrons. \citet{Godard2013} and \citet{Zanchet2013} find that with the chemical destruction rates considered typical for PDR environments, all levels of CH$^+$ ($J\ge2$) are very sensitive to the formation pumping since the time-scale for the chemical reaction becomes comparable or shorter than that of the nonreactive collision. Collisional excitation with H$_2$, H, and electrons could, however, be important for the lowest rotational transitions. In contrast, radiative pumping is predicted to have only a marginal effect even in a strong FIR radiation field. The OH radical is a key intermediary molecule in forming other important PDR tracers like H$_2$O, CO$^+$, O$_2$, NO and SO$^+$ in the ISM. On the surface of high FUV-flux PDRs the OH formation is expected to be dominated by the endothermic reaction with H$_2$ and atomic oxygen O$^0$ \citep{Goicoechea2011, Hollenbach2012}. The route is endothermic by 0.08~eV ($\sim$900~K) and with an activation barrier of 0.4~eV ($\sim$4800~K). Similarly to CH$^+$, the possibility of the FUV-pumped vibrationally excited H$_2$ enhancing the abundance and excitation of OH has been suggested. \citet{Goicoechea2011} made the first detection of rotationally excited CH$^+$ and OH emission lines with \textit{Herschel}/PACS towards the \textit{CO$^+$ peak} \citep{Storzer1995} at the edge of the Orion Bar PDR, one of the nearest, nearly edge-on luminous PDRs (with a FUV radiation field of a few 10$^4$ in Draine units). While their pointed observations cover a small area (47$\arcsec \times$~47$\arcsec$), they hint at a possible spatial correlation between these lines and suggest that the rotational OH emission originates in small irradiated dense structures ($n_H\sim$10$^{6-7}$~cm$^{-3}$ and T$_k\sim$160$-$220~K). The Orion Bar is thought to contain different physical structures with an interclump medium at medium density ($n_H\sim 10^4-10^{5}$~cm$^{-3}$) and high density clumps ($n_H\sim$10$^{6-7}$~cm$^{-3}$) \citep[e.g., ][]{Tielens1993, Tauber1994, Lis2003, Lee2013}. A recent analysis of the high-J CO, H$_2$, OH, and CH$^+$ emission lines using the PDR Meudon code by Joblin et al. (in prep.) shows that these lines are very sensitive to the thermal pressure and suggests $P\sim2 \times 10^8$~K~cm$^{-3}$ for the emitting structures. However, this study uses pointed observations and in order to understand the CH$^+$ and OH formation and excitation in detail, we need to understand their spatial distribution and how they compare with other tracers. Using high spectral resolution \textit{Herschel}/HIFI observations, \citet{Nagy2013} find that the CH$^+$ J=1-0 and J=2-1 lines have line widths of $\Delta V \sim 5$~km~s$^{-1}$ towards the Orion Bar, larger than those of other molecular lines, which typically show line widths of 2$-$3~km~s$^{-1}$. This line broadening could result from formation pumping leading to a non-thermalized velocity distribution. In this paper, we study the spatial distribution of the rotationally excited CH$^+$ and OH emission lines using fully sampled PACS maps of CH$^+$ transition J=3-2 at 119.8~$\muup$m and OH 84~$\muup$m $\Lambda$-doublet towards the Orion Bar. These are the first fully sampled maps of these emission lines in a PDR. Observational constraints on the excited CH$^+$ and OH spatial distribution and comparison with tracers of warm and dense gas and vibrational excited H$_2$ are needed to establish their physical origin and their main excitation mechanisms (collisions versus pumping). Given the similar critical densities ($\sim$10$^{10}$~cm$^{-3}$) and upper level energies (E/k$\sim$250~K) of the targeted lines, our observations allow us to study how the chemical reaction with H$_2 \, (v>0)$ affects the formation and excitation of CH$^+$ and OH. We first describe the \textit{Herschel} observations and data reduction of the CH$^+$ and OH lines, as well as the lines observed for comparison, in Sect. \ref{sect:observations}. We discuss the line detection, spatial morphology and origin of the CH$^+$ J=3-2 and OH 84~$\muup$m lines in Sect. \ref{sect:morphology}. In Sect. \ref{sect:H2_comparison} we investigate the CH$^+$ and OH formation and chemical pumping excitation mechanism via reaction with vibrationally excited H$_2$. In this section we also compare OH with H$_2$O, since OH can also be the product of H$_2$O photodissociation in the gas unshielded against FUV radiation. In Sect. \ref{sect:HIFI}, we discuss the CH$^+$ velocity dispersion. The detection of a proplyd in the OH map is described in Sect. \ref{sect:proplyd}. We conclude and summarize the findings of the paper in Sect. \ref{sect:conclusions}. | \label{sect:conclusions} We have presented \textit{Herschel} fully sampled maps of the FIR rotationally excited emission lines of CH$^+$ and OH towards the Orion Bar over a large area ($\sim$110$\arcsec$ $\times$ 110$\arcsec$). The CH$^+$ and OH lines delineate the edge of the Bar indicating that they are good tracers of the warm molecular zone and very sensitive to the physical conditions. Our main results are summarized as follows. \begin{enumerate} \item We confirm the correlation of OH and CH$^+$ with high-J CO emission (a tracer of dense gas). We find that the spatial thickness of the observed line emission layers results from the Bar being tilted towards the observer, an effect already inferred in other studies. OH, CH$^+$, and high-J CO originate in the inclined thin irradiated surface of the Bar with embedded dense structures. The emission peaks of these lines are associated with the surface of the largest clumps (density enhancements) seen with high density tracers, such as HCN or CS. \item Although excited CH$^+$ and OH display similar overall spatial distribution, there are also relevant differences in their morphology. The good correlation of CH$^+$ with vibrationally excited H$_2$ supports that CH$^+$ is formed via the reaction $C^+ \, + \, H_2 \, (v > 0)$. We provide observational evidence to the fact that the excitation of the CH$^+$ rotational levels is affected by formation pumping which dominates the excitation for J$\ge$2 levels, as suggested by previous models. The CH$^+$ emission is limited by the abundance of vibrationally excited H$_2$ and not by the C$^+$ abundance. We find the C$^+$ emission to be more extended than the CH$^+$ emission. \item OH is less well correlated with the vibrationally excited H$_2$, a fact pointed out by theoretical studies predicting the reaction $O^0 \, + \, H_2 \, (v=0) \to H + OH$ to dominate the OH formation. The other formation route for OH via the photodissociation of H$_2$O appears not to be as important since the bulk of the OH and water emission seem to arise from different cloud depths. \item We find a broad line width of $\sim$3$-$4~km~s$^{-1}$ for CH$^+$ \mbox{J=2-1}, which is broader than other molecular line tracing dense gas in the Orion Bar. The line width most likely reflects the excess energy transferred to translational energy during the formation pumping of CH$^+$. \item Interestingly, the peak of the OH 84 $\muup$m emission corresponds to the position of the proplyd 244-440. FIR OH emission could be a good diagnostic of hot gas in the externally illuminated protoplanetary disks. \end{enumerate} | 16 | 9 | 1609.04359 |
1609 | 1609.02396_arXiv.txt | The origin of rings around giant planets remains elusive. Saturn's rings are massive and made of 90-95\% of water ice with a mass of $\sim 10^{19}$ kg. In contrast, the much less massive rings of Uranus and Neptune are dark and likely to have higher rock fraction. According to the so-called "Nice model", at the time of the Late Heavy Bombardment, giant planets could have experienced a significant number of close encounters with bodies scattered from the primordial Kuiper Belt. This belt could have been massive in the past and may have contained a larger number of big objects ($M_{\rm body}=10^{22}$kg) than what is currently observed in the Kuiper Belt. Here we investigate, for the first time, the tidal disruption of a passing object, including the subsequent formation of planetary rings. First, we perform SPH simulations of the tidal destruction of big differentiated objects ($M_{\rm body}=10^{21}$ and $10^{23}$kg) that experience close encounters with Saturn or Uranus. We find that about $0.1-10$\% of the mass of the passing body is gravitationally captured around the planet. However, these fragments are initially big chunks and have highly eccentric orbits around the planet. In order to see their long-term evolution, we perform N-body simulations including the planet's oblateness up to $J_4$ starting with data obtained from the SPH simulations. Our N-body simulations show that the chunks are tidally destroyed during their next several orbits and become collections of smaller particles. Their individual orbits then start to precess incoherently around the planet's equator, which enhances their encounter velocities on longer-term evolution, resulting in more destructive impacts. These collisions would damp their eccentricities resulting in a progressive collapse of the debris cloud into a thin equatorial and low-eccentricity ring. These high energy impacts are expected to be catastrophic enough to produce small particles. Our numerical results also show that the mass of formed rings is large enough to explain current rings including inner regular satellites around Saturn and Uranus. In the case of Uranus, a body can go deeper inside the planet's Roche limit resulting in a more efficient capture of rocky material compared to Saturn's case in which mostly ice is captured. Thus, our results can naturally explain the compositional difference between the rings of Saturn, Uranus and Neptune. | \label{sec:intro} The origin of planetary rings is still a debated question. Saturn's main rings are unique as they are made of 90-95\% water ice \citep{Cuz98, Pou03, Nic05} with a mass of $\sim 10^{19}$ kg \citep{Esp83, Cha09}. In contrast, the much less massive rings of Uranus and Neptune are dark and likely to have a higher rock content \citep{Tis13} than Saturn's rings. Whereas dusty Saturn's E and G rings are likely to be formed via the destruction or surface erosion of the nearby present satellites \citep{Esp93, Col94, Bur01, Hed07, Por06}, Saturn's main rings cannot result from the same process as there is no obvious source of material to feed them today. Note, however, that a recent study shows that the origin of Saturn's F ring and Uranian $\epsilon$ ring could be a natural consequence of the collisional destruction between small satellites just outside the main rings \citep{Hyo15b} that is formed by the spreading of ancient rings \citep{Cha10, Cri12, Hyo15a}.\\ Several ring formation scenarios have been proposed for massive rings, like those of Saturn: (1) primordial satellite collisional destruction by passing comet \citep{Pol73, Pol75, Har84}, (2) tidal destruction of a primordial satellite at the Roche Limit after inward migration due to tidal interaction with the circum-Saturn gas disk \citep{Can10} and (3) tidal disruption of passing objects \citep{Don91}. The inward migration of a primordial Titan-sized satellite and the removal of only its pure icy mantle could beautifully explain the silicate deficit of Saturn's rings. However, it requires some fine-tuning of the timing of the event (likely at the end of the evolution of the circum-Saturn circumplanetary disk) so that the disk is still massive enough to allow inward migration of the satellite, but light enough in order to prevent the rapid infall of debris into the planet because of gas drag. In addition, it would be difficult to directly form centimetre- to meter-sized particles that are currently seen in Saturn's main rings by tidal destruction alone. \cite{Can10} proposes collision between fragments can form small particles, but detailed studies are still required. Tidal disruption of a passing differentiated object could also potentially explain the high ice/rock fraction by capturing only the icy mantle of the incoming body and letting the remnant core escape from Saturn's gravity field. However, so far this scenario has scarcely been studied and only used a simplified analytical model of a homogeneous body \citep{Don91, Cha09}. Thus, direct numerical simulation of the tidal splitting of a big differentiated body is necessary now to investigate this scenario in detail. In addition, even though some mass capturing occurs, the fragments are expected to have highly eccentric orbits around the planet and the long-term evolution of such fragments remains unclear, in particular by which process a ring of cm-sized particles forms. In this work, we investigate, for the first time, the details of tidal disruption of a passing large differentiated object and the long-term fate of its debris by using direct simulations.\\ Such an event may have occurred, with the most probability, either during the phase of planet formation where the giant-planets' cores are expected to scatter the neighbouring planetesimals efficiently, or later during the Late Heavy Bombardment (LHB). The well known "Nice model" explains not only the Lunar cataclysm or today's orbital architecture of giant planets \citep{Gom05, Tsi05}, but also the implantation of Jupiter's Trojan asteroids as well as the irregular satellites around giant planets \citep{Mor05, Nes07}. During this instability phase, giant planets could have experienced a significant number of close encounters with bodies scattered from the primordial Kuiper Belt that surrounded the giant planets. This belt could have been significantly massive and may have contained a larger number of big objects than what is currently observed in the Kuiper Belt \citep{Lev08,Nes16}. \cite{Cha09} estimated an encounter rate of the primordial Kuiper Belt Objects (KBOs) with the giant planets during the LHB and investigated the captured mass around giant planets using a simplified analytical model assuming homogeneous small bodies like comets. The flux of such undifferentiated small bodies is enormous and isotropic, and thus the average angular momentum of captured mass should be almost zero, resulting in no contribution to the formation of the rings. However, only a single tidal disruption of a large object that deviates from the average could decide the story, and such large objects could be expected to be differentiated like Pluto.\\ Here we use two different direct simulations and investigate successive processes from tidal destruction of a passing object to the possible formation of planetary rings (Figure \ref{summary}). Our work will address (1) the captured mass as well as its ice/silicate fraction due to the tidal disruption of differentiated bodies at different planets, (2) the orbits of the captured fragments, and (3) the long-term orbital and collisional evolution of the captured fragments. To do so, we first investigate the physics of tidal disruption of a differentiated object that is initially on a hyperbolic orbit about Saturn or Uranus, and calculate how much mass is gravitationally captured around these planets by using smooth particle hydrodynamics (SPH) simulations. Then we perform direct N-body simulations in order to see the longer-term evolution of such captured fragments around Saturn, including the effect of oblateness potential of the planet. In section 2, in light of our newly derived semi-analytical models that take into account spin and self-gravity of a differentiated object, we briefly review a previous analytical formula of the capture efficiency of tidal disruption. In section 3, we explain our SPH method and model. In section 4, we show the results from SPH simulations and discuss the mass capture efficiency as well as orbits of captured fragments. In section 5, using the data obtained from SPH simulations as initial conditions, we perform N-body simulations of subsequent long-term evolution of captured fragments. Then, using results of the simulations as well as analytic estimation, we discuss the fate of the captured fragments. Section 6 summarises our results and discusses the origin of planetary rings. | The origin of rings around planets is still debated. In this work, we investigate the possibility that a single close encounter of a large differentiated body may form rings around giant planets (Figure \ref{summary}). We perform two different direct simulations (SPH simulations and N-body simulations) in order to understand the physics of the tidal disruption of a passing large differentiated primordial KBO and long-term evolution of the resultant captured fragments. At the time of LHB, we assume that giant planets could suffer a significant number of encounters with primordial KBOs \citep[see also][]{Cha09}. In the previous works, such a process was investigated using a simplified analytical formula for the capture efficiency of mass around a planet \citep{Don91, Cha09}. However, the physical model in these works only considered ballistic orbits of the constituent particles and neglected the self-gravity and spin state of the passing body. Recently, \cite{Nes16} have argued that the primordial Kuiper belt should have had $1000$ to $4000$ Pluto-sized bodies ($M_{\rm body} \sim 10^{22}$kg) in order to explain the observed abundance of objects on non-resonant orbits in the present Kuiper belt. Computation of encounter probabilities (see Section 2.3) suggests that all giant planets may have experienced a few to several tens of encounters inside their Roche limit with Pluto-sized objects during the LHB. Thus, this big reservoir of impactors might have delivered significant mass to the giant planet system through tidal disruption. Using semi-analytical arguments and SPH simulations, we first investigate the detailed physical process of tidal disruption of a passing large differentiated object. Our semi-analytical model shows significant effects of the self-gravity on the captured mass and we find that Dones' (1991) formula for the capture efficiency is not always consistent with our simulation results. When the object is large enough ($M_{\rm body}=10^{23}$kg), the capture efficiency tends to be smaller than Dones' formula regardless of the spin state of the object. In contrast, when the object is smaller ($M_{\rm body}=10^{21}$kg), the capture efficiency can be larger and smaller than Dones' formula depending on the direction of the spin of the object. Such deviation from Dones' formula becomes larger as the pericenter distance becomes larger. We also investigated the effect of an initial spin of the passing body and found that destruction becomes more significant and capture efficiency becomes larger when the body has an initial prograde spin with respect to the direction of the encounter, while the capture efficiency becomes smaller when the body has an initial retrograde spin. SPH simulations show that, during a close encounter, tidal forces spin up the body in the prograde direction. Therefore, a body with an initial prograde spin is more easily and efficiently deformed (smoothly elongated) and destroyed than a body with an initial retrograde spin. If the body's pericenter is well inside the planet's Roche limit and above the planet's radius, then the body is disrupted and about 0.1-10\% of its mass is captured and expected to end up in a ring-structure. Assuming an average capture probability of about 1\% and encounter rates reported in Section 2.3, the total mass delivered to giant planets through the tidal break-up of objects with the mass of $M_{\rm body}=10^{22}$kg may range from 8, 4, 2 and $2 \times 10^{20}$kg (for Jupiter, Saturn, Uranus and Neptune, respectively) up to 32, 16, 8 and 8 $\times 10^{20}$kg. Close encounters of small bodies with planets is a natural consequence of the LHB. As was also discussed in \cite{Cha09}, tidal disruption of passing objects could form massive rings around all giant planets. In addition, recent works have shown that the inner regular satellites of Saturn, Neptune and Uranus could form by spreading of ancient massive rings \citep{Cha10,Cri12,Hyo15a}, and narrow rings such as Saturn's F ring and Uranian $\epsilon$ ring with their shepherding satellites could be the natural consequence at the last stage of such ring spreading across the Roche Limit \citep{Hyo15b}. During a close encounter of a $M_{\rm body} = 10^{21}$kg passing object with Saturn, our numerical simulations show that only enough mass to explain Saturn's current ring ($ \sim 10^{19}$kg) can be embedded. On the other hand, when the object has the mass $M_{\rm body} = 10^{23}$kg, enough mass not only for the current rings but also for the total mass of its regular satellites (up to and including Rhea) can be embedded. In addition, we find a small fraction of silicate material in the captured mass because the object's mantle is preferentially disrupted. In the case of Uranus, the captured mass could explain only the mass of the current rings (mass of $10^{15}-10^{16}$ kg \citep{Fre91}), assuming an encounter with a body with mass $M_{\rm body} = 10^{21}$kg. In contrast, ring mass as well as the mass of all satellites up to Oberon ($\sim 10^{22}$kg) can be explained by $M_{\rm body} = 10^{23}$kg body. Furthermore, in the case of Uranus, due to its higher density, the width between the planet's surface and its Roche limit is larger than in the case of Saturn. Thus, a body can pass deeper potential field of the planet. As a result, the tidal destruction could be significant enough to disrupt not only the body's icy mantle but also its silicate core, and thus silicate components can be more efficiently captured than in the case of Saturn. This would also be applicable to Neptune since it is also denser than Saturn. Therefore, this could explain the fact that the rings of Uranus and that of Neptune are darker than that of Saturn \citep{Tis13} \\ Soon after the capture, the fragments are in the form of big chunks ($m \sim 10^{18-19}$kg in the case of $M_{\rm body} = 10^{21}$kg and $m \sim 10^{20-21}$kg in the case of $M_{\rm body} = 10^{23}$kg) and they have very large eccentricities ($e \sim 0.9-0.98$). Therefore, their long-term evolution needs to be investigated to see if the system becomes circular equatorial rings as currently seen in the ring systems around planets. Thus, using the data obtained from SPH simulations, we perform N-body simulations of the captured fragments including the oblate potential of the planets. Since their pericenter distances are deep inside the planet's Roche limit, after several Keplerian orbits ($\sim$ year), the fragments are further tidally destroyed and form a ringlet-like structure which still has a large eccentricity on the same plane. However, they would still be large particles ($\sim$km to 100km) assuming their physical tensile strength for pure ice at low temperatures \citep{Har78}.\\ Due to orbital precession, these individual particles form a torus-like structure ($\sim$ hundreds-to-thousand years). In this configuration, orbital crossing between particles is enhanced and particles could experience high velocity collisions ($v_{\rm coll} \sim$ few km/s). Tidal disruption itself can only form km sized fragments. However, such highly energetic collisions would be catastrophic enough to grind them down to centimetre-to-meter sized particles that are currently seen in Saturn's main rings or some vaporisation of particles might occur. Eventually, destruction or inelastic collisions leads to energy dissipation of random velocities and the cloud of debris would form a thin equatorial ring inside the planet's Roche limit. Note that direct simulations are still necessary, including such destruction and vaporisation, in order to understand more details about the final state of the ring systems. However, such complicated physics is beyond the scope of this paper and we will leave it to future works.\\ The above scenario seems to be very consistent with the satellite formation picture depicted in \cite{Cha11}, where an initial massive ring system spreads and gives birth to satellites at the Roche Limit. The silicate content of Saturn's satellites may come from big chunks of silicate initially implanted at random in the primordial ring system. The tidal disruption scenario may naturally form such big chunks of silicate. These chunks may also constitute the "silicate shards" of material thought to be embedded in the core of propeller structures observed in rings by Cassini \citep{Tis06, Por07}.\\ In conclusion, the tidal disruption of a passing large differentiated KBO could form massive rings around giant planets and could explain the compositional differences between the rings of Saturn and those of Uranus (and possibly those of Neptune too). However, the major caveat of this model is the question of angular momentum \citep{Esp91}. Rings and satellite systems of the four giant planets rotate in the prograde direction. However, encounters of KBOs with giant planets happened from random directions, so that the net total angular momentum should be close to zero in the case of a very large number of encounters (relevant for comet-sized objects). However, the number of encounters with Pluto-sized objects is quite small. So small number statistics may apply so that, by chance, only a couple of encounters may dominate the total angular momentum budget for a given planet. Therefore, having a system rotating in the prograde direction may mostly be a matter of chance. Whereas this explanation may hold for one planet, it is hard to believe that the same encounter's history happened for all giant planets. We think this problem might be solved by considering the effect of tides on retrograde objects. All fragments on retrograde orbits will experience a stong negative tidal torque from the giant planet leading to their orbital decay and crash into the planet. The elimination timescale depends strongly on the size of the fragments, but such information is not accurately provided by the current SPH simulations. However, if the tidal decay time is smaller than the collisional flattening time, this would imply that all retrograde material could be removed before the disk is formed in the equatorial plane. This process is far beyond the scope of our paper and thus we leave this matter to future work. One of the most striking features of giant planet ring systems is their diversity. Whereas Saturn's rings are massive, those of Jupiter, Uranus and Neptune are much less massive and show different structures. Our model suggests that all giant planets could have initially had massive rings as in \cite{Cri12}. However, they could have different subsequent evolution: as noted in \cite{Cha09b}, Saturn's ring is the only system with a synchronous orbit inside the Roche limit, whereas the other planets have opposite relative locations. This means that any satellite forming at the edge of the Roche limit of Jupiter, Uranus or Neptune should tidally decay inside the ring system and may push the rings inwards due to resonant interactions, leading to a faster removal of the ring system. This might explain the diversity of today's ring system even though they have all started from massive rings. However, the satellite might be tidally destroyed once it enters deep inside the Roche limit and additional ring material could be supplied. This process is also far beyond the scope of our paper and thus we leave this matter to future work. | 16 | 9 | 1609.02396 |
1609 | 1609.07266_arXiv.txt | In addition to their anomalous abundances, $^3$He-rich solar energetic particles (SEPs) show puzzling energy spectral shapes varying from rounded forms to power laws where the later are characteristics of shock acceleration. Solar sources of these particles have been often associated with jets and narrow CMEs, which are the signatures of magnetic reconnection involving open field. Recent reports on new associations with large-scale EUV waves bring new insights on acceleration and transport of $^3$He-rich SEPs in the corona. We examined energy spectra for 32 $^3$He-rich SEP events observed by ACE at L1 near solar minimum in 2007-2010 and compared the spectral shapes with solar flare signatures obtained from STEREO EUV images. We found the events with jets or brightenings tend to be associated with rounded spectra and the events with coronal waves with power laws. This suggests that coronal waves may be related to the unknown second stage mechanism commonly used to interpret spectral forms of $^3$He-rich SEPs. | Small $^3$He-rich solar energetic particle (SEP) events remain $\sim$40 years after their discovery one of the less explored phenomenon in solar physics. The anomalous abundances of $^3$He-rich SEPs have been explained by various mechanisms including gyro-resonant interactions with plasma wave turbulence \cite{fis78, mil98} or interactions with multiple magnetic islands \cite{dra12}. The energy spectra of major ion species in some $^3$He-rich SEP events are power laws or broken power laws but in other events $^3$He and Fe exhibit rounded spectra toward the low energies with $^3$He rollovers at $\sim$100-600\,keV\,nucleon$^{-1}$ and Fe rollovers below $\sim$100\,keV\,nucleon$^{-1}$ \cite{mas00, mas02}. It has been suggested that rounded spectra arise from the basic mechanism of $^3$He enrichment and power laws involve a further stage of acceleration. It is well known that power law spectra are produced by shock acceleration models (see e.g. \cite{ver15}, for a review) but shock detection is not typically reported in $^3$He-rich SEP events (e.g., via type II radio bursts). Notice that the spectral shape can deviate from power law when shock propagates into an inhomogeneous solar wind \cite{zan00}. $^3$He-rich SEP sources have been associated with EUV jets and jet-like CMEs \cite{kah01, wan06, nit06}. These are signatures of magnetic reconnection involving open field \cite{shi92} which may create turbulence and magnetic islands for a particle acceleration. With new observing capabilities on STEREO and SDO it has been recently shown that some $^3$He-rich SEP events were associated with large-scale coronal EUV waves \cite{nit15, buc15}. These waves were found without CMEs or only with weak coronal outflows. A common view on EUV waves is that they present or contain magnetosonic waves which can even steepen to the shocks (see \cite{war15}, for a review). The aim of this paper is to examine relation between energy spectral characteristics of $^3$He-rich SEPs and their flare signatures in the corona. | We examine energy spectra in 32 $^3$He-rich SEP events observed at L1 during solar minimum conditions in 2007-2010. The events have been examined in detail in our recent survey \cite{buc16}. Four events contain a high energy ($>$25\,MeV) but relatively low intensity solar proton component \cite{ric14}, which might not be typical $^3$He events but rather large gradual events or a mix of both. Three of these events show still very high $^3$He enrichment and all have high Fe/O ratio, typical of $^3$He-rich SEPs. Note that such cases if not identified may present one source of variability in $^3$He-rich SEP events. Table~\ref{tab1} lists all the investigated events. Column 1 gives the event number, column 2 the event start day, column 3 the $^3$He energy spectral form at $\sim$0.1-2\,MeV\,nucleon$^{-1}$, column 4 indicates whether the ion event had velocity dispersion, column 5 the EUV flare shape and column 6 the CME speed reported in SOHO/LASCO catalog. The EUV flares were examined using STEREO/EUVI \cite{how08}; the energetic ions were detected by ACE/ULEIS \cite{mas98} and/or ACE/SIS \cite{sto98}. Figure~\ref{fig1} shows an example of EUV wave associated with $^3$He-rich SEP event on 2010 February 19. Shown are running differences images at three times to illustrate the wave expansion. Figure~\ref{fig2} presents the event integrated $^4$He, $^3$He, O, Fe fluence energy spectra for 30 events. The spectra for two remaining events (2007-May-23 and 2008-Feb-4) have been shown in \cite{mas09}. The energetic ion fluences for events in 2009 were very weak. Specifically the ULEIS did not observe statistically significant $^3$He fluences in 2009 November 3 event though at higher energies the SIS observed a $^3$He-rich SEP event \cite{wie13}. \begin{figure} \begin{center} \includegraphics[width=\textwidth]{test3_run1.eps} \end{center} \caption{\label{fig1}STEREO-A 195 {\AA} EUV 2.5 min running differences images of the solar disk on 2010 February 18. Yellow arcs indicate the west limb from the Earth view and red pluses the L1 magnetic foot-point. At this time STEREO-A was separated from the Sun-Earth line by 65$^{\circ}$.} \end{figure} \begin{table} \centering \begin{threeparttable} \caption{\label{tab1}$^3$He-rich SEP events.} \begin{tabular}{clcccc} \br \#&Start day&$^3$He spectrum\tnote{a}&Dispersion&EUV flare\tnote{b}&CME\tnote{c}\\ \mr 1&2007-Jan-14&PL&no&B&...\\ 2&2007-Jan-24&R&yes&B&...\\ 3&2007-May-23&PL&no&W&544\\ 4&2008-Feb-4&?&no&B&...\\ 5&2008-Jun-16&?(PL)&?&J&...\\ 6&2008-Nov-4&?(PL)&yes&W&732\\ 7&2009-Apr-29&...&?&W&239\\ 8&2009-May-1&...(R)&?&W&...\\ 9&2009-Jul-5&...&?&W&yes\\ 10&2009-Oct-6&PL&?&B&...\\ 11&2009-Nov-3&...&?&W&yes\\ 12&2009-Dec-22&PL&no&W&318\\ 13&2010-Jan-16&PL&yes&J&...\\ 14&2010-Jan-27&?(PL)&yes&W&...\\ 15&2010-Jan-31&...(PL)&?&W&219\\ 16&2010-Feb-8&PL&no&W&yes\\ 17&2010-Feb-12&PL&no&W&509\\ 18&2010-Feb-19&R&yes&W/B&223\\ 19&2010-Mar-4&PL&yes&W&374\\ 20&2010-Mar-19&PL&yes&W&...\\ 21&2010-Jun-12&PL&no&W&486\\ 22&2010-Sep-1&R&no&W&1304\\ 23&2010-Sep-2&R&no&J&...\\ 24&2010-Sep-4&R&yes&B&...\\ 25&2010-Sep-17&?&no&J&yes\\ 26&2010-Oct-17&PL&yes&W&304\\ 27&2010-Oct-19&R&yes&B&385\\ 28&2010-Nov-2&R&yes&W&253\\ 29&2010-Nov-14&R&no&B&442\\ 30&2010-Nov-17&R&yes&J&639\\ 31&2010-Nov-29&R&?&W&505\\ 32&2010-Dec-10&PL&?&W&299\\ \br \end{tabular} \begin{tablenotes} \item[a] at 0.1-2\,MeV\,nucleon$^{-1}$; PL-power law, R-rounded, ?-unclear; the spectral form in parentheses is for Fe and extends down to 30\,keV\,nucleon$^{-1}$; spectra in events \#2-6, 14, 26-30 reported in \cite{wie00,mas09, buc15, nit15} \item[b] B-brighthening, J-jet, W-wave \item[c] CME speed in km\,s$^{-1}$ from LASCO catalog (cdaw.gsfc.nasa.gov/CME\_list); yes-CME observed with STEREO/COR-1 \end{tablenotes} \end{threeparttable} \end{table} \begin{figure} \begin{center} \includegraphics[width=\textwidth]{uleis_fluence_spectra_allnew2.eps} \end{center} \caption{\label{fig2}Energy spectra of $^4$He, $^3$He, O, Fe at $\sim$0.03-2\,MeV\,nucleon$^{-1}$ observed with ACE/ULEIS for 30 $^3$He-rich SEP events in 2007-2010.} \end{figure} We are able to assign the $^3$He spectral shape for 22 from 32 investigated $^3$He-rich SEP events (see Table~\ref{tab1}). In the remaining 10 events $^3$He spectral points show complex variations (5 events) or we have insufficient number of spectral points (2 points as a maximum). The power law spectrum has been found in 12 events, where 3 events are with jet/brightening and 9 with a coronal wave. The rounded spectral shape is found in 10 events, where 6 are with jet/brightening and 4 with a wave. Since in the two class scheme the Fe spectrum behaves as $^3$He, we examined the Fe spectrum in the cases where $^3$He is unclear. This adds 5 more events. We found 16 events with $^3$He or Fe having power laws where 4 events were with jet/brightening and 12 with a coronal wave. This confirms the previous finding depending solely on $^3$He spectrum. Including Fe we now have 11 events with rounded spectra where 6 are with jet/brightening and 5 with a wave, again consistent with the previous result. This shows that events with jet/brightening have a slight tendency to have rounded $^3$He spectra (6 of 9) and events with a wave to have power law $^3$He spectra (9 of 13). This is consistent with a recent study reporting similar tendency for 13 $^3$He-rich SEP events with jets and 4 events with EUV waves \cite{nit15}. Excluding 4 events with high energy solar proton components (\#12, 17, 21, 22), which are associated with EUV waves and all but one have power law $^3$He spectra, the results will not change dramatically. The $^3$He-rich SEP events with EUV waves have still a tendency to have power law $^3$He spectra (6 of 9) or power law $^3$He or Fe spectra (9 of 13). Exclusion of all events with a high energy proton component may be too conservative as for example one such event, 2009-Dec-22, have very high $^3$He and Fe enrichments (386 keV\,nucleon$^{-1}$ $^3$He/$^4$He$\sim$0.23, Fe/O$\sim$1.16). The 2 of 3 above events with the proton component which do not have an occulted source active region (AR) have reported type II radio bursts (ftp.ngdc.noaa.gov). The CMEs which accompany these solar proton events were generally not too fast (318, 486, 509, 544, and 1304 km\,s$^{-1}$ as reported by LASCO catalog). The slowest one in 2009-Dec-22 event was also associated with a type II radio burst. We get similar results if we consider coronal wave events only with slow CMEs ($<$350 km\,s$^{-1}$) or with no CME: $^3$He power law spectrum in 5 from 7 events and $^3$He (or Fe) power law in 7 from 10 events. In our earlier study \cite{buc15} we report rounded $^3$He, Fe spectra in the event with a weak EUV wave, probably without a significant influence on energetic ions. Therefore we look at the wave associated events with $^3$He rounded spectra (2010-Feb-19, 2010-Sep-1, 2010-Nov-2, 2010-Nov-29). The coronal wave in February 19 event was ejected about 30 min earlier, not at the time of the type III radio burst. Note this is the only case in this survey when the wave was not co-temporal with the flare and/or type III burst. Though overall the wave looks quite significant in the February 19 event (see Figure~\ref{fig1}), maybe it was ejected too early to have an effect on the later particle injection. The wave in the September 1 event had a large spatial scale but faded before reaching the L1 foot-point. The February 19 and September 1 events have all $^4$He, $^3$He, O, Fe spectra rounded, the feature has not been previously reported (see later). The November 2 and November 29 events are associated with quite narrow waves with less bright fronts. Furthermore the November 2 wave has the narrowest front in this study and the source AR in the November 29 event was extraordinary separated from the nominal magnetic connection. Thus the influence of the waves on energetic ion spectra in these two events would be minimal. We have also 3 events with power law $^3$He spectra (2007-Jan-14, 2009-Oct-6, 2010-Jan-16) associated with a brightening or a jet. The October 6 event spectrum could be still rounded as we do not have spectral points below 400 keV\,nucleon$^{-1}$. Other two events have one, the lowest energy (100 keV\,nucleon$^{-1}$) spectral point, deviated from the power law dependence but this can be an instrumental issue. It has been discussed that propagation through interplanetary space is probably not a cause of a common spectral shapes observed in $^3$He-rich SEP events \cite{mas02}. Indeed, the 2007 January 14 event is a short duration event with a non-dispersive onset while the 2010 January 16 event is a dispersive one, though both have similar, power law, spectral shapes. The 2009 October 6 event is too weak to observe a velocity dispersion. In this study about half of the events with rounded $^3$He spectra have a dispersive onset (6 of 10; 1 event with unclear dispersion) indicating nearly scatter-free propagation and half of the events with power law $^3$He spectra have non-dispersive onsets (6 of 12; 2 events are unclear). We found several $^3$He-rich SEP events which differ from the two class scheme introduced in \cite{mas00}. For example the following two events have only the $^3$He spectrum rounded, while the Fe is power law: the 2010 November 17 event and maybe also the 2010 November 14 event. Neither of these is associated with a coronal wave and both have the same solar source. Since we could not see this feature in other events (and no other reports are known) this may be due to specific conditions in a particular solar source region. However note on some other events where Fe shows a clear power law shape, $^3$He looks more complex (e.g., 2008-Nov-4, 2010-Jan-27) which might be of the previous type. Furthermore we found other events where all $^4$He, $^3$He, O, Fe spectra are rounded. Typical examples are the events of 2010 September 1 (already mentioned) and 2010 September 4. The September 1 event is particularly unusual, with the $^4$He spectrum continuing to rise at the lowest energies after the turnover. These two events are associated with different ARs: September 1 with a coronal wave and a high speed CME (with $>$25\,MeV proton event) and September 4 with a brightening. There is possibly another $^3$He-rich SEP event with all spectra rounded, the above mentioned 2010 February 19 event. Here after a turnover the Fe spectrum continues to rise at the lowest energies. This event was associated with a brightening but the wave ejected earlier was still present during the type III burst. Rounded $^4$He spectra have been predicted by a MHD turbulence model at lower ($<$20\,keV\,nucleon$^{-1}$) energies \cite{liu06} not covered by available instruments. | 16 | 9 | 1609.07266 |
1609 | 1609.07487.txt | The internal properties of stars in the red-giant phase undergo significant changes on relatively short timescales. Long near-{\orange un}interrupted high-precision photometric timeseries observations from dedicated space missions such as CoRoT and {\it Kepler} have provided seismic inferences of the global and internal properties of a large number of evolved stars, including red giants. These inferences are confronted with predictions from theoretical models to improve our understanding of stellar structure and evolution. Our knowledge and understanding of red giants have indeed increased tremendously using these seismic inferences, and we anticipate that more information is still hidden in the data. Unraveling this will further improve our understanding of stellar evolution. This will also have significant impact on our knowledge of the Milky Way Galaxy as well as on exo-planet host stars. The latter is important for our understanding of the formation and structure of planetary systems. | 16 | 9 | 1609.07487 |
||
1609 | 1609.01506_arXiv.txt | We present interferometric observations of CO lines ($^{12}$CO(1-0, 2-1) and $^{13}$CO(1-0, 2-1)) and dense gas tracers (HCN(1-0), HCO$^+$(1-0), HNC(1-0) and HNCO(4-3)) in two nearby edge-on barred lenticular galaxies, NGC~4710 and NGC~5866, with most of the gas concentrated in a nuclear disc and an inner ring in each galaxy. We probe the physical conditions of a two-component molecular interstellar medium in each galaxy and each kinematic component by using molecular line ratio diagnostics in three complementary ways. First, we measure the ratios of the position-velocity diagrams of different lines, second we measure the ratios of each kinematic component's integrated line intensities as a function of projected position, and third we model these line ratios using a non-local thermodynamic equilibrium radiative transfer code. Overall, the nuclear discs appear to have a tenuous molecular gas component that is hotter, optically thinner and with a larger dense gas fraction than that in the inner rings, suggesting more dense clumps immersed in a hotter more diffuse molecular medium. This is consistent with evidence that the physical conditions in the nuclear discs are similar to those in photo-dissociation regions. A similar picture emerges when comparing the observed molecular line ratios with those of other galaxy types. The physical conditions of the molecular gas in the nuclear discs of NGC~4710 and NGC~5866 thus appear intermediate between those of spiral galaxies and starbursts, while the star formation in their inner rings is even milder. | \label{sec:intro} Molecular clouds are the stellar nurseries of galaxies, and probing their physical properties in different galaxy types has the potential to answer many important open questions regarding star formation processes and galaxy evolution. Actively star-forming spiral galaxies, including our own Milky Way, are rich in cold gas and their molecular gas reservoirs have been studied for many years. However, since so-called 'red and dead' early-type galaxies (ETGs; lenticulars and ellipticals) are generally thought to be very poor in molecular gas, star formation within them is thought to have largely stopped. Nevertheless, roughly $10$ years after the first detection of molecular gas in external spiral galaxies \citep{r75,s75}, different phases of the interstellar medium (ISM) of ETGs were also studied, through observations of X-rays \citep[e.g.][]{fct85}, optical emission lines \citep[e.g.][]{col84}, H\,{\small I} \citep[e.g.][]{ktc85} and CO \citep[e.g.][]{w86,ws03,sa07}. \cite{y11} carried out the most extensive survey of molecular gas ($^{12}$CO(1-0)) in ETGs so far, in the $260$ galaxies of the volume-limited ATLAS$^{\rm 3D}$ sample\footnote{http://www-astro.physics.ox.ac.uk/atlas3d/} \citep{cap11}. They obtained a $22\%$ detection rate, with H$_2$ masses ranging from $10^7$ to $10^9$~$M_\odot$. The CO-rich ETGs in the ATLAS$^{\rm 3D}$ sample were further studied to probe the molecular gas properties in more details. Interferometric observations of $^{12}$CO(1-0) in $40$ objects were presented in \citet{al13}, revealing a variety of CO morphologies (discs, rings, bars, and spiral arms), with sizes smaller than in spirals in absolute terms but similar when compared to their optical extent \citep{d13}. The molecular gas kinematics is generally regular \citep{d13}, allowing one to easily probe the Tully-Fisher (luminosity-rotational velocity; \citealt{t77}) relation of ETGs \citep{d11}. However, gas-star kinematic misalignments indicate that the molecular gas has an external origin in at least one third of the systems, with significant field--cluster environmental differences \citep{dav11}. The molecular gas can also be used to study the physical conditions (temperature, density, column density, opacity, excitation mechanism, etc) within the dense cold gas of ETGs, where star formation takes place. For example, different transitions of a given molecule are good proxies for the gas temperature (e.g.\ $^{12}$CO(2-1)/$\,^{12}$CO(1-0)), isotopologues probe the gas optical depth and column density (e.g.\ $^{13}$CO/$\,^{12}$CO), and complex molecules (e.g.\ HCN, HCO$^+$, HNC, and HNCO) require much larger critical densities (up to $n_{\rm crit}\approx10^{6}$~cm$^{-3}$) to be excited compared to simpler ones (typically $n_{\rm crit}\approx10^{3}$~cm$^{-3}$). More subtle effects also exist. HCN and its isotopomer HNC trace respectively the warm-dense and cool-slightly less dense parts of a cloud, while HCO$^+$ traces even more tenuous regions \citep{hu95}. HCO$^+$ can also be enhanced in shocks associated with young supernova remnants (SNRs), due to cosmic rays (CRs) in the shocked material \citep{dic80,wo81,el83}, and is therefore also an important tracer of CR-dominated regions of the ISM. HNCO, on the other hand, is a good tracer of shocked gas \citep{mei05, mei12, rod10, ot14}, and it correlates well with SiO, a well-known shock tracer \citep{zin00}. Focusing on the physical conditions of the ISM through single-dish observations of several $^{12}$CO transitions, the $^{13}$CO isotopologue, and other molecules, \citet{ka10} and later \citet{c12} (see also \citealt{tim13}) found that the molecular line ratios of ETGs are generally similar to those of spirals and Seyferts, but different from those of starbursts and (ultra-) luminous infrared galaxies ((U)LIRGs). Interestingly, the line ratios are statistically correlated with several other ISM and stellar properties, e.g.\ the molecular-to-atomic gas ratio, dust temperature, dust morphology, K-band absolute magnitudes and stellar population age \citep{c12}. The $^{13}$CO(1-0)/$\,^{12}$CO(1-0) ratio of ETGs also seems to depend on environment \citep{ala15}. Here, we conduct interferometric observations of two typical edge-on lenticular galaxies (with large bulges and central dust lanes), NGC~4710 and NGC~5866 (morphological type $-0.9$ and $-1.2$, respectively; HyperLEDA\footnote{http://leda.univ-lyon1.fr/}). Both galaxies are fast rotating and their ionised gas is kinematically-aligned with the stellar kinematics, indicating that the gas is likely supplied by internal processes (e.g.\ stellar mass loss; \citealt{dav11}) or is left over from the galaxy formation event itself. NGC~4710 is a member the Virgo Cluster \citep{kra82} at a distance of $16.8$~Mpc \citep{t98}, while NGC~5866 is in a small group including two spirals at a distance of $15.3$~Mpc \citep{t98}. However, since NGC~4710 is located in the outskirts of the cluster and the distance between NGC~5866 and its nearest companion is rather large, both galaxies are unlikely to have had recent significant interactions with other galaxies or to have been affected by environmental effects. The general properties of the galaxies are listed in Table~\ref{tab:gprop}. NGC~4710 and NGC~5866 also happen to be CO-bright, unusual for supposedly 'red and dead' systems, allowing the first spatially-resolved study of multiple molecular tracers in an early-type galaxy. We map the entire discs of NGC~4710 and NGC~5866 in common low-$J$ CO lines such as $^{12}$CO(1-0), $^{12}$CO(2-1), $^{13}$CO(1-0) and $^{13}$CO(2-1), thus probing relatively tenuous molecular gas, as well lines of more complex molecules such as HCN(1-0), HCO$^+$(1-0), HNC(1-0) and HNCO(4-3) (the latter two lines being detected for the first time in those galaxies), thus probing denser gas. \begin{table} \begin{center} \caption{General properties of NGC~4710 and NGC~5866.} \begin{tabular}{@{}l l r r r r r@{}} \hline Galaxy & Property & Value & Reference \\ \hline NGC~4710 & Type & S0$_3$(9) & a\\ & RA~(J2000) & $12^{\rm h}49^{\rm m}38.8^{\rm s}$ & b\\ & Dec~(J2000) & $15^{\rm d}09^{\rm m}56^{\rm s}$ & b\\ & Distance (Mpc) & $16.8$ & c\\ & $\log(M_{{\rm H}_2}/M_\odot)$ & $8.72\,\pm\,0.01$ & d\\ & SFR$_{\rm 22\micron}$~($M_\odot$~yr$^{-1}$) & $0.11\,\pm\,0.02$ & e\\ & V$_{\rm sys}$~(km~s$^{-1}$) & $1102$ & f\\ & Major diameter & $4\farcm9$ & c\\ & Minor diameter & $1\farcm2$ & c\\ & Position angle & $207\degr$ & g\\ & Inclination & $86\degr$ & h\\ \hline NGC~5866 & Type & S0$_3$(8) & a\\ & RA~(J2000) & $15^{\rm h}06^{\rm m}29.5^{\rm s}$ & b\\ & Dec~(J2000) & $55^{\rm d}45^{\rm m}48^{\rm s}$ & b\\ & Distance (Mpc)& $15.3$ & c\\ & $\log(M_{{\rm H}_2}/M_\odot)$ & $8.47\,\pm\,0.01$ & f\\ & SFR$_{\rm 22\micron}$~($M_\odot$~yr$^{-1}$) & $0.21\,\pm\,0.04$ & e\\ & V$_{\rm sys}$~(km~s$^{-1}$) & $755$ & f\\ & Major diameter & $4\farcm7$ & c\\ & Minor diameter & $1\farcm9$ & c\\ & Position angle & $127\degr$ & g\\ & Inclination & $89\degr$ & h\\ \hline \end{tabular} \label{tab:gprop} \end{center} References: $^{\rm a}$ \citet{san94}; $^{\rm b}$ Nasa/Ipac Extragalactic Database (NED); $^{\rm c}$ \citet{t98}; $^{\rm d}$ \citet{y11}; $^{\rm e}$ \citet{tim14}; $^{\rm f}$ \citet{cap11}; $^{\rm g}$ \citet{dav11}; $^{\rm h}$ \citet{d11}. \end{table} The most abundant tracers reveal an X-shaped position-velocity diagram (PVD) in both galaxies, indicating the presence of an edge-on barred disc \citep[see][]{ba99,ab99,ma99}, with gas concentrated in a nuclear disc within the inner Lindblad resonance and in an inner ring around the end of the bar (corotation) and possibly farther out. Infrared observations also support a nuclear disc surrounded by a ring-like structure in the disc of NGC~5866. Indeed, \citet{xi04} found that the $6.75$ and $15$~$\micron$ emission in NGC~5866 peaks in the centre and at $\approx4$~kpc on either side of it (see their Fig.~10). The barred nature of NGC~4710 is consistent with its box/peanut-shaped bulge, but that of NGC~5866 is more surprising given its classical bulge. Should the ring-like structure in NGC~5866 have a different origin, however, the barred nature of NGC~5866 would have to be revisited. Nevertheless, we discuss our empirical and model results in light of these facts throughout. The first goal of our study is thus to exploit the variations of the molecular line ratios along the galaxy discs (as a function of projected radius and velocity), to study the physical properties of the molecular gas in each of those two dynamical components independently (i.e.\ nuclear disc and inner ring). We also do this quantitatively by modeling the molecular line ratios with a non-local thermodynamic equilibrium (non-LTE) code (RADEX; \citealt{van07}). A secondary goal is to compare the star formation activity in ETGs to that in other types of galaxies. For this, the observed line ratios are compared with those at the centre of spirals, starbursts, Seyferts and other lenticulars, as well as with those of some giant molecular clouds (GMCs) in the spiral arms and inter-arm regions of nearby galaxies \citep{so02,br05,baa08,ka10,c12}. The paper is divided as follows. Section~\ref{sec:obsredu} describes the observations and data reduction, while \S~\ref{sec:iman} presents the results and a basic analysis of the data. We discuss the line ratio diagnostics in \S~\ref{sec:rat}, both empirically and through modelling, the latter detailed in Appendix~\ref{sec:LVG}. A detailed discussion is presented in \S~\ref{sec:result} and we conclude briefly in \S~\ref{sec:conc}. | \label{sec:conc} Interferometric observations of tenuous ($^{12}$CO(1-0), $^{12}$CO(2-1), $^{13}$CO(1-0) and $^{13}$CO2-1)) and dense (HCN(1-0), HCO$^+$(1-0), HNC(1-0) and HNCO(4-3)) molecular gas tracers were presented for the edge-on lenticular galaxies NGC~4710 and NGC~5866. Our main conclusions are: \begin{enumerate} \item The PVDs of the CO lines are X-shaped and reveal that the gas is constrained to two bar-driven kinematic components in both galaxies, a nuclear disc (contained within the inner Lindblad resonance) and an inner ring (around corotation). Although brighter in the nuclear discs, the tenuous molecular gas is clearly detected in both kinematic components and is radially more extended than the dense gas, the latter being generally detected in the nuclear discs only. However, as suggested by the HCN(1-0) detection in the inner ring of NGC~5866, it is likely that the inner rings also contain dense gas below our detection thresholds. Both components appear clumpy, and no molecular gas is detected beyond the inner rings. \item A comparison of our interferometric data with published single-dish data (with much smaller primary beams) reveals that the latter were missing significant flux associated with the CO lines in the radially-extended inner rings. \item Molecular line ratios were probed empirically by studying the ratios of the PVDs of CO lines only, dense gas tracer lines only, and CO to dense gas tracer lines, as well as by extracting the integrated line intensity ratios of these same lines as a function of projected position along the galaxy discs, this for each kinematic component separately. The CO(1-0)/CO(2-1) ratios are smaller in the nuclear discs than the inner rings, suggesting that the nuclear discs have higher tenuous molecular gas temperatures. The $^{12}$CO/$^{13}$CO ratios are slightly larger in the nuclear discs, suggesting that the tenuous gas there has slightly smaller optical depths and column densities (at the very least in NGC~4710). The line ratios of the dense gas tracers only (detected only in the nuclear discs) are all larger than $1$ and HCN(1-0)/HCO$^+$(1-0) $<$ HCN(1-0)/HNC(1-0) $<$ HCN(1-0)/HNCO(4-3), suggesting that the environment is similar to PDRs, with a chemical enhancement of HCN via UV radiation from young massive OB stars and relatively few (supernova explosion-related) CRs. The ratios of CO to dense gas tracers (e.g.\ $^{12}$CO/HCN(1-0) or $^{13}$CO/HCN(1-0)) are significantly lower in the nuclear discs than in the inner rings, suggesting a higher fraction of dense gas there, possibly linked to a higher ambient pressure. Overall, the picture that emerges from these empirical line ratio diagnostics is that of nuclear discs that have a more inhomogeneous ISM, with more dense clumps immersed in a hotter and optically thinner molecular gas medium, consistent with a more intense star formation activity (conversely for the inner rings). \item LVG (RADEX) modeling was also carried out, considering a two-component molecular ISM traced by the CO lines only (tenuous component) and dense gas tracer lines only (dense component). The results are however inconclusive. The best-fit models within a single kinematic component often cover the entire range of parameters allowed by the models and are often driven to the edge of the model grid, presumably as a result of the shallowness and extent of the $\chi^2$ contours. The most likely model parameters have very large uncertainties (particularly for $T_{\rm K}$), due to the shape of the marginalised probability distribution functions. As such, they do not reveal clear differences between the inner rings and nuclear discs. This is exacerbated by the facts that most model results are upper limits in the inner ring of NGC~4710, and that we are unable to model the tenuous gas component in the disc of NGC~5866. \item We further compared the line ratios measured in NGC~4710 and NGC~5866 with those obtained in other galaxy types, revealing interesting contrasts. The CO(1-0)/CO(2-1) ratio is larger and thus the tenuous molecular gas temperature lower in NGC~4710 than in most other lenticular galaxies. The $^{12}$CO/$^{13}$CO ratios in NGC~5866 are larger than those in NGC~4710 and similar to those in starbursts, suggesting an optically thinner tenuous component similar to that in starbursts. The range of ratios in other lenticulars and Seyferts is however much larger than that in either NGC~4710 or NGC~5866. While the $^{12}$CO(1-0)/HCN(1-0) ratios in the nuclear discs of NGC~4710 and NGC~5866 are similar to those of starbursts near the lower end of the range for spirals, the ratios in the inner rings are larger and rather similar to those observed in some Seyferts and lenticulars near the upper end of the spiral range (where M31's GMCs are found). The $^{13}$CO(1-0)/HCN(1-0) ratios of both the inner rings and nuclear discs are however larger than those of starbursts (in that order), indicating smaller dense gas fractions. The HCN(1-0)/HCO$^+$(1-0) ratio is greater than unity everywhere in NGC~4710 and NGC~5866, as in most starbursts, while it is smaller than unity in spatially-resolved GMCs. High $^{13}$CO(1-0)/HCO$^+$(1-0) ratios in NGC~4710 and NGC~5866 further indicate relatively low HCO$^+$ enhancement (few CRs from supernova explosions) compared to that seen in starbursts and GMCs. Overall, the molecular line ratios in the nuclear discs of NGC~4710 and NGC~5866 thus suggest that the physical conditions of the molecular gas are intermediate between those of spiral galaxies and starbursts, with intense but not extreme star-formation activity, while the inner rings host even milder star formation. Interestingly, the star formation efficiency of both galaxies is much lower than that of normal spirals, suggesting that the line ratios are not sensitive to whatever is suppressing star formation. \end{enumerate} In summary, based on empirical line ratios, star formation feedback is likely to be stronger in the nuclear discs of NGC~4710 and NGC~5866 than in their inner rings, leading to hotter and optically thinner CO gas with a higher fraction of dense gas clumps. However, due to their large uncertainties, the most likely model results are unable to confirm this apparent dichotomy. As the resolution of our observations ($\approx300$~pc for the CO lines) is not enough to resolve individual GMCs ($<80$~pc), the physical conditions estimated either empirically or via LVG modeling are only averages over GMC associations. Higher angular resolution observations of high-$J$ CO lines and dense gas tracers with ALMA will ultimately resolve the GMCs, and therefore allow us to verify and expand these statements with much greater accuracy. | 16 | 9 | 1609.01506 |
1609 | 1609.06326_arXiv.txt | Neutrino masses and the number of light neutrino species can be tested in a variety of laboratory experiments and also can be constrained by particle astrophysics and precision cosmology. A conflict between these various results could be an indication of new physics in the neutrino sector. In this paper we explore the possibility for reconciliation of otherwise discrepant results in a simple model containing a light scalar field which produces Mass Varying Neutrinos (MaVaNs). We extend previous work on MaVaNs to consider issues of neutrino clumping, the effects of additional contributions to neutrino mass, and reconciliation of eV mass sterile neutrinos with cosmology. | \label{sec:intro} Over the past twenty years, definitive evidence for neutrino oscillations from a host of experiments has revealed that neutrino masses are nonzero. Because neutrino masses cannot be accounted for in the standard model (SM), this is a clue to physics beyond the SM. Since oscillation experiments are only sensitive to the difference in (squared) masses, the overall mass scale is not known and only two mass differences have been conclusively established, the so-called solar and atmospheric mass splittings, $\Delta m_{\odot}^2\simeq7.5\times10^{-5}~\rm eV^2$ and $\Delta m_{\rm atm}^2\simeq2.4\times10^{-3}~\rm eV^2$. It has also been well established that the neutrino mixing matrix, characterizing the mismatch between weak interaction and mass eigenstates, involves large mixing angles, in stark contrast to the quark sector. While they may seem like an uninteresting example of new physics---we have seen other chiral fermions obtain masses in the SM---neutrino masses could differ in a fundamental way from other fermion masses. Because neutrinos are not charged under electromagnetism, their mass generation mechanism could involve Majorana masses, violating lepton number, while the nonzero charges of the other fermions requires their masses to be of purely Dirac form. To generate neutrino masses in a way that does not disturb the successful picture we have of electroweak symmetry breaking, new neutrino states that are uncharged under the electroweak gauge group (or ``sterile'' neutrinos, as opposed to the ``active'' ones that carry electroweak charge) are typically invoked. Since they are gauge singlets, mass terms for these sterile neutrinos need not involve Higgs fields, which means that the mass scale in the sterile neutrino sector is largely a free parameter. While there may be theoretical bias for this scale to be very large compared to the weak scale, it is entirely possible and self-consistent that it is within reach of current experiments. Indeed, there are phenomenological reasons to consider a mass scale in the sterile neutrino sector as small as an eV. Alongside this standard three neutrino picture, there have been a number of experimental hints of neutrino oscillations characterized by a squared mass splitting of $\Delta m^2\sim{\cal O}\left(1~\rm eV^2\right)$ and a mixing angle $\theta\sim{\cal O}\left(0.1\right)$; these include short-baseline reactor experiments~\cite{Mueller:2011nm,*Huber:2011wv}, the flux of neutrinos from radioactive sources in gallium solar neutrino experiments~\cite{Acero:2007su,*Giunti:2010zu}, and electron (anti)neutrino appearance in muon (anti)neutrino beams~\cite{Aguilar:2001ty,*Aguilar-Arevalo:2013pmq}. To interpret these data in terms of neutrino oscillations requires an additional (sterile) neutrino around an eV and a large mixing angle with the active neutrinos. For detailed analyses, see, e.g.,~\cite{Kopp:2013vaa,*Gonzalez-Garcia:2015qrr}. It should be noted that there is generally tension between disappearance and appearance data, as a recent search for $\nu_\mu$ and $\bar\nu_\mu$ disappearance at IceCube~\cite{TheIceCube:2016oqi} shows, disfavoring the sterile neutrino interpretation of electron (anti)neutrino appearance data. However, a global fit including the IceCube results claims that a relatively large active-sterile mixing is allowed~\cite{Collin:2016aqd}. Additionally, progress in the direct search for neutrino masses has been ongoing. Searches using the endpoint in tritium $\beta$-decay currently limit the electron neutrino mass to less $2.05~\rm eV$ at 95\% CL~\cite{Aseev:2011dq}. The upcoming KATRIN~\cite{Robertson:2013ziv} and Project-8~\cite{Doe:2013jfe} experiments hope to probe masses down to about $0.1-0.2~\rm eV$ in the near future. Conceivably, the PTOLEMY experiment could use inverse beta decay to be sensitive to cosmological neutrinos~\cite{Weinberg:1962zza,Betts:2013uya,Long:2014zva}. In parallel with progress in neutrino measurements, cosmology has entered an era of impressive precision, enabling cosmological tests of physics beyond the standard model. We have direct observational evidence of the state of the Universe up to temperatures of a few MeV, corresponding to the time of neutrino decoupling and primordial nucleosynthesis (BBN). The precise picture of the Universe we now have at these temperatures and below allows for new physics below an MeV, even if weakly coupled, to be confronted with observation. Because (active or sterile) neutrinos interact very weakly with the rest of the Universe after decoupling--acting as a form of noninteracting radiation until they become nonrelativistic when they begin to act like dark matter--their observational consequences are relatively easy to understand. At early times, the cosmic microwave background (CMB), structure formation, and BBN are all sensitive to the energy density in neutrinos, which can be related to their masses and, in the case of sterile neutrinos, their mixing with the active neutrinos. The general agreement of these data with the standard cosmological picture based on three (essentially massless) neutrinos allows constraints to be placed on additional sterile neutrinos or on the masses of the active neutrinos. However, at late times, the neutrino energy density is the only SM component that can have nonstandard cosmology, which makes finding probes of this behavior crucial. In the case of a single massive sterile neutrino that is fully thermalized at early times, an up-to-date fit to cosmological observations give an upper bound on its mass (assuming the light, mostly active neutrinos' masses are negligible) of $0.53~\rm eV$~\cite{Archidiacono:2016kkh}. In the standard case of only three (active) neutrinos, the Planck analysis of only CMB data constrains the sum of the light neutrino masses to $0.675~\rm eV$~\cite{Ade:2015xua}. Including further cosmological data improves the bounds to $0.3~\rm eV$~\cite{Ade:2015xua,Thomas:2009ae,*RiemerSorensen:2011fe,*Zhao:2012xw}. Note that these upper limits are all at 95\%~CL. Improved observations could allow values of the sum of the active neutrino masses as small as $0.06~\rm eV$ to be probed~\cite{Feng:2014uja}. Sterile neutrinos which are much lighter than an eV and have similar abundance to the active neutrinos are disfavored by the Planck determination of $N_{\rm eff}$~\cite{Ade:2015xua}. At first glance, the null results from cosmological analyses are in strong tension with the sterile neutrino interpretation of the short baseline anomalies. In addition, the projected reach in the limit on the sum of the neutrino masses in the standard three neutrino scenario coming from cosmology seems to imply that current laboratory searches will not be sensitive enough to see nonzero neutrino masses. However, these conclusions rely on the assumption of a standard cosmological history. Thus, one should view contemporary terrestrial experiments seeking to measure neutrino masses or to test the sterile neutrino solution to short baseline anomalies as nontrivial probes of cosmology. Some scenarios that allow for cosmological observations to be compatible with eV mass neutrinos include coupling the sterile (with respect to the SM) neutrinos to a new U(1) gauge boson~\cite{Hannestad:2013ana,*Dasgupta:2013zpn,*Bringmann:2013vra,*Saviano:2014esa,*Chu:2015ipa,*Cherry:2016jol} or pseudoscalar~\cite{Archidiacono:2014nda,*Archidiacono:2016kkh}, or allowing the sterile neutrinos to be chiral under a new gauge group~\cite{Berezhiani:1995yi,*Vecchi:2016lty}. In this paper we will focus on the reconciliation of eV mass neutrinos with cosmology via the dependence of the neutrino masses and mixing angles on the expectation value of a non constant light scalar field. This possibility was originally motivated to address the puzzle of dark energy~\cite{Fardon:2003eh,Fardon:2005wc}, but we will consider this possibility more generally, including the possibility of additional contributions to neutrino mass, the effects of neutrino clustering, and models which do not give dark energy. We will consider two scenarios. In~\S~\ref{sec:logpotential} we consider a MaVaN model containing a light scalar field with a logarithmic potential, and find parameters such that eV mass sterile neutrinos with sizable mixing angles are allowed today which were always heavy enough at earlier times so that these states were never populated in the early universe and have no observable effect on cosmology. In~\S~\ref{sec:mass} we consider a scenario which allows the observed, active neutrinos to have a mass which is today around an eV. Because the masses were much lighter at high redshift, cosmological observations indicate a much smaller mass. In~\S~\ref{sec:susy} we discuss a cosmologically viable supersymmetric MaVaN scenario which could allow eV mass sterile neutrinos to appear in terrestrial experiments. | \label{sec:conclusions} In this paper we have described several situations involving mass-varying neutrinos that allow for either active neutrinos or sterile neutrinos with a large active-sterile mixing to have masses around an eV and yet still be compatible with strong limits from cosmological observations. Along with ``secret'' neutrino interactions~\cite{Hannestad:2013ana,*Dasgupta:2013zpn,*Bringmann:2013vra,*Saviano:2014esa,*Chu:2015ipa,Archidiacono:2014nda,Archidiacono:2016kkh}, this possibility illustrates the necessity of combining cosmological probes of neutrino properties with terrestrial experiments. Combining both probes allows us to {\it test} whether neutrinos have richer structure than expected in ways that cosmological observations or terrestrial experiments alone cannot. MaVaNs are motivated by attempts to understand dark energy. It is interesting that they can also modify neutrino cosmology to allow recent hints for eV-scale sterile neutrinos to be reconciled with cosmological observations, or for the active neutrinos to have masses within the reach of near future experiments today. In addition, it is worth mentioning that unlike the case of ``secret'' neutrino interactions, the (mostly) sterile neutrinos in this scenario with a logarithmic potential are kinematically forbidden from being produced at late times, and therefore do not suffer from the problem that they are produced in late-time collisions, upsetting agreement with cosmological observations~\cite{Mirizzi:2014ama}. The requirement of a light scalar field for the MaVaNs scenario suggests that this sector could be supersymmetric. Properly supersymmetrizing the theory adds additional constraints and requires introducing very weak couplings of this light field to charged fermions in order to make the sterile neutrinos heavy at early times. Besides cosmological observations and neutrino experiments, these weak couplings offer perhaps the best way of testing the scenario, through, e.g., searches for fifth forces at large distance scales~\cite{Adelberger:2003zx} or electron-density--dependent neutrino masses~\cite{Kaplan:2004dq,Zurek:2004vd}. Interactions between light scalars and active neutrinos can prevent active neutrinos from freely streaming and lead to observable signatures in the CMB~\cite{Friedland:2007vv,*Basboll:2008fx,*Smith:2011es,*Cyr-Racine:2013jua,*Archidiacono:2013dua}. In the scenarios we have considered, the active neutrino-scalar coupling is too small to be constrained by these considerations. It would be interesting to study such CMB signatures and their complementarity with other terrestrial observables in nonstandard neutrino scenarios. We have extended earlier work on MaVaNs to take into account the effects of neutrino clustering, to add additional contributions to neutrino mass and additional couplings in the scalar sector. We have been able to exhibit models in which eV mass sterile neutrinos appear in present day neutrino oscillation experiments but do not affect precision cosmology, and in which active neutrinos could have eV scale masses today but not in the early universe. Neutrino masses, inflation, dark matter, baryogenesis, and dark energy all show that there must be new physics beyond the standard model. We do not know the energy scale, but neutrino masses and dark energy indicate a physical scale of order $10^{-4}-10^{-2}$ eV. It would be remiss of us to simply adopt theoretical prejudice and assume that such new physics does not involve any new light particles. If there is such new physics, then precision cosmology and laboratory measurements are not necessarily simply different ways of measuring the same neutrino properties. As an illustration of the possibilities, in this paper we have explored models containing light sterile neutrinos coupled to a light scalar in which laboratory and cosmological measurements which would give seemingly inconsistent results can instead be interpreted as evidence for a new light sector, which could be the origin of dark energy. \appendix\label{appendix} | 16 | 9 | 1609.06326 |
1609 | 1609.01992_arXiv.txt | {The majority of bright extragalactic $\gamma$-ray sources are blazars. Only a few radio galaxies have been detected by \textsl{Fermi}/LAT. Recently, the GHz-peaked spectrum source PKS\,1718--649 was confirmed to be $\gamma$-ray bright, providing further evidence for the existence of a population of $\gamma$-ray loud, compact radio galaxies. A spectral turnover in the radio spectrum in the MHz to GHz range is a characteristic feature of these objects, which are thought to be young due to their small linear sizes. The multiwavelength properties of the $\gamma$-ray source \object{PMN\,J1603--4904} suggest that it is a member of this source class.} {The known radio spectrum of \pmn can be described by a power law above 1\,GHz. Using observations from the Giant Metrewave Radio Telescope (GMRT) at 150, 325, and 610\,MHz, we investigate the behavior of the spectrum at lower frequencies to search for a low-frequency turnover.} {Data from the TIFR GMRT Sky Survey (TGSS ADR) catalog and archival GMRT observations were used to construct the first MHz to GHz spectrum of \pmn.} {We detect a low-frequency turnover of the spectrum and measure the peak position at about 490\,MHz (rest-frame), which, using the known relation of peak frequency and linear size, translates into a maximum linear source size of $\sim$1.4\,kpc.} {The detection of the MHz peak indicates that \pmn is part of this population of radio galaxies with turnover frequencies in the MHz to GHz regime. Therefore it can be considered the second confirmed object of this kind detected in $\gamma$-rays. Establishing this $\gamma$-ray source class will help to investigate the $\gamma$-ray production sites and to test broadband emission models. } | \label{sec:intro} Most sources detected at $\gamma$-ray energies by the \textsl{Fermi} Large Area Telescope (LAT) are active galactic nuclei (AGN), in particular blazars, which are extragalactic jets observed at small angles to the line of sight \citep{3fgl}. Only a few, nearby non-blazar sources are detected \citep[see also, e.g.,][]{Abdo2010_misaligned,3fgl}. In contrast to blazars, jets in radio galaxies or misaligned sources are seen at larger inclination angles, hence, they are less or not at all relativistically beamed. A typical broadband spectrum can be explained by inverse Compton or hadronic processes in the jet \citep[see, e.g.,][and references therein]{Massaro2016_Fermireview}. The location and mechanism of the high-energy emission are still subject to much debate. Possible origins include core, parsec-scale jets, and the kiloparsec-scale lobes. In nearby sources \citep[Centaurus~A and Fornax~A;][]{Abdo2010_CenAlobes,Ackermann2016_FornaxA} the $\gamma$-ray emission from the lobes can be directly imaged with the \textsl{Fermi}/LAT. The smaller versions of evolved radio galaxies are of a few kpc or less in size \citep[e.g.,][]{Readhead1996b,Readhead1996a,Odea1998,Kunert-Bajraszewska2010}. Since their morphologies are reminiscent of the evolved double sources, they are called compact symmetric objects' ($\lesssim$\,1\,kpc, CSOs) or medium-sized symmetric objects' ($\lesssim$\,1-15\,kpc, MSOs). Kinematic measurements of their compact lobes suggest that these sources are younger than full-sized radio galaxies \citep[e.g.,][]{Owsianik1998,An2012}. An alternative explanation of the smaller extensions is the scenario of frustrated jets, i.e., the source is confined owing to the interaction with a dense, surrounding medium \citep{Bicknell1997,Carvalho1998}. Thus, these sources play an important role in the study of AGN evolution and the interaction of AGN jets with the ambient medium. Typically, CSOs show a turnover or cutoff in their radio flux-density spectrum around 1\,GHz, where the spectrum changes from flat or steep at higher frequencies to an inverted spectrum. The peak of MSOs is typically lower, in the upper megahertz range. The convex spectral shape of their radio spectrum is one of the defining properties of these small radio sources. Based on the peak frequency $\nu_\mathrm{peak}$, one refers to MHz-peaked spectrum (MPS) or compact steep spectrum (CSS) sources for $\nu_\mathrm{peak}<1$\,GHz and GHz-peaked sources (GPS) for $\nu_\mathrm{peak}>1$\,GHz \citep{Odea1998}. The reason for this turnover could be either due to synchrotron self-absorption \citep[SSA; e.g.,][]{Snellen2000} by relativistic electrons from the emitting source, or free-free absorption due to an external dense medium \citep[FFA; e.g.,][]{Bicknell1997}. Well-sampled radio spectra are necessary to identify the underlying absorption mechanism \citep[e.g.,][]{Callingham2015}. Several studies found that the linear size $LS$ of the radio morphology anticorrelates with the rest-frame peak frequency of the radio spectrum \citep{Fanti1995,Odea1997,Odea1998}, \begin{equation}\label{eq:Odea} \log\nu_\mathrm{peak} \approx -0.21(\pm 0.05) - 0.65(\pm0.05)\log LS ,\end{equation} with LS in kpc and $\nu_\mathrm{peak}$ in GHz. This relation indicates that the mechanism causing the spectral turnover is related to the source size and is explained well in the SSA scenario. Therefore, a typical GPS radio galaxy has an extent of less than 1\,kpc. A typical MPS or CSS is up to 20\,kpc in size according to this relation. Evolved radio galaxies have been established as $\gamma$-ray loud objects \citep[e.g.,][]{Abdo2010_misaligned}. Theoretical broadband emission models also \citep[][]{Kino2009,Kino2011,Kino2013,Stawarz2008,Ostorero2010} predict $\gamma$-ray emission from compact, likely young, radio galaxies. Because of the interaction of the evolving radio source and the interstellar medium, high-energy emission can be produced through the non-thermal inverse Compton process or thermal bremsstrahlung. The interaction of particles in the lobes with the ambient medium is expected to dominate the emission, while the contribution of the jet is smaller because of Doppler deboosting \citep{Stawarz2008}. \citet{Kino2009} predict thermal bremsstrahlung $\gamma$-ray emission in the initial phase of expansion at a detectable level for \textsl{Fermi}/LAT observations of nearby young radio galaxies. Two CSS sources, 3FGL J1330.5+3023 (3C 286) and 3FGL J0824.9+3916 (4C +39.23B), were reported in the latest \textsl{Fermi}/LAT AGN catalog \citep[3LAC,][]{3lac}. However, GPS sources are still not confirmed as a $\gamma$-ray bright source class. Only recently, \citet{Migliori2016} reported of the first detection of an established CSO/GPS, \object{PKS\,1718--649}. This result follows the discussions of a few $\gamma$-ray CSO-candidates detected by \textsl{Fermi}/LAT \citep[see also][for a summary]{DAmmando2016}: 4C+55.1 \citep[][]{McConville2011}, PKS\,1413+135 \citep[][]{Gugliucci2005}, 2234+282 \citep{An2016}, and PMN\,J1603--4904 \citep{Mueller2014a}. An extensive multiwavelength study of \pmn \citep{Mueller2014a,Mueller2015a}, conducted within the framework of the TANAMI program \citep{Ojha2010a,Kadler2015}, suggested it was not a blazar. Very Long Baseline Interferometry (VLBI) observations indicate a CSO-like morphology. The kinematic study on timescales of $\sim$15\,months shows no significant motion of the eastern and western component. Lower resolution radio observations with the ATCA ($>$1\,GHz) indicate extended, unresolved emission at sub-arcsecond scales. X-ray observations with \textsl{XMM-Newton} and \textsl{Suzaku} measure an emission line in the X-ray spectrum on top of an absorbed power-law emission. Together with the optical-spectroscopic redshift of $z=0.23$ from X-shooter observations \citep{Goldoni2016} this X-ray feature could be interpreted as emission from a highly ionized plasma, but the lack of a neutral iron line is puzzling. The \textsl{Fermi}/LAT counterpart 3FGL\,1603.9$-$4903 (with 95\%-confidence semimajor axis of $\sim$0.8\,arcmin)\footnote{A false association of the $\gamma$-ray source with PMN J1603–4904 is still possible, however, as pointed out in \citet{Mueller2015a} a more exotic explanation for the $\gamma$-ray emission origin would be required.} shows a hard $\gamma$-ray spectrum but with a spectral break above 50\,GeV and only mild variability \citep{1fhl,3fgl,2fhl}. These broadband properties led to the conclusion that \pmn is either a very peculiar blazar or a compact, possibly young, radio galaxy. Opposite motion of the radio lobes and a turnover in the radio spectrum would hence be expected and would further confirm the CSO nature. Here we present a detailed analysis of the radio spectrum of \pmn, extending the existing information down to 150\,MHz. | \label{sec:conclusions} We have presented the first MHz to GHz spectrum for the $\gamma$-ray loud extragalactic jet source \pmn. The radio spectrum clearly shows a rest-frame spectral turnover at $\sim 490$\,MHz. Using the established anticorrelation of peak frequency and linear size, we determine the linear size of \pmn to be $\sim 1.4$\,kpc. With VLBI, only the inner $\sim$40\,pc of the small radio source are detected \citep{Mueller2014a}. The spectral index of the optically thin emission is comparably flat for GPS or CSS sources \citep{Odea1998,Fanti2001}, and is likely due to the bright compact emission. The spectral shape can be explained by the superposition of different components. Both, the SSA and FFA models can describe the overall spectrum. The turnover can hence be attributed either to self-absorption or to external absorption by a dense medium. The sparse spectral coverage below 1\,GHz does not allow the testing of more complex models or distinguishing between SSA and FFA. More sensitive observations with southern telescopes like the upgraded GMRT \citep{RaoBandari2013} or the Murchinson Widefield Array \citep[MWA;][]{MWA}, covering a wider frequency range in the MHz regime would be required. Higher resolution observations with sufficient sensitivity could allow us to image the extended emission. Currently, only ALMA is capable of addressing these scales at this low declination. Because of its multiwavelength properties, \pmn has been discussed as a possible $\gamma$-ray loud young radio galaxy. With the observations reported here, this classification is supported by the detection of an MPS-like radio spectrum. \pmn adds to the class of so far very rare extragalactic jets that show a spectral turnover in the MHz to GHz range, and are detected at $\gamma$-rays. As it is the second confirmed object of this type, following \pks \citep{Migliori2016}, there are some noteworthy differences. First, \pks is a nearby galaxy (z=0.014) in contrast to \pmn at z=0.23. Furthermore, \pmn has a hard GeV spectrum with a photon index $\Gamma\sim2$ \citep[][]{1fgl,2fgl,3fgl}, while for \pks a $\Gamma\sim2.9$ was found \citep{Migliori2016}, that is more comparable with the spectrum of the Cen~A lobes \citep{Abdo2010_CenAlobes}. Hence, as also discussed by \citet{Migliori2016}, this could indicate significant differences in the intrinsic physical mechanism producing $\gamma$-rays. This is also suggested by the VLBI structure of both sources: while \pks has an irregular lobe-dominated morphology \citep{Tingay2003,Ojha2010a}, \pmn is a more symmetric, core dominated source. Future \textsl{Fermi}/LAT observations using the Pass8 analysis \citep{Atwood2013_Pass8} will have the required sensitivity to detect further GPS and CSS sources at $\gamma$-rays. By establishing this source class and its defining properties one will be able to test theoretical emission models \citep[e.g.,][]{Kino2007,Kino2009,Stawarz2008}. For \pmn, modeling of the broadband SED will help us to understand the high-energy emission mechanism. The high Compton dominance \citep[see Fig.~7 in][]{Mueller2014a} and the flat $\gamma$-ray spectrum challenges current models. | 16 | 9 | 1609.01992 |
1609 | 1609.03997_arXiv.txt | CMB Stage-4 experiments will reduce the uncertainties on the gravitational lensing potential by an order of magnitude compared to current measurements, and will also produce a Sunyaev-Zel'dovich (SZ) cluster catalog containing $\sim10^{5}$ objects, two orders of magnitudes higher than what is currently available. In this paper we propose to combine these two observables and show that it is possible to calibrate the masses of the full Stage-4 cluster catalog internally owing to the high signal to noise measurement of the CMB lensing convergence field. We find that a CMB Stage-4 experiment will constrain the hydrostatic bias parameter to sub-percent accuracy. We also show constraints on a non parametric $Y-M$ relationship which could be used to study its evolution with mass and redshift. Finally we present a joint likelihood for thermal SZ (tSZ) flux and mass measurements, and show that it could lead to a $\sim5\sigma$ detection of the lower limit on the sum of the neutrino masses in the normal hierarchy ($\sum m_{\nu}=60 \textrm{meV}$) once combined with measurements of the primordial CMB and CMB lensing power spectra. | \label{sec:intro} The number of galaxy clusters as a function of mass and redshift is a prediction of the LCDM model, and by accurately reconstructing the cluster mass function we can put constraints on cosmological parameters such as the matter density $\Omega_m$, the sum on the neutrinos masses $\sum m_{\nu}$ and the normalisation of the linear matter power spectrum $\sigma_{8}$. This observable has recently gained a vivid interest following the publication of the Planck cluster catalog ($\approx 10^{3}$ clusters) and the slight discrepancy between cosmological parameters inferred from the primary CMB and from the distribution of cluster masses \cite{2015arXiv150201598P,2015arXiv150201597P}. A classical method to estimate cluster masses is to use measurements of the Compton-$y$ parameter, a measurement of the integrated flux of the thermal Sunyaev-Zel'dovich effect at the position of the cluster. The size of the effect is proportional to the total thermal energy of the cluster gas and is therefore correlated with cluster mass \cite{2008ApJ...675..106B, 2012ApJ...754..119M,2013ApJ...772...25S}. This $Y-M$ scaling relation has traditionally been determined empirically from X-ray observations of clusters \cite{2010A&A...517A..92A, 2014A&A...571A..20P}. However, X-ray-inferred cluster masses rely on the assumption that clusters have reached hydrostatic equilibrium, and departure from this equilibrium can bias the estimated cluster masses. Physical phenomena causing this departure include bulk motions in the gas or non-thermal sources of pressure (such as magnetic fields or cosmic rays). Numerical simulations have shown that this can lead to an underestimation of the true cluster masses by 10 to 15 $\%$ \cite{2008A&A...491...71P,2007ApJ...668....1N,2010A&A...514A..93M}. Moreover, instrumental systematics in the X-ray analysis could propagate into the cosmological results \cite{2012MNRAS.426.2046A}, and an independent method for calibrating the scaling relation is extremely valuable. Following the first detections of CMB lensing by clusters \cite{2015PhRvL.114o1302M, 2015ApJ...806..247B}, a new method for self-calibrating the cluster masses using measurements of the lensing convergence at the cluster positions has recently been proposed \cite{2015A&A...578A..21M}. It has been demonstrated on simulations and successfully applied to Planck data resulting in a $5 \sigma$ measurement of the hydrostatic bias parameter \cite{2015arXiv150201597P}. The aim of this work is to discuss extensions of this method in the era of CMB Stage-4 (S4), a next-generation CMB experiment that will achieve a cosmic-variance-limited reconstruction of the convergence field up to multipoles $\ell \sim 1000$. This paper is structured as follows. In Section \ref{sec:method} we describe our cluster lensing model as well as a maximum likelihood estimator for the cluster masses from a lensing convergence map. We also discuss the impact of possible foregrounds contamination and atmospheric noise. In Section \ref{sec:calibration} we propose two parametric methods to calibrate the $Y-M$ relationship using cluster lensing. First we forecast constraints on the hydrostatic parameter following the method proposed in \cite{2015A&A...578A..21M}, and then extend the formalism and forecast constraints on more general scaling relations, including a non-parametric model that can be used to study the mass and redshift dependence freely. In Section \ref{sec:cosmopar} we study how the availability of joint tSZ and lensing mass measurements improves the cosmological constraints achievable by a cluster survey carried out with S4 by consistently accounting for the uncertainties in the $Y-M$ scaling relation. We summarise our main conclusions in Section \ref{sec:conclusion}. Throughout this paper we adopt a fiducial cosmology with $\Omega_{m}=0.315$, $\Omega_{b}=0.049$, $\Omega_{\Lambda}= 0.685$, $H_{0}=67\, {\textrm{km s}^{-1}\textrm{Mpc}^{-1}}$, $A_s=2.2\times10^{-9}$, $n_s=0.96$ and $\tau=0.06$, compatible with \cite{2015arXiv150201589P}. We will use cluster masses $M_{500}$ defined as the mass measured within a radius $R_{500}$ that encloses a mean density $500$ times larger than the critical density at the cluster redshift. We will also estimate the number density of haloes as a function of mass using the measurements of the mass function by \cite{2008ApJ...688..709T}. | \label{sec:conclusion} We have studied the potential of using CMB lensing by clusters to calibrate their masses in the era of CMB Stage-4 experiments. We have found that CMB S4 will allow a sub-percent determination of the hydrostatic bias parameter relating the mass inferred from X-ray observation to the true cluster mass. We have then extended the model and shown that the large number of detected clusters and the low uncertainties in the reconstructed convergence maps could be used to calibrate the $Y-M$ relationship using solely CMB data. We have also studied the constraints on a non-parametric reconstruction of the $Y-M$ relationship which can be used to study the evolution of this relation with mass and redshift with high significance over a wide range of masses. Throughout the paper we have studied the impact of discarding temperature data in the reconstruction of the convergence field. This is an important comparison, since the temperature data from Stage-4 CMB experiments will suffer from atmospheric contamination and residual foregrounds. Finally we have presented a joint likelihood for tSZ and lensing mass measurements allowing us to forecast constraints on cosmological parameters while consistently accounting for the uncertainties in the $Y-M$ scaling relation. The method presented here relies on a number of assumptions. Testing and characterising the effect of each of these assumptions is the subject of future work but it is worth quoting the following caveats, which might affect future analyses with real data: \begin{itemize} \item Throughout this paper, we have modelled the lensing field and the lensing reconstruction noise as Gaussian fields. At the level of precision achieved by CMB Stage-4 experiments this assumption may break down \cite{2012PhRvD..86l3008B,2016arXiv160803169L}, and the non-Gaussian contribution to the signal and/or the noise should be studied in detail. \item We used a Poisson likelihood for cluster number counts. While this is commonly used for current cluster catalogs, the high number of clusters detected in the CMB S4 catalog might require a more sophisticated treatment. Studies of possible departure from Poisson for cluster number count can be found in \cite{2003ApJ...584..702H, 2004PhRvD..70d3504L, 2011MNRAS.418..729S}, where a sample variance term is added to the Poisson shot-noise term to account for the fact that cluster are peaks of the same underlying density field. In our analysis we have used wide redshift bins to reduce the impact of sample variance. \item We have not included the covariance between the CMB lensing power spectrum and cluster counts calibrated using CMB lensing when constraining cosmological parameters. In practice, the full covariance could be constructed from Monte-Carlo simulations while analysing CMB S4 data (e.g. see \cite{2016arXiv160105779K}). \item The cluster mass function used to determine cosmological parameters is a fit to N-body simulations \cite{2008ApJ...688..709T}. In order to achieve the accuracy required for S4, in particular for non-standard scenarios such as massive neutrinos \cite{2013JCAP...12..012C}, a better understanding of the theoretical uncertainties in the mass function is necessary, in particular regarding the effects of baryonic physics \cite{2012MNRAS.423.2279C}. \item Finally, we have assumed knowledge of the cluster mass \cite{1996ApJ...462..563N} and pressure \cite{2010A&A...517A..92A} profiles, and this assumption allows us to devise a minimum-variance estimate of $Y$ and $M$ from the data. However, this needs to be validated using hydrodynamic simulations at the level of precision corresponding to the uncertainties on cosmological parameters. A more careful analysis of cluster de-blending is also necessary \end{itemize} While more work need to be done to address each of these issues, we have shown that using cluster lensing to calibrate cluster masses has an important potential and could play a role in the future determination of the sum of neutrinos masses. This, combined with particle physics measurements, could allow us to distinguish between the normal and inverted hierarchies, thus opening a new window on fundamental physics. | 16 | 9 | 1609.03997 |
1609 | 1609.06056_arXiv.txt | In protoplanetary discs, planetary cores must be at least 0.1~$\me$ at 1 au for migration to be significant; this mass rises to 1~$\me$ at 5~au. Planet formation models indicate that these cores form on million year timescales. We report here a study of the evolution of 0.1~$\me$ and 1~$\me$ cores, migrating from about 2 and 5~au respectively, in million year old photoevaporating discs. In such a disc, a gap opens up at around 2~au after a few million years. The inner region subsequently accrete onto the star on a smaller timescale. We find that, typically, the smallest cores form systems of non--resonant planets beyond 0.5~au with masses up to about 1.5~$\me$. In low mass discs, the same cores may evolve {\em in situ}. More massive cores form systems of a few earth masses planets. They migrate within the inner edge of the disc gap only in the most massive discs. Delivery of material to the inner parts of the disc ceases with opening of the gap. Interestingly, when the heavy cores do not migrate significantly, the type of systems that are produced resembles our solar system. This study suggests that low mm flux transition discs may not form systems of planets on short orbits but may instead harbour earth mass planets in the habitable zone. | \label{sec:intro} Before the discovery of extrasolar planets, explaining the formation of terrestrial planets in our solar system was already a challenging task. The detection of a large variety of planetary systems containing Earth--mass or super--Earth planets by the {\em Kepler} satellite has made the task even more arduous. Terrestrial mass planets are usually considered to form either {\em in situ} or through inward migration. {\em In situ } formation was the early scenario proposed for forming planets in our solar system (Wetherill 1988, Lissauer 1993 and references therein), and has been put forward as a way to explain the {\em Kepler} candidates (Hansen \& Murray 2013). However, it has been argued that this model is not consistent with the distribution of solids in discs (Raymond \& Cossou~2014, Schlichting~2014). Also, {\em in situ} formation models have difficulties forming cores of giant planets before the gas dissipates (Thommes et al. 2003, Chambers 2016). Formation through inward migration is an efficient way of obtaining tight systems of super--Earths with short periods (Terquem \& Papaloizou 2007, Haghighipour 2013 and references therein), but fails to explain terrestrial planets similar to those in our solar system. However, migration of planets in dissipating discs has recently been studied (Coleman \& Nelson 2014, Cossou et al. 2014, Coleman \& Nelson 2016) and seems to offer a way of combining the advantages of both models for forming terrestrial planets. Coleman \& Nelson (2014, 2016) studied the evolution of systems containing initially 36 or 52 cores with masses of $0.3$ or $0.1 \; \me$, respectively, spread between 1 and 20~au. In addition, there were thousands of planetesimals with masses 10, 20 or 50 times smaller than that of the embryos and distributed in between the planets. The disc had a surface gas density at least equal to that of the minimum solar mass nebula, i.e. $\Sigma \propto r^{-3/2}$ and $\Sigma=1.7 \times 10^3 \; \gcm$ at 1~au, and was subject to photoevaporation and viscous evolution. They found that terrestrial--mass planets and super--Earths formed in the discs with the lowest masses. In the models with small abundance of solids and large planetesimals, growth was found to be limited so that migration was inefficient. Such models resulted in systems of low mass planets spread out through the disc and in which mean motion resonances were destroyed after the disc dissipated. Cossou et al. (2014) started with cores with masses between 0.1 and 2~$\me$ spread between 1 and 20~au. The total mass in the cores was between 21 and 84~$\me$. They considered a disc with a surface gas density of $300 \; \gcm$ at 1~au and $\Sigma \propto r^{-1/2}$. In some of their simulations, the disc's mass was decreased exponentially to mimic disc's viscous evolution and photoevaporation. Systems of hot super--Earths in mean motion resonances were produced in discs which were not dissipating, whereas the systems were more spread out and not in resonances when dissipation was included. In these studies, all types of planets formed from the same parent population and, in general, the planetary systems that were produced at around 1~au contained planets more massive than the terrestrial planets in our solar system. In this paper, we consider a model where different parent populations exist at different locations and the disc is photoevaporating. In typical protoplanetary discs, the type--I migration timescale becomes comparable to the disc lifetime for cores at least as massive as 0.1 and 1~$\me$ at 1 and 5~au, respectively. Planet formation models suggest that it takes at least 1~Myr for 0.1~$\me$ cores to form at $\sim 1$~au. More massive cores, with a mass $\sim 1 \; \me$, may form on this timescale beyond the snow line, at around 5~au. This prompts us to consider a model where the initial conditions are a population of 0.1~$\me$ cores between 1 and 5~au and a population of 1~$\me$ cores beyond 5~au, which start migrating when the disc is $\sim 1$~Myr--old. The total mass in these populations of cores is set by the initial surface density in the disc. We evolve these initial populations in a disc which undergoes a transition due to X--ray photoevaporation. Transition discs are defined as discs lacking emission in the near--infrared, which means that there is a (large) hole in the dust distribution in their inner parts. Observations strongly suggest two types of transition discs: those with low mm flux, which have low accretion rates and hole sizes smaller than about 20~au, and those with high mm flux, which have higher accretion rates and hole sizes larger than 20~au (Owen \& Clarke 2012, Owen 2016 and references therein). It is believed that low mm flux transition discs are in the process of dispersing, whereas high mm flux transition discs are not. The most commonly accepted interpretation for low mm flux transition discs is X--ray photoevaporation. In this model, a gap opens up after about 3~Myr (75\% of the disc lifetime) at $\sim 2$~au, where the accretion rate in the disc matches the photoevaporation rate (Owen, Ercolano \& Clarke 2011). The inner parts of the disc then become decoupled from the outer parts and cannot be resupplied in gas and dust. They subsequently accrete onto the central star whereas the outer edge of the gap recedes due to photoevaporation. By contrast, it is believed that, in high mm flux transition discs, a massive planet (with a mass of a few Jupiter masses) is responsible for creating a gap. The above discussion suggests that, in low mm flux transition discs, 0.1~$\me$ and 1~$\me$ cores forming between 1 and 5~au and beyond 5~au, respectively, on a timescale of 1~Myr, would start to migrate a million years or so before a gap opens up at around 2~au. As their migration timescale itself is on the order of a million years, it can be expected that the dispersion of the disc will prevent migration of cores to very small radii and also that the formation of the gap will prevent massive cores to be delivered to the region of terrestrial planets. This is the model we explore in this paper. The plan of the paper is as follows. In section~\ref{sec:formation}, we describe the disc evolution model and justify the parameters that are used in the simulations. We also give expressions for the planet migration and eccentricity damping timescales and briefly discuss core formation timescales. In section~\ref{sec:sim}, we present $N$--body simulations of cores migrating in photoevaporating discs. We first describe the numerical scheme and the initial set up. We then present the results of the simulations. We show that, for reasonable parameters that are consistent with the observations, a population of $0.1 \; \me$ cores originating from between 2 and 4~au migrate down to 0.5--1~au. The final masses of planets are between a fraction of an Earth mass and $\sim 1.5 \; \me$. As for the population of 1~$\me$ cores originating from $\sim 5$~au, it forms a few cores of a few Earth masses which may migrate down below the inner edge of the gap only in the most massive discs considered here. In less massive discs, cores with a mass comparable to that of Jupiter may be left at a few au from the star. Finally, in section~\ref{sec:discussion}, we summarize and discuss our results. | \label{sec:discussion} In this paper, we have studied the outcome of core migration in transition discs produced by photoevaporation. The initial surface mass density in discs at around 1~au is at most on the order $10^3 \; \gcm$, and it decreases with time. Therefore, planetary cores that form at this distance from the star migrate on a timescale smaller than the disc lifetime only if their mass is larger than about 0.1~$\me$. As it takes at least about 1~Myr to form such cores at around 1~au, migration of cores forming below the snow line starts to be relevant only when the disc is at least 1~Myr old. Similarly, if the surface density varies as $r^{-1}$, cores at $\sim 5$~au have a migration timescale shorter than the disc lifetime only when their mass reaches about 1~$\me$. Assuming these cores form on a timescale of about 1~Myr, they will also start to migrate significantly when the disc is about 1~Myr old. In low~mm flux transition discs, a gap is believed to open up at around 2--3~au due to X--ray photoevaporation when the disc is a few Myr. The disc's inner parts are subsequently accreted onto the central star as a result of viscous evolution, while the outer edge of the gap recedes under the effect of photoevaporation. In this context, we find that $\sim 0.1 \; \me$ cores that form within $\sim 1$~Myr between 1--4~au end up forming a system of a few planets with masses between a fraction of an Earth mass and 1.5~$\me$ at most. In general, these cores do not migrate down further than $\sim 0.5 $~au, and may even evolve {\em in situ} in low mass discs. They are not in mean motion resonances. Such resonances, which appear during the migration phase, are usually destroyed when the gas dissipates (as previously found by Cossou et al. 2014 and Coleman \& Nelson 2016). It is difficult to form planets at smaller radii, unless we start with heavier cores or start migrating the cores earlier (as, e.g., in Terquem \& Papaloizou~2007). In both cases, it would require a more efficient planet formation process than envisioned here. If $\sim 1$~$\me$ cores can form beyond the snow line also within $\sim 1$~Myr, they migrate on the same timescale as the smaller cores and, depending on the surface density in the disc, they may or may not reach the inner parts of the disc. After the gap has opened up, the inner parts of the disc become isolated from the outer parts, and no more material is delivered there. We have found that, in a disc with an initial surface mass density of about $10^3 \; \gcm$ at 1~au, it was possible to obtain a system of cores similar to that of the Solar system, with a few planets with masses of a few tenths of an earth mass to an earth mass in the terrestrial zone and a more massive core a little bit further away (fig.~[\ref{fig2}] and lower panel of fig.~[\ref{fig4}]). However, as the photoevaporation model predicts that the outer edge of the gap is receding rather fast, it is difficult to envision within this model how the massive core could accrete a gaseous envelope to become a giant planet. Such a system of low mass planets is similar to those obtained by Cossou et al. (2014) and Coleman \& Nelson (2016) in their models with inefficient migration. The exact parameters that should be used in such simulations are of course not known, and different outcomes could be produced by changing them. However, the point of this paper is to show that, by adopting reasonable parameters that fit the observations and the current understanding of planet formation and migration, systems of cores/planets can be obtained where all the planets do not end up on resonant chains near the disc's inner edge and delivery of material to the inner parts from the outer parts of the disc can be avoided. To that extent, the type of systems that we obtain resembles more our Solar system than the systems observed by {\em Kepler}, which is biased towards detecting planets on very tight orbits. Our results predict that low mm flux transition disc may harbour terrestrial planets in the habitable zone. Note that, as collisions between cores occur until rather late in the disc's evolution, dust may be produced at distances below 1 or 2~au from the central star even after the disc's inners parts have been accreted. In that case, the disc would not appear as transition disc. This study suggests that low mm flux transition discs may not be able to form giant planets before X--ray photoevaporation opens up a gap. These discs would then form planetary systems of the type obtained in this study: predominantly low mass planets at around 1~au with possibly more massive cores further away. High mm flux transition discs, by contrast, are associated with more massive stars and may form giant planets of a few Jupiter masses, massive enough to open up a gap, before photoevaporation could proceed. | 16 | 9 | 1609.06056 |
1609 | 1609.04786_arXiv.txt | Terrestrial exoplanets in the canonical habitable zone may have a variety of initial water fractions due to random volatile delivery by planetesimals. If the total planetary water complement is high, the entire surface may be covered in water, forming a ``waterworld.'' On a planet with active tectonics, competing mechanisms act to regulate the abundance of water on the surface by determining the partitioning of water between interior and surface. Here we explore how the incorporation of different mechanisms for the degassing and regassing of water changes the volatile evolution of a planet. For all of the models considered, volatile cycling reaches an approximate steady-state after $\sim 2 \ \mathrm{Gyr}$. Using these steady-states, we find that if volatile cycling is either solely dependent on temperature or seafloor pressure, exoplanets require a high abundance ($\gtrsim 0.3\%$ of total mass) of water to have fully inundated surfaces. However, if degassing is more dependent on seafloor pressure and regassing mainly dependent on mantle temperature, the degassing rate is relatively large at late times and a steady-state between degassing and regassing is reached with a substantial surface water fraction. If this hybrid model is physical, super-Earths with a total water fraction similar to that of the Earth can become waterworlds. As a result, further understanding of the processes that drive volatile cycling on terrestrial planets is needed to determine the water fraction at which they are likely to become waterworlds. | \subsection{Surface water abundance and habitability} \indent To date, the suite of observed exoplanets from \textit{Kepler} has proven that Earth-sized planets are common in the universe ($\approx 0.16$ per star, \citealp{Fressin2013,Morton2014}). Though we do not yet have a detailed understanding of the atmospheric composition of an extrasolar terrestrial planet, spectra of many extrasolar gas giants \citep{Kreidberg2015a,Sing2015} and a smaller Neptune-sized planet \citep{Fraine2014} have shown that water is likely abundant in other Solar Systems. Calculations of volatile delivery rates to terrestrial planets via planetesimals (e.g. \citealp{Raymond:2004,Ciesla2015}) have shown that planets can have a wide range of initial water fractions, with some planets being $1\%$ water by mass or more. Both observations and simulations hence point towards the likelihood that terrestrial planets are also born with abundant water. However, the intertwined effects of climate \citep{Kasting1993} and mantle-surface volatile interchange \citep{Hirschmann2006,Cowan2015} determine whether there is abundant liquid water on the present-day surfaces of terrestrial exoplanets. Additionally, atmospheric escape (especially early in the atmospheric evolution) can cause loss of copious amounts of water \citep{Ramirez2014,Luger2015,Tian2015,Schaefer2016}, with $\gtrsim 10$ Earth oceans possibly lost from planets in the habitable zone of M-dwarfs. \\ \indent The extent of the traditional habitable zone is determined by the continental silicate weathering thermostat \citep{Kasting1988}, in which silicate minerals react with $\mathrm{C}\mathrm{O}_2$ and rainwater to produce carbonates \citep{Walker1981}. Silicate weathering is extremely efficient at stabilizing the climate because the process runs faster with increasing temperature. This is due to faster reaction rates and increased rain in warmer climates. However, the silicate weathering thermostat itself depends on the surface water abundance. \\ \indent If there is no surface water, the silicate weathering thermostat cannot operate due to the lack of reactants, and if the planet surface is completely water-covered the negative feedback does not operate unless seafloor weathering is also temperature dependent \citep{Abbot2012}. Note that even if seafloor weathering is temperature-dependent, it might be insufficient to stabilize the climate \citep{Foley2015}. A waterworld state is likely stable \citep{Wordsworth2013}, as water loss rates would be low because the atmosphere would be $\mathrm{C}\mathrm{O}_2$-rich due to the lack of a silicate-weathering feedback. However, if water loss rates remain high due to a large incident stellar flux, it is possible that brief exposures of land can allow for a ``waterworld self-arrest'' process in which the planet adjusts out of the moist greenhouse state \citep{Abbot2012}. This can occur if the timescale for $\mathrm{C}\mathrm{O}_2$ drawdown by the silicate-weathering feedback is shorter than the timescale for water loss to space, which is probable for Earth parameters. \\ \indent From the above discussion, we conclude that although waterworlds are by definition in the habitable zone (having liquid water on the surface), they may not actually be temperate and conducive to life. It is instead likely that waterworlds are less habitable than worlds with continents, and so determining whether or not waterworlds are common is important. To determine whether or not waterworlds should be common, we must look to the deep-water cycle, that is, the mantle-surface interchange of water over geologic time. \subsection{Earth's deep-water cycle} \label{sec:deepintro} \indent To understand the deep-water cycle on exoplanets, we look to Earth as an analogue, as it is the only planet known with continuous (not episodic) mantle-surface water interchange due to plate tectonics. On present-day Earth, water is largely expelled from the mantle to the surface (degassed) through volcanism at mid-ocean ridges and volcanic arcs \citep{Hirschmann2006}. Water is lost from the surface to the mantle (regassed) through subduction of hydrated basalt. The relative strength of regassing and degassing determines whether the surface water abundance increases or decreases with time. \\ \indent It has long been suggested that Earth's surface water fraction is in effective steady-state \citep{McGovern1989,Kasting1992}, due to the constancy of continental freeboard since the Archean ($\sim 2.5$ Gya). However, this may simply be due to isostasy, that is, the adjustment of the continental freeboard under varying surface loads \citep{Rowley2013,Cowan2014a}. A more convincing argument is that the degassing and regassing rates on Earth are high enough that if they did not nearly balance each other the surface would have long ago become either completely dry or water-covered \citep{Cowan2014a}. However, some studies of volatile cycling on Earth that utilized parameterized convection to determine the upper mantle temperature and hence the degassing and regassing rates have not found such a steady state \citep{McGovern1989,Crowley2011,Sandu2011}. If the Earth is indeed near steady-state, this mismatch could be because there are many secondary processes, e.g. loss of water into the transition zone \citep{Pearson2014} and early mantle degassing \citep{Elkins-Tanton2011}, that are difficult to incorporate into a simplified volatile cycling model. Also, it is possible that our understanding of what processes control the release of water from the mantle and return of water to it via subduction is incomplete. \\ \indent Using the maximum allowed fraction of water in mantle minerals \citep{Hauri2006,Inoue2010}, \cite{Cowan2014a} estimate that Earth's mantle water capacity is $\approx 12$ times the current surface water mass. However, measurements of the electrical conductivity of Earth's mantle \citep{Dai2009} have found only $\sim 1-2$ ocean masses of water in the mantle, which is much less than the maximally allowed value. This measurement may vary spatially \citep{Huang2005} and by method \citep{Khan2012}, but it is likely constrained to within a factor of a few. This implies that dynamic effects lead to a first-order balance between degassing and regassing on Earth, rather than the surface water complement being in steady-state simply because the mantle is saturated. \subsection{Previous work: the deep-water cycle on super-Earths} \indent Using a steady-state model wherein the degassing and regassing of water is regulated by seafloor pressure, \cite{Cowan2014a} applied our knowledge of Earth's deep-water cycle to terrestrial exoplanets. They showed that terrestrial exoplanets require large amounts ($\sim 1\%$ by mass) of delivered water to become waterworlds. Applying a time-dependent model and including the effects of mantle convection, \cite{Schaefer:2015} found that the amount of surface water is strongly dependent on the details of the convection parameterization. These works rely on other planets being in a plate-tectonic regime similar to Earth. However, it is important to note that there is debate about whether or not plate tectonics is a typical outcome of planetary evolution (e.g. \citealp{ONeill2007,Valencia2007,Valencia2009,Korenaga2010}), potentially because plate tectonics is a history-dependent phenomenon \citep{Lenardic2012}. In this work, we also assume plate tectonics. We do so because our understanding of habitability is most informed by Earth and it enables us to examine how processes that are known to occur on Earth affect water cycling on exoplanets. As a result, we assume that continents are present, and that isostasy determines the depths of ocean basins. In the future, exploring other tectonic regimes (e.g. stagnant lid) may be of interest to exoplanet studies and potential investigations of Earth's future evolution \citep{Sleep2015}. \\ \indent The studies of volatile cycling on super-Earths discussed above used drastically different approaches, with \cite{Cowan2014a} applying a two-box steady-state model of volatile cycling, and \cite{Schaefer:2015} extending the time-dependent coupled volatile cycling-mantle convection model of \cite{Sandu2011} to exoplanets. As a result, these works made different assumptions about which processes control water partitioning between ocean and mantle. The degassing parameterization of \cite{Cowan2014a}, based on the model of \cite{Kite2009a}, utilized the negative feedback between surface water inventory and volatile degassing rate that results from pressure reducing degassing. Their regassing rate was also related to the surface water inventory, using the prediction of \cite{Kasting1992} that the hydration depth increases with increasing surface water abundance up to the limit where the hydration depth is equal to the crustal thickness. Meanwhile, the degassing and regassing parameterizations of \cite{Schaefer:2015} were both related directly to the mantle temperature, with the degassing rate determined by the abundance of water in melt and the regassing rate set by the depth of the hydrated basalt (serpentinized) layer, which is determined by the depth at which the temperature reaches the serpentinization temperature. \\ \indent In this work, we seek to identify how different assumptions about regassing and degassing determine the surface water mass fraction. To do so, we utilize simplified models of convection and volatile cycling that separately incorporate the key features of both the \cite{Cowan2014a} and \cite{Schaefer:2015} volatile cycling parameterizations. The latter model builds upon the analytic work of \cite{Crowley2011}, who developed an analytic model that captures the key processes in the numerical models of \cite{Sandu2011} and \cite{Schaefer:2015}. However, here we further simplify and also non-dimensionalize the \cite{Crowley2011} model, enabling us to elucidate the dependencies of water abundance on mantle temperature and planetary parameters. We then combine the models of \cite{Cowan2014a} and \cite{Schaefer:2015}, utilizing surface water budget-dependent degassing and temperature-dependent regassing. We do so because it is likely the most physically relevant choice, as temperature affects serpentinization depths (and resulting regassing rates) more directly than seafloor pressure. Additionally, temperature-dependent degassing would become small at late times while seafloor pressure-dependent degassing would not, and it has been shown by \cite{Kite2009a} that degassing should be pressure-dependent. This is more in line with the approximate steady-state water cycling on Earth is currently in, as if both regassing and degassing are temperature-dependent regassing will dominate at late times. We find that the choice of volatile cycling parameterization greatly impacts the end-state surface water mass reservoir. We also find that, regardless of volatile cycling parameterization, the water partitioning reaches a steady-state after a few billion years of evolution due to the cooling of the mantle to below the melting temperature, which causes the effective end of temperature-dependent degassing and regassing. \\ \indent This paper is organized as follows. In \Sec{sec:theory}, we describe our parameterized convection model and the various volatile cycling parameterizations we explore, along with the consequences these have for the temporal evolution of mantle temperature and water mass fraction. Detailed derivations of the volatile cycling models can be found in \App{app:deriv}. In \Sec{sec:waterworld} we explore where in water mass fraction-planet mass parameter space each volatile cycling model predicts the waterworld boundary to lie. We discuss our results in \Sec{sec:discussion}, performing a sensitivity analysis of the waterworld boundary on key controlling parameters, comparing this work to previous works, and discussing our limitations and potential avenues for future work. Importantly, we also show how our model with pressure-dependent degassing and temperature-dependent regassing could in principle be observationally distinguished from the models of \cite{Cowan2014a} and \cite{Schaefer:2015}. Lastly, we express conclusions in \Sec{sec:conc}. | \label{sec:conc} \begin{enumerate} \item Volatile cycling on terrestrial exoplanets with plate tectonics should reach an approximate steady-state on the timescale of a few billion years, independent of the volatile cycling parameterization used. Given that Earth is likely near a steady-state in surface water mass fraction, this gives us confidence that many terrestrial exoplanets around main-sequence stars are also at or near steady-state. The steady states in the temperature-dependent and hybrid models may be substantially different from present-day Earth, as both these models store approximately an order of magnitude more water in the hydrated crust than Earth itself. \item Models considering either temperature-dependent degassing and regassing or pressure-dependent degassing and regassing predict that copious amounts of water ($\sim 0.3-1\%$ of total planetary mass) must be present to form a waterworld. These models have their mantles saturated with water, and if the total water mass fraction is high they are at or near the petrological limit for how much water the mantle can hold. The waterworld boundary for the solely temperature-dependent volatile cycling model is determined by this limit. As a result, if a super-Earth mantle can hold more water, the waterworld boundary will move upward by a similar factor. This would make it even less likely for super-Earths to be waterworlds. \item If seafloor pressure is important for the degassing rate of water but not for regassing, it is more likely that super-Earths will be waterworlds. In this case, a super-Earth with the same total water mass fraction as Earth could become a waterworld. These planets would be less likely to be habitable, as unlucky planets with a large amount of initial water delivery may lack a silicate weathering feedback to stabilize their climates. Understanding further which processes determine volatile cycling on Earth will help us understand what processes control mid-ocean ridge degassing and subduction rates of water on exoplanets with surface oceans. \end{enumerate} | 16 | 9 | 1609.04786 |
1609 | 1609.00592_arXiv.txt | It has long been known that rotation plays a central role in the dynamical and chemical evolution of stars (\cite{meynetmaeder00}, \cite{maedermeynet09}). Through the rotationnal mixing driven by the differential rotation, the meridional circulation and the turbulence it sustains, chemical elements and angular momentum are transported through the radiation zones of stars (e.g. \cite{zahn92}, \cite{maederzahn98}, \cite{mathiszahn04}, \cite{MR06}, \cite{ELR07}) in a way that needs to be precisely characterized and modeled. \begin{figure} \centering \includegraphics[trim = 2cm 3.5cm 2cm 3.5cm,width=0.8\linewidth,angle=-90]{image2.ps} \caption{Sketch of a radiative core (yellow region) upon which lies a differentially rotating convective envelope (orange region); $R_c$ is the radius of the radiative core. At the interface $r=R_c$, the rotation rate is imposed: $\Omega_{cz}(r=R_c,\theta)= \Omega_0 + \Delta\Omega\sin^2\theta$ where $\Omega_0$ is the rotation rate of the pole of the radiative core, i.e. $\Omega_0=\Omega(r=R_c,\theta=0)$. The equator rotates faster than the pole when $\Delta\Omega>0$ and slower than the pole when $\Delta\Omega<0$. \label{fig1} } \end{figure} \begin{figure} \centering \includegraphics[width=1\linewidth]{profilesBV1Msun.eps} \caption{Squared Brunt-V\"ais\"al\"a frequency from a 1D $1 M_\odot$ MESA model near the ZAMS (with $Z=0.02$, $\alpha_{\rm MLT}=2$).} \label{fig2} \end{figure} Indeed, helioseismology provides the internal rotation profile deep within the Sun (until $0.2R_\odot$) where the radiation zone has a quasi solid rotation under the tachocline (\cite{couvidat03}, \cite{garcia07}) which 1D rotating stellar models fail to reproduce (\cite{TC05}, \cite{TC10}). One reason probably lies in the need for other physical mechanisms such as internal gravity waves (\cite{zahn97}, \cite{TC05}) or an internal magnetic field (\cite{GM98}, \cite{spruit99}) or/and in the multidimensional nature of the differential rotation and of the meridional circulation \cite[][]{MR06}. Moreover, one wants to go beyond the slow rotation hypothesis underlying the 1D modeling of rotating stellar radiation zones to properly describe the dynamics occuring during the early stages of the evolution of stars during which they are rapidly rotating (\cite{GB13}, \cite{GB15}). To tackle this issue, we build a 2D numerical model of a fast rotating sphere representing such a radiative core at the top of which we impose a latitudinal shear so as to reproduce the conical differential rotation applied by the convective zone. Numerical simulations (\cite{matt11}, \cite{kapyla14}, \cite{gastine14}, \cite{varela16}) reveal that the differential rotation of the convective envelope of low mass stars may be solar (with slow pole and fast equatorial regions) or anti-solar (with fast pole and slow equator) depending on the value of the convective Rossby number, which quantifies the ratio of the rotation period and of the convective turnover time (a small Rossby number corresponds to the rapidly rotating regime; see e.g. \cite{brun15}). We propose to perform a systematic parameter study over the boundary condition describing the convective differential rotation applied at the top of the radiation core between the pole and the equator. We provide a full 2D description of the differential rotation and the associated meridional circulation computed self-consistently using the Boussinesq approximation. Since previous 2D modelings (\cite{friedlander76}, \cite{garaud02}) are limited to the solar case, this setting lays a first important step for the next generation of rotating models of low mass stars in 2D. | In this work, we propose a first 2D description of the dynamics of fast rotating radiative cores of low mass stars undergoing the shear induced by the differential rotation of the convective envelope at its upper boundary. This simple Boussinesq model shows the necessity to resort to a 2D approach. Indeed, the arising flow from the shear is cylindrical, which must be described using a large number of spherical harmonics. Such a setting also permits to take into account the Brunt-V\"ais\"al\"a frequency from selected evolutionary stages and stellar masses allowing the exploration of the HR diagram in 2D. In the solar case, the shear is evaluated close to $b_\odot\simeq10$. Therefore, the third plot of Fig. \ref{fig3} exhibits the best fast rotating solar model we may provide within the current setting. We compute a core to surface rotation rate ratio lower than $1$ which is in agreement with what \cite{benomar15} deduced from observations. But it also tells us to expect the differential rotation to be cylindrical until half of the radiative core depth with a fast equatorial region. If we compare to the actual solar rotation profile, this calls for other physical processes responsible for transport of angular momentum deep within the internal regions, which is not taken into account in the present setting. Internal gravity waves \cite[][]{zahn97}, anisotropic turbulence (\cite{zahn92}, \cite{maeder03}, \cite{mathiszahn04}) and magnetic fields (\cite{GM98}, \cite{spruit99}, \cite{strugarek11}, \cite{AGW13}) are the best candidates and will be introduced in our 2D model in future works. | 16 | 9 | 1609.00592 |
|
1609 | 1609.07698_arXiv.txt | Transition-region explosive events (EEs) are characterized by non-Gaussian line profiles with enhanced wings at Doppler velocities of 50--150\,\kms. They are believed to be the signature of solar phenomena that are one of the main contributors to coronal heating. The aim of this study is to investigate the link of EEs to dynamic phenomena in the transition region and chromosphere in an active region. We analyze observations simultaneously taken by the Interface Region Imaging Spectrograph (IRIS) in the \siiv\ 1394\,\AA\ line and the slit-jaw (SJ) 1400\,\AA\ images, and the Swedish 1-m Solar Telescope (SST) in the \halpha\ line. In total 24 events were found. They are associated with small-scale loop brightenings in SJ~1400\,\AA\ images. Only four events show a counterpart in the \halpha$-35$\,\kms\ and \halpha$+35$\,\kms\ images. Two of them represent brightenings in the conjunction region of several loops that are also related to a bright region (granular lane) in the \halpha$-35$\,\kms\ and \halpha$+35$\,\kms\ images. Sixteen are general loop brightenings that do not show any discernible response in the \halpha\ images. Six EEs appear as propagating loop brightenings, from which two are associated with dark jet-like features clearly seen in the \halpha\,$-35$\,\kms\ images. We found that chromospheric events with jet-like appearance seen in the wings of the \halpha\ line can trigger EEs in the transition region and in this case the IRIS \siiv\ 1394\,\AA\ line profiles are seeded with absorption components resulting from Fe\,{\sc ii} and Ni\,{\sc ii}. Our study indicates that EEs occurring in active regions have mostly upper-chromosphere/transition-region origin. We suggest that magnetic reconnection resulting from the braidings of small-scale transition region loops is one of the possible mechanisms of energy release that are responsible for the EEs reported in this paper. | \label{sect_intro} Explosive events (EEs) are small-scale transients (duration~$<$600\,s, size~$<$5\arcsec) observed in the solar transition region, and describe non-Gaussian line profiles with enhanced wings at Doppler velocities of 50--150~\kms\,\citep{1983ApJ...272..329B,1989SoPh..123...41D}. The physical nature of EEs is under investigation for more than two decades. EEs were firstly suggested to be the spectral footprint of bi-directional jets caused by magnetic reconnection\,\citep{1991JGR....96.9399D,1997Natur.386..811I}. Later, chromospheric upflow events \citep{1998ApJ...504L.123C}, siphon flows in small-scale loops \citep{2004A&A...427.1065T}, surges\,\citep{2009ApJ...701..253M} and transient brightenings and X-ray jets \citep{2012A&A...545A..67M} have been associated with EEs. \par \citet{1991ApJ...370..775P} reported that explosive events occur in the solar magnetic network lanes. This has been later confirmed by many further studies \citep[e.g.][etc.]{2003A&A...403..731M,2004A&A...419.1141N,2004A&A...427.1065T,2008ApJ...687.1398M}. \citet{1998ApJ...497L.109C} established that the majority of explosive events are associated with the cancellation of photospheric magnetic flux, which was recently confirmed by \citet{2014ApJ...797...88H} and \citet{2015ApJ...809...82G}. EEs were modelled by \citet{1999SoPh..185..127I}, \citet{2002A&A...383..697R}, \citet{2001A&A...380..719R}, \citet{2001A&A...375..228R}, \citet{2001A&A...370..298R}, and \citet{2015arXiv150908837I} in two-dimensional numerical simulations as the product of magnetic reconnection. \par Non-Gaussian line profiles in the solar transition region are intensively investigated since the first flight of the Naval Research Laboratory (NRL) High Resolution Telescope and Spectrograph (HRTS) in 1975. \citet{1989SoPh..123...41D} reported on the various shapes of C~{\sc iv} line profiles (see their Figures~2, 3, 6, 8, 9, 11, 13, and 15) that were named as ``explosive events''. Later \citet{1991JGR....96.9399D} suggested that EE line profiles are the spectroscopic signature of magnetic reconnection. Recently, \citet{2014Sci...346C.315P} reported similar Si~{\sc iv} line profiles observed by \textit{IRIS} with superimposed absorption lines from singly ionised ion, neutral atom and/or molecular lines \citep{2014A&A...569L...7S} in active region dot-like events that the authors called ``hot explosions''. The electron density of the hot explosion was estimated to exceed $10^{13}$\,cm$^{-3}$. \citet{2014Sci...346C.315P} suggested that these line profiles are the products of small-scale magnetic reconnection occurring in the photosphere while \citet{2015ApJ...808..116J} put forward the idea that they are rather the product of Alfv\'enic turbulence originating in the chromosphere or above. \par The Interface Region Imaging Spectrograph\,\citep[IRIS,][]{2014SoPh..tmp...25D} launched in 2013 obtains observations of the solar transition region at unprecedented spatial and spectral resolution providing great opportunities to investigate the physical processes that generate EEs. \citet{2014ApJ...797...88H} studied an EE that occurred at the boundary of the quiet Sun and a coronal hole observed with {\it IRIS}, the Atmospheric Imaging Assembly {\it (AIA)} and the Helioseismic and Magnetic Imager {\it (HMI)}. The authors found that the EE is associated with a complex loop system seen in the AIA 171\,\AA\ passband. The magnetic cancellation rate during the event was of $5\times10^{14}$\,Mx\,s$^{-1}$. The EE reached a temperature of at least $2.3~\times~10^5$\,K. \citet{2014ApJ...797...88H} suggested that this EE is caused by magnetic reconnection within the complex loop system. A recent study by \citet{2015ApJ...810...46H} found EEs in a footpoint of a cool transition region loop system and also at the footpoint junction of two loop systems where cancelling opposite magnetic polarities were present. \citet{2015arXiv150908837I} reported IRIS observations of 15 explosive events occurred in active region. They analyzed in detail the core and wing emission of the explosive event spectra, and found that core and wing emission are spatially coincident, and do not move significantly during the typical event duration. The numerical experiment shows that multiple magnetic islands and acceleration sites characterising the plasmoid instability in fast magnetic reconnection can reproduce the observed profiles. \citet{2015ApJ...809...82G} found short-period variability (30 s and 60--90 s) within EE bursts in IRIS observations. \par In the present work we aim to investigate the link of EEs observed in the transition region by IRIS in a Si~{\sc iv} line to chromospheric phenomena registered in the wings of H$\alpha$ spectral imaging data. After identifying the EEs in the \siiv\ 1394~\AA\ line, the IRIS slit-jaw images at 1400\,\AA\ passband and \halpha\ images taken by the CRisp Imaging SpectroPolarimeter \,\citep[CRISP,][]{2008ApJ...689L..69S} installed at the Swedish 1-m Solar Telescope\,\citep[SST,][]{2003SPIE.4853..341S} in La Palma are used to identify the \halpha\ wing counterparts of the EEs. \par In the following, we describe the observations and the data analysis in Section\,\ref{sect_obs}. The results and discussion are presented in Section\,\ref{sect_res}. We give the conclusions in Section\,\ref{sect_concl}. \begin{figure} \includegraphics[width=0.5\textwidth,clip,trim=0.5cm 1cm .5cm 2cm]{iris_fov_on_aia.eps} \caption{Partial view of the solar disk observed in the AIA 171\,\AA\ passband. The analyzed IRIS and CRISP field-of-views are denoted by the black box. The region between the dashed lines (in blue) is the place where the IRIS spectra were obtained.} \label{fig_fov} \end{figure} \begin{figure} \includegraphics[width=0.5\textwidth,clip,trim=0.1cm 1.8cm 0cm 0.5cm]{ee_ids_on_map2.eps} \caption{\siiv\ 1394\,\AA\ radiance image of the region observed by the IRIS in the quasi ``sit-and-stare'' mode. The plus symbols are the locations of EE line profiles detected by the automatic procedure. The red symbols present those detected by step 3 and the green ones are from step 5. The selected events (24 in total) are denoted by boxes with numbers.} \label{fig_ee_id} \end{figure} \begin{figure*} \includegraphics[width=17cm,clip,trim=0cm 0.1cm 0cm 0.1cm]{ee_on_map_smp_new.eps} \caption{ Locations of the EE line profiles (red plus signs: identified by step 3 of the automatic method, green plus signs: identified by step 5 of the automatic method) superimposed on the IRIS SJ~1400\,\AA\ (the first panel), the \halpha\ $-35$\,\kms\ (the second panel) and \halpha\ $+35$\,\kms\ (the third panel) images. White solid lines denote the slit of the IRIS spectrograph. The slit image is shown in the fourth panel, where the spectral range is from $-100$\,\kms\ to 100\,\kms, in order to clearly present the wing enhancement. (An animation of this figure is available online).} \label{fig_ee_on_map} \end{figure*} \begin{figure*} \includegraphics[width=17cm,clip,trim=1.8cm 2.2cm 6cm 0cm]{ee_no4_imgs.eps} \caption{\siiv\ 1394\,\AA\ EE spectra found in a conjunction region of the footpoints of a few loop systems (EVENT 4): The top row shows the conjunction region viewed in the IRIS SJ~1400\,\AA\ (a), CRISP \halpha$-35$\,\kms\ (b) and \halpha$+35$\,\kms\ (c). The dotted lines (in black) in panel (a) outline some of the loops connecting in the conjunction region, and the dark vertical bar is the location where the light is blocked by the IRIS spectrometer slit, which is also over-plotted as dotted line in panels (b) and (c). The box (in black) denotes the region from which the lightcurves shown in panels (f) are generated. The plus sign (in green) denotes the location of the EE \siiv\ spectrum shown in panel (d). Bottom row: (d) \siiv\ 1394\,\AA\ line profile (in black) taken from the location marked by a plus sign in the top row together with an average spectrum (in red) taken from the full field of view shown in Fig.\,\ref{fig_ee_id}; (e) \halpha\ line profile from the same location where the EE \siiv\ spectrum emitted. The observed time of each profile is labeled; (f) the lightcurves of the boxed region in the SJ 1400\,\AA\ (black), \halpha$-35$\,\kms\ (green), \halpha$+35$\,\kms\ (red).} \label{fig_no4} \end{figure*} \begin{figure*} \includegraphics[width=17cm,clip,trim=1.8cm 3.8cm 0.5cm 0cm]{ee_no20_imgs.eps} \caption{Si~{\sc iv} 1394\,\AA\ EE spectra found in a loop brightening (EVENT 20). From top to bottom, first three rows show the evolution of the loop region seen in the IRIS SJ~1400\,\AA, \halpha$-35$\,\kms\ (\halpha B) and \halpha$+35$\,\kms\ (\halpha R), respectively. The dotted lines on the \halpha\ images show the locations of the spectrometer slit. The plus signs (in red and green) are the locations of the identified EE spectra. The black box on the SJ image at 08:03:29\,UT denotes the region, from which the lightcurve (bottom row of this figure) is obtained. Bottom row: The left panel displays the \siiv\ radiance image (top) and RB asymmetry at 45--55~\kms\ (bottom) of the region, in which the event is outlined by a contour line (black solid line). The spectra shown in the middle panel are taken from the pixels marked with diamond symbols in the left panel. The dashed line denotes the time (08:03:29\,UT) when the \siiv\ spectra turns to single-Gaussian. The middle panel displays the variation of the \siiv\ spectra taken from the pixels shown in the left panel (time is denoted by colors). The right panel shows a SJ 1400\,\AA\ lightcurve taken from the box region marked in the SJ 1400\,\AA\ image, where the dashed line marks the time at 08:03:29~UT.} \label{fig_no20} \end{figure*} \begin{figure*} \includegraphics[width=17cm,clip,trim=0.5cm 4.5cm 0.5cm 0cm]{ee_no9_imgs.eps} \caption{EE spectra found in a propagating loop brightening. From top to bottom, first three rows show evolution of the region seen in the IRIS SJ~1400\,\AA, \halpha$-35$\,\kms\ (\halpha B) and \halpha$+35$\,\kms\ (\halpha R) respectively. The dotted lines on the \halpha\ images denote the location of the spectrometer slit. The plus signs (in red) are the locations of the identified EE spectra. A bright structure is found to move along a sigmoid path marked with a dashed line on the SJ~1400\,\AA\ image at 07:37:43\,UT. A time-slice image along this path is given in the bottom panel of this figure. Bottom row: The left panel displays the time-slice plot from the cut denoted on the SJ~1400\,\AA\ image at 07:37:43\,UT. The right panels show EE spectra at different times obtained from the locations marked by plus symbols on the SJ images. The arrows denote the absorption components in the \siiv\ spectra, and the reference profile is shown in red.} \label{fig_no9} \end{figure*} | \label{sect_concl} In the present study, we investigate transition region explosive events and their possible counterpart in \halpha\ wing images taken at $\pm$35\kms\ by analysing spectral imaging co-observations taken by IRIS and CRISP/SST. We identified 103 non-Gaussian spectra that were then grouped in 24 events. The evolution of these events was analyzed in the IRIS SJ 1400\,\AA\ and CRISP \halpha\ co-observations. \par We found that all EE spectra are related to loop brightenings seen in SJ 1400\,\AA\ images. From 24 events, 2 are associated with brightenings in the footpoint conjunction region of several loops, 16 are found in general loop brightenings and 6 are related to propagating loop brightenings. Although all the identified EE spectra are related to loop brightenings, not all loop brightenings seen in the SJ\,1400\,\AA\ images emit EE spectra. \par Only four events are found to have a counterpart in the \halpha\ wings. In the \halpha$-35$\,\kms\ and \halpha$+35$\,\kms\ images, two events are found in the loop conjunction region that appear as bright granular lane and two propagating loop brightenings in the SJ~1400\,\AA\ appear to correspond to dark jet-like features. The EE spectra found in general loop brightenings seen in SJ 1400\,\AA\ do not show any discernible response in the \halpha\ wing images. Absorption components from Fe\,{\sc ii} and Ni\,{\sc ii} in \siiv\ spectra are found in the events associated with propagating loop brightenings in the SJ~1400\,\AA\ images and dark jet-like features in \halpha\,$-35$\,\kms\ images. The absorption components seeded in the \siiv\ spectra might be the indication of the ejection of cold plasma observed as dark jet-like features in the \halpha\ blue wing images. \citet{1998ApJ...504L.123C} reported that upflow chromospheric events (observed with the \halpha$-$0.5~\AA\ in the BBSO spectrograph) in the quiet Sun are associated with Si~{\sc iv} 1402.8~\AA\ explosive events (observed with the Solar Ultraviolet Measurements of Emitted Radiation (SUMER) spectrograph). The chromospheric upflow events were suggested by the author to be the manifestation of cool plasma material flowing into magnetically diffusive regions, while explosive events were believed to be hot plasma material flowing out of the same regions. The second part of our investigation in the quiet-Sun region\,\citep{madj2016} will shed more light on this issue. \par Our study indicates that explosive events in active regions are associated with transition-region loop brightenings that mostly do not have \halpha\ wing counterparts. This indicates that the energy release witnessed by most of the EEs occurs in the upper chromosphere and/or transition region of the solar atmosphere and it could be related to small-scale loop braiding at this height. A forward modeling with output of plasma observables (e.g. spectral lines, images, etc.) based on 3D MHD simulations of magnetic reconnection in loop braidings is crucial to test this idea. Although rare, chromospheric events may eventually trigger explosive events in the transition region and in these cases they are characterized by absorption components (Fe\,{\sc ii} and Ni\,{\sc ii}) seeded in \siiv~1394\,\AA\ spectra. | 16 | 9 | 1609.07698 |
1609 | 1609.00137_arXiv.txt | Spectroscopic measurements on the night sky of Xinglong Observatory for a period of 12 years from 2004 to 2015 are presented. The spectra were obtained in moonless clear nights by using OMR spectrograph mounted on 2.16-m reflector with wavelength coverage of 4000-7000 {\AA}. The night sky spectrum shows the presence of emission lines from Hg~{\sc i} and Na~{\sc i} due to local artificial sources, along with the atmospheric emission lines, i.e. O~{\sc i} and OH molecules, indicates the existence of light pollution. We have monitored the night sky brightness during whole night and found little decrement in the sky brightness with time, but the change is not significant. Also, we monitored the light pollution level in different azimuthal direction and found that the influence of light pollution from the direction of Beijing city is stronger compared with that from the direction of Tangshan city and others. The analysis from night sky spectra for the entire data set suggested that the zenith sky brightness of Xinglong Observatory has brightened about 0.5 mag arcsec$^{-2}$ in V and B bands from 2004 to 2015. We recommend consecutive spectroscopic measurements of the night sky brightness at Xinglong Observatory in future, not only for monitoring but also for scientific reference. | Xinglong Observatory of National Astronomical Observatories, Chinese Academy of Sciences (NAOC) is one of the major optical observatories in China, located at a distance of 120~km towards North-East to Beijing, the capital city of China. The observatory hosts nine optical telescopes with apertures ranging from 0.5 to 4-m diameter. There are about 63$\%$ spectroscopic nights per year to perform observations in this site (See \citealt{2015PASP..127.1292Z}). Night sky brightness is one of the fundamental parameters of an optical observatory that restricts the limiting magnitude for any planned observations. The brightness of moonless night sky are generated from natural sources mainly contributed by airglow, zodiacal light, and integrated starlight \citep{1998A&AS..127....1L}, and from artificial sources due to the lighting system of neighboring towns \citep{1999A&AS..140..345D}. With the economic development and population growth of the surrounding cities, their attendant light pollution also grows. \cite{1999AcApS..19..220J} presented an identification of the night sky emission lines of Xinglong Observatory with spectral coverage from 5300 to 8200 {\AA} during 1996-1998, and found that Na~{\sc i} and Hg~{\sc i} lines from artificial sources are quite weak. Previous studies on sky brightness are mainly measured using broadband photometry. However, such measurements sometimes may be misleading, as it encompasses both natural airglow and artificial sources \citep{2000PASP..112..566M}. In order to better understand the contribution of atmospheric and artificial light sources, spectrophotometric measurements on sky brightness is suggested, which can distinguish the artificial sources from the natural sources clearly. \cite{2010PASP..122.1246N} presented a way to identify the contribution from specific elements that influence overall sky brightness. The spectrophotometric measurements has been widely used at various international optical observatories. \cite{1999A&AS..140..345D} presented a survey in Venezuela and Italy using a small spectrograph with spectral coverage from 4100 to 6400 {\AA}. Night sky spectra of the Kitt Peak, during 1998, were analyzed with wavelength coverage from 3800 to 6500 {\AA} by \cite{1990PASP..102.1046M}. Then \cite{2000PASP..112..566M} presented an absolute spectrophotometry of the night sky from $\sim$3700 to 6700 {\AA} over two astronomical sites in southern Arizona, Kitt Peak and Mount Hopkins, and measured for different azimuthal directions and different zenith distances, then converted to broadband magnitudes and gave the comparison with the night sky spectra in 1988. \cite{2010PASP..122.1246N} presented new absolute spectrophotometry of the Kitt Peak night sky during 2009-2010, and they strove to use the same observation and data reduction manner of \citet{2000PASP..112..566M}, and compared with published data. \cite{2004JKAS...37...87S} presented the spectrophotometry of night sky over Bohyunsan Optical Astronomy Observatory (BOAO), which is located on top of Mount Bohyun, with the nearly entire visible wavelength from 3600 to 8600 {\AA}, and the authors compared the night sky spectrum with Kitt peak. Site testing for observatories also used the night sky spectra to analyze local light pollution (e.g. \citealt{2007PASP..119.1186S}, \citealt{2010PASP..122..363M}). These night sky studies are mainly based on relative low-resolution spectra, there may be different spectral coverage and blended lines. In order to identify as many lines as possible from the contribution of light pollution, \cite{2003PASP..115..869S} presented the night sky spectrum of light pollution at the Lick Observatory from 3800 to 9200 {\AA} with a high spectral resolution(R$\sim$45000), and identified a large variety of lines from light pollution. In this work, we study the night sky brightness for a period of 12 years through the spectroscopic measurements in visible wavelength. This paper consists of four sections. Section~1 gives a brief introduction of Xinglong Observatory and the research basis of night sky spectra around the world. Section~2 describes the details of data acquisition and reduction. Analysis of the night sky spectrum at Xinglong Observatory and its results are presented in Section~3. We discuss our results and conclude in Section~4. | We made an attempt to study the night sky brightness at Xinglong Observatory based on spectroscopic measurements. Overnight monitoring of sky brightness suggests the zenith night sky brightness decrease with time in night, but not significant. We noticed the strong emission lines from Hg~{\sc i} and Na~{\sc i} in the spectra, apart from natural light emission lines, indicates the influence of light pollution from the usage of Mercury and Sodium lamps in surrounding cities of Xinglong Observatory. Influence of light pollution from the direction of Beijing city is stronger compared with that from the direction of Tangshan city and others. We compared the night sky spectra of twelve years from 2004 to 2015 and found increasing trend in the sky brightness during these years. The convolution of broadband magnitudes suggests that zenith sky has brightened about 0.5 mag arcsec$^{-2}$ in V and B bands. Consecutive spectroscopic measurements in the night sky brightness at Xinglong Observatory in the future is apparently essential, not only for monitoring, but also for scientific reference. We thank observers and night assistants of 2.16-m telescope for their kind support in obtaining data. This work is partly supported by National Natural Science Foundation of China under grant No.11373003 and National Key Basic Research Program of China (973 Program) No. 2015CB857002. Y.B.K. thanks the Chinese Academy of Sciences Visiting Fellowship for Researchers from Developing Countries for support through grant 2013FFJB0008. | 16 | 9 | 1609.00137 |
1609 | 1609.00906_arXiv.txt | {The mean-field dynamo model is employed to study the non-linear dynamo regimes in a fully convective star of mass 0.3$M_{\odot}$ rotating with period of 10 days. For the intermediate value of the parameter of the turbulent magnetic Prandl number, $Pm_{T}=3$ we found the oscillating dynamo regimes with period about 40Yr. The higher $Pm_{T}$ results to longer dynamo periods. If the large-scale flows is fixed we find that the dynamo transits from axisymmetric to non-axisymmetric regimes for the overcritical parameter of the $\alpha$effect. The change of dynamo regime occurs because of the non-axisymmetric non-linear $\alpha$-effect. The situation persists in the fully non-linear dynamo models with regards of the magnetic feedback on the angular momentum balance and the heat transport in the star. It is found that the large-scale magnetic field quenches the latitudinal shear in the bulk of the star. However, the strong radial shear operates in the subsurface layer of the star. In the nonlinear case the profile of the angular velocity inside the star become close to the spherical surfaces. This supports the equator-ward migration of the axisymmetric magnetic field dynamo waves. It was found that, the magnetic configuration of the star dominates by the regular non-axisymmetric mode m=1, forming Yin Yang magnetic polarity pattern with the strong (>500 G) poloidal magnetic field in polar regions.} | Stars with the extended convective envelopes demonstrate the high level of magnetic activity \citep{mdwa,2009ARAA_donat,L1-2015SSRv}. It is commonly believed that the magnetic activity of these stars origins from the hydromagnetic turbulent dynamo action \citep{brsu05,Brunrev14}. Extremely high magnetic activity was found on the fully-convective low-mass stars which belong to the M-dwarfs branch of the low main sequence of the Hertzsprung-Rawssel diagram. Observations of the M-dwarfs indicated the rather strong large-scale magnetic field with strength of several kG \citep{S1-1985ApJ,S2-1986ApJ,J1-1996ApJ,L1-2015SSRv}. The magnetic topology of the M-dwarfs is likely depends on the mass and the rotation period of a star \citep{2009ARAA_donat,2016MNRAS1129S}. Observations indicate that the early type M-dwarfs with the moderate period of rotation about 4-5 days demonstrate the strong non-axisymmetric magnetic field with the dominant toroidal component \citet{D2-2008MNRAS}. The extremely fast rotating early type M-dwarfs with period of rotation less than 1 day indicate the transition to the axisymmetric dynamo with the dominant poloidal component of the large-scale magnetic field. Situation become complicated on the mid and late-type M-stars which have the masses less than $0.2M_{\odot}$ as they could show either the strong axisymmetric dipole-kind large-scale magnetic field, or the low-strength non-axisymmetric magnetic field \citep{M1=2008MNRAS,M2-2010MNRAS}. Thus we can conclude about three basic states of the dynamo on the fast rotating M-dwarfs, they are: the strong multipole magnetic field (hereafter, SM), the strong dipole field (hereafter SD) and the weak multipole (hereafter WM) magnetic field. We follow notation suggested by \citet{2011MNRAS.418L.133M}. Interesting that simultaneously with multiply states of the dynamo regimes, the dynamo generated total magnetic flux do not show the rotation-activity connection which is known among the solar type stars \citep{2003ApJ583.451M}. Observed magnetic properties of the M-stars initiated the number of the theoretical studies employing the mean-field models (see, \citealt{2006AA446.1027C,elst07,KMS14,shul15}) and the direct numerical simulations (e.g., \citealt{2006ApJ...638..336D,2008ApJ676.1262B,2013IAUS294.163D,2014AA564A.78S}). Using the results of the numerical simulations, \citet{2011MNRAS.418L.133M} suggests that the bi-stability of the magnetic topology on the late-type M-stars could result from two types of the convection regimes occurred in the fast rotating convective bodies \citep{1988GApFD44.3R}. Also, the direct numerical simulations show the differential rotation is important part of the dynamo in the fully convective stars. Similar conclusions were suggested by \citet{shul15} after studying the linear dynamo regimes. Current interpretation of the dynamo bi-stability given by \citet{2011MNRAS.418L.133M} suggests that the strength of the large-scale magnetic field is compatible with the nonlinear balance between the Lorentz and Coriolis force in case of the SD-type magnetism and it is established by the Lorentz-inertia force balance in case of the WM magnetism. Later, \citet{2014AA564A.78S} found that in the anelastic simulations the separation between the SD and WM magnetism is less profound than they as well as others (e.g., \citealt{2009EL8519001S}) found with the Boussinescue approximation. The origin of the strong multipolar magnetic field on the moderate rotating early M-stars is barely studied. Results of the mean-field models and the numerical simulations suggest the dynamo on these stars could operate with help of the differential rotation. The linear analysis of \citet{KMS14} show that the axisymmetric magnetic field modes have the smaller critical threshold of the dynamo instability than the non-axisymmetric ones. Thus, the transition from axisymmetric to non-axisymmetric dynamo can occur only in the nonlinear regime. The paper we study the nonlinear dynamo models for a fully convective star. Here we restrict ourselves to the same case of the star discussed earlier by \citet{shul15}, i.e., the star of mass 0.3$M_{\odot}$ of 1Gyr age and rotating with period of 10 days. We will address the axisymmetric and non-axisymmetric dynamo regimes with regards for the non-linear back reaction of the large-scale magnetic field on the $\alpha$-effect and the large-scale flow. The solution of the dynamo problem is coupled with the solution of the mean angular momentum balance and the mean heat transport in the convective sphere. The main goal of the paper is to find the typical topology of the large-scale magnetic field in the nonlinear dynamo for the given rotation period and investigate the nonlinear effects on the dynamo. | The previous consideration of the mean-field models of the fully convective stars was restricted to analysis of the eigenvalue problems \citep{elst07,shul15} or the kinematic case with uniform density stratification and the algebraic non-linearity of the $\alpha$-effect \citep{2006AA446.1027C}. The main progress in theoretical understanding of the dynamo on the the fully convective stars were made with help of the direct numerical simulations (see, e.g., \citealt{2006ApJ...638..336D,2008ApJ676.1262B,2012AA546A19G,2015ApJ813L31Y}). The paper for the first time presents results of the non-linear mean-field dynamo models of the fully convective star rotating with period 10 days. The key reasons to study the mean-field models is to study behavior of the dynamo in varying the governing dynamo parameters. At the first step, let us discuss the kinematic dynamos with the nonlinear $\alpha$-effect. The angular velocity profile in this case is different to the cylinder-like pattern, which was discussed in the literature (see, e.g., \citealt{moss04,moss05,2006AA446.1027C}) and which appears in the direct numerical simulations. Our model include effect of the meridional circulation which is important in the subsurface layer for the case of $Pm_{T}>1$. For the case $Pm_{T}=3$, the model M1 shows the strong axisymmetric dipole-like magnetic field with magnitude of the polar field about 1kG. The dominance of the antisymmetric relative to equator magnetic field disappears in the nonlinear mixed parity solution if we neglect the meridional circulation. The dynamo waves show the solar-like time-latitude diagrams with toroidal field drifting to the equator and the radial field drifting to the pole. \begin{figure} \includegraphics[width=0.98\columnwidth]{alpha-hc} \caption{\label{fig:alph}Snapshots of the model M5, a) the nonlinear $\alpha$ effects (volume contours for $\pm$3cm/s); b) the small-scale magnetic helicity density (from the Eq(\ref{eq:helcon-1})), (volume contours for $\pm$1.05$\cdot10^{10}$G$^{2}$/M)} \end{figure} The eigenvalue analysis shows that generation of the non-axisymmetric magnetic field for the case of $Pm_{T}>1$ is less efficient than the axisymmetric dynamo because the critical parameter of the dynamo instability is smaller in the second case. This is general conclusion of the most studies of the mean field dynamo starting from the seminal paper by \citet{rad86AN}. The conclusion lead to ignorance of the non-axisymmetric dynamos even for the super-critical regimes of the axisymmetric dynamo (cf, \citealt{2016JFM799R6R}). However the model M2 show that in case of the dynamo instability of the non-axisymmetric field, the non-axisymmetric regime can beat the axisymmetric one. The interaction between axisymmetric and non-axisymmetric magnetic field goes via the nonlinear effects. Those are the conservation of the magnetic helicity and the magnetic buoyancy. Contributions of the magnetic helicity on the $\alpha$-effect can not be ignored in the mean-field solar dynamos \citep{2007NJPh....9..305B}. They are important in the non-axisymmetric dynamo, as well. The change of the dynamo regime for the overcritical $C_{\alpha}$ is because of the non-axisymmetric $\alpha$-effect, which is produced by the magnetic helicity conservation in the non-axisymmetric large-scale dynamo. Figures\ref{fig:alph}(a,b) show snapshots of the $\alpha_{\phi\phi}$ (see, Eq.(\ref{alp2d}) and the mean magnetic helicity density of the small-scale field, which is generated because of the magnetic helicity conservation in the model M5. Models M2 and M3 show similar distributions. The models produce the non-axisymmetric non-linear $\alpha$ effect and this supports dominance the non-axisymmetric magnetic field in the dynamo. In the solar dynamo models the non-axisymmetric $\alpha$-effect was employed for explanation of the so-called active longitudes of the sunspot formations \citep{bigruz,berd06}. In our models this effect stems naturally from magnetic helicity conservation. The magnetic feedback on the differential rotation reduces efficiency of the axisymmetric dynamo. The strength of the large-scale magnetic field in the model M4 is less than in the model M1. The cyclic effect of the large and small-scale Lorentz force on the angular momentum fluxes produces phenomena known in the solar magnetic acitvity like the zonal variations of the angular velocity and variations of the meridional flow. Both of them predicted to have much smaller amplitude than for the Sun. The rotational velocity at the equator is $1.44$ km/s, then the predicted magnitude of the latitudinal shear between equator and pole is only about 14 m/s. Therefore our models demonstrate the dynamo induced zonal variations are about of 10 percent of magnitude of the mean latitudinal shear. The relative variations of the meridional circulation are about 1 percent of the mean flow which is much smaller than it is observed on the Sun. Note, the M-dwarf has much denser plasma than the Sun and for the 1kG magnetic field at the top of the integration domain ($0.98R_{\star}$) the Alfven velocity is less than $6$m/s. In the model the toroidal field does not penetrate to the surface because of the vacuum boundary conditions and this reduces the magnitude of the large-scale flow variations on the surface. Unlike the Sun (see, eg, \citealt{2011JPhCS.271a2001B,2011JPhCS271a2074H,2013ASPC..479..395K}) the predicted torsional oscillations have the equal magnitudes in the bulk of the star and at the surface. Variations of the meridional circulation are concentrated to the surface. Note , that the radial profile of the meridional circulation is still unclear in the case of the Sun, see preliminary results in the papers by \citet{hath12} and \citet{Zhao13m}, who supports concentration of 11-th year variations of the solar meridional circulation to the surface. It is predicted that magnetic activity produces rather strong distortion of the angular velocity profile inside the star leaving the structure of the meridional flow nearly the same as it is in the kinematic models. The same results were found in the direct numerical simulations of \citet{2008ApJ676.1262B} and \citet{2015ApJ813L31Y}. Figure \ref{fig:modes}a allows comparison to their results. We find that in the magnetic case (the model M5) the latitudinal shear persists only in the upper layer of the star. Also, there the strong radial shear presents near the equator. The same was found in the direct numerical simulation by \citet{2015ApJ813L31Y}. The model of \citet{2008ApJ676.1262B} showed the uniform angular velocity profile in the magnetic case. We find that in the nonlinear model M5 the positive radial shear in the equatorial region is stronger than in the kinematic model M1. Also we see formation of the radial shear at the surface in the polar region in the model M5. The increase of the magnitude of the subsurface shear as a result of the magnetic field influence on the angular momentum fluxes is also in agreement with the recent numerical simulations on the solar-like stars (\citealt{guer2013,2014AA...570A..43K,2016ApJ819.104G}). \begin{figure} \includegraphics[width=1\columnwidth]{dr-modes}\caption{\label{fig:modes}a) The angular velocity radial profile in the kinematic (red lines) and nonlinear models M1 and M5 for the equator ($0^{\circ}$) and 60$^{\circ}$ latitudes; b)Modes.} \end{figure} Our results show that the strength of the surface poloidal magnetic field is only factor two or three lesser than the strength of the toroidal magnetic field inside the star, see the Table 1. All the models show rather strong polar magnetic field, 1kG in the kinematic models and from 100 to 500 G in the nonlinear models. Current observations of the stellar magnetic activity inform us a lot about the topological and spectral properties of the magnetic field distributions at M-dwarfs and cool stars \citep{M2-2010MNRAS,2016MNRAS1129S}. Figure \ref{fig:modes}b presents results of the spherical harmonic decomposition for magnetic field predicted by the fully nonlinear models M4 (axisymmetric one) and M5. In the axisymmetric model M4 we don't expect any toroidal field out of the surface because of the boundary conditions. In this case the energy of the magnetic field outside the star is dominated by $\ell=3$ and $\ell=5$ harmonics which is similar to the Sun \citep{sten88,2013AARv2166S,2016MNRAS4591533V}. The non-axisymmetric dynamo model M5 show the dominance of the mode m=1 and $\ell=1$ of the large-scale toroidal magnetic field. The ratio of the energy of the non-axisymmetric and axisymmetric poloidal magnetic field in the model M5 is about factor order of the magnitude. The given results are in agreement with \citet{M2-2010MNRAS} for the magnetic field observations for the early types of the M-dwarfs with a moderate rotation rates. Let's summarize the main findings of the paper. Our study confirm the previous conclusions of \citet{shul15} that the weak differential rotation of the M-dwarfs can support the axisymmetric dynamo especially for the case $Pm_{T}>1.$ For the case $Pm_{T}=3$ we find that the generation threshold $\alpha$-effect parameter $C_{\alpha}$ is lower for axisymmetric magnetic field. However for the overcritical $\alpha$-effect the non-axisymmetric dynamo become preferable. The situation is reproduced both in the kinematic and in the fully nonlinear dynamo models. In the non-linear case the differential rotation of the star deviates strongly from the kinematic case. For the most complete non-linear dynamo model we found the non-axisymmetric magnetic field of strength about 0.5kG at at the surface mid latitude, it is rigidly rotating and it is perturbed by the axisymmetric dynamo waves propagating out of the rotational axis. The predicted dynamo period of the axisymmetric dynamo waves in the model is about 40 Yr for the $Pm_T=3$ and it is longer for the higher $Pm_T$. \subsection*{Acknowledgements} I appreciate Prof D.D.~Sokoloff, Prof D.~Moss and Dr D.Shulyak for discussions and comments. I thank the financial support of the project II.16.3.1 of ISTP SB RAS and the partial support of the RFBR grants 15-02-01407-a, 16-52-50077-jaf. | 16 | 9 | 1609.00906 |
1609 | 1609.00301_arXiv.txt | We have searched for the signature of cosmic voids in the CMB, in both the Planck temperature and lensing-convergence maps; voids should give decrements in both. We use \zobov\ voids from the DR12 SDSS CMASS galaxy sample. We base our analysis on $N$-body simulations, to avoid {\it a posteriori} bias. For the first time, we detect the signature of voids in CMB lensing: the significance is \tcb{$3.2\sigma$}, close to $\Lambda$CDM in both amplitude and projected density-profile shape. A temperature dip is also seen, at modest significance (\tcb{$2.3\sigma$}), with amplitude about 6 times the prediction. This temperature signal is induced mostly by voids with radius between 100 and 150 $\Mpc$, while the lensing signal is mostly contributed by smaller voids -- as expected; lensing relates directly to density, while ISW depends on gravitational potential. The void abundance in observations and simulations agree, as well. We also repeated the analysis excluding lower-significance voids: no lensing signal is detected, with an upper limit of about twice the $\Lambda$CDM prediction. But the mean temperature decrement now becomes non-zero at the \tcb{$3.7\sigma$} level (similar to that found by Granett et al.), with amplitude about 20 times the prediction. However, the observed dependence of temperature on void size is in poor agreement with simulations, whereas the lensing results are consistent with $\Lambda$CDM theory. Thus, the overall tension between theory and observations does not favour non-standard theories of gravity, despite the hints of an enhanced amplitude for the ISW effect from voids. | In a $\Lambda$CDM universe, dark energy stretches cosmic voids, causing their gravitational potential to decay. Photons from the Cosmic Microwave background (CMB) then lose energy when traversing a void, so that the CMB temperature is expected to be colder when a void sits along the line of sight. This is the Integrated Sachs-Wolfe effect (ISW: \citejap{Sachs1967}), and its detection would give direct evidence of dark energy, at least for large voids that evolve quasi-linearly. But this imprint has not been detected with unquestionable significance, owing to the large effective noise term from the superimposed primordial CMB temperature fluctuations. This noise can be reduced by stacking CMB imprints from many voids, and several papers have followed such a strategy \citep{Granett08, Ilic2014, Cai2014, PlanckISW2014, PlanckISW2015, Hotchkiss2015}. The highest S/N measurement of this kind was reported in \citeauthor{Granett08} (\citeyear{Granett08}; G08). They stacked the WMAP7 temperature maps for 50 voids from the SDSS DR6 galaxy sample, yielding a temperature decrement of approximately $-10 \mu K$ at the $3.7\sigma$ level. This signal is rather high compared to expectations from $\Lambda$CDM, but the result was reproduced with the same G08 catalogue using Planck CMB temperature maps \citep{PlanckISW2014, PlanckISW2015}. A limitation of G08 is that their voids were found in a photometric redshift catalogue, with large redshift uncertainties compared to spectroscopic redshift samples. But the photometric-redshift smearing may even help to detect very elongated structures along the line of sight, which may have the highest ISW signals \citep{Granett2015}. Our goal in the present study is therefore to conduct a similar analysis using the larger SDSS-DR12 CMASS spectroscopic redshift sample, which covers the same redshift range ($0.4<z<0.7$) and the same volume as that of the DR6 photometric redshift sample in the NGC region. We also include the SGC region from the CMASS sample in this study. A number of ISW searches using voids from the SDSS DR7 spectroscopic redshift samples at low $z$ found less significant results, i.e. at around the $2\sigma$ level \citep{Ilic2014, Cai2014, PlanckISW2014}, or a null detection \citep{Hotchkiss2015}. All of these studies used the \zobov\ algorithm \citep{Neyrinck2005,Neyrinck08} to find voids. The variety of results reported by different groups is largely due to the differences in the way void catalogues are pruned. This suggests that the details of void selection are important for studies of this kind. A number of factors may affect the stacked ISW signal. First, voids found in the galaxy field may not necessarily correspond to sites of maximal coldness in the ISW signal (potential maxima, in linear theory). There may be spurious voids due to the discreteness of the galaxy sample. Second, the edges of voids in over-dense environments (the so-called voids-in-clouds: \citejap{ShethVDWeygaert2004}) may be contracting. Their underlying potentials are negative rather than positive at the scale of the void, which reverses the sign of the ISW signal \citep{Cai2014}. Finally, it is important to note that the selection of voids has to be conducted on physical grounds {\it prior} to the measurement of the signal. Failure to do so can introduce {\it a posteriori\/} bias and overestimation of the statistical significance of the measurement. The above issues can either introduce noise or cause biases for the ISW signal. They can be reduced to some extent by calibrating the void catalogues using simulations, as demonstrated in \citet{Cai2014}. In this paper we analyse voids found in the SDSS-DR12 CMASS sample, following a procedure similar to that of \citet{Cai2014}. Furthermore, we also carry out a stacking analysis using the Planck lensing convergence map. Even though CMB lensing is dominated by structures at $z\simeq 2$, low-$z$ structures also contribute. Voids should be associated with density minima in order to cause an ISW temperature decrement, and this underdensity should be detectable via CMB lensing. Weak gravitational lensing by voids has been predicted in the literature \citep{Amendola1999, Krause2013, Higuchi2013} and it has been measured using weak galaxy shear \citep{Melchior2014, Clampitt2015, Gruen2016, Sanchez2016}. But for the more distant galaxies, the use of CMB lensing should be a better probe. Evidence for the co-existence of the ISW and CMB lensing signals would help to confirm the reality of each effect. This dual probe is valuable from the point of view of modified gravity, since the two effects are closely related: lensing depends on the sum of metric potentials $\Phi+\Psi$, whereas ISW depends on the time derivative of this same combination. Hints of the general coexistence of both the ISW and CMB lensing signatures have been found by \citet{PlanckISW2015}. This paper also showed some evidence for a mean lensing signal from the G08 supervoids (but not from the G08 superclusters). This is the issue that we intend to explore in more detail, with a larger void sample. This paper is organised as follows: In Section 2, we define our void catalogues and describe our simulations for the ISW and lensing signal associated with voids. Section 3 presents the main results of stacking voids with the CMB temperature map and the lensing convergence map, focusing on the estimation of signal-to-noise. We conclude and discuss our results in Section 4. | By taking voids at $0.4<z<0.7$ from the DR12 SDSS CMASS galaxy sample, and using Planck CMB data, we have measured the stacked CMB temperature ($\Delta T$) and lensing convergence ($\Delta \kappa$) at the void locations. An important aspect of our analysis is to use $N$-body simulations to calibrate the void catalogue, which enables us to select voids with physical motivation without introducing {\it a posteriori\/} bias. We have demonstrated that the simulated voids are good matches to the CMASS void data in terms of abundance, but the simulations also indicate that some of the catalogued voids are not true matter underdensities -- particularly the largest systems, with $r_{\rm v} \gs 150 \Mpc$. In this way, we have found the following results concerning the imprint of voids on the CMB: (1) There is a relatively low (\tcb{2.3$\sigma$}) significance for the void-CMB temperature cross-correlation, which is contributed mainly by large voids with radii greater than $100\Mpc$. The void-CMB lensing association is much stronger, at the \tcb{3.2$\sigma$} level, contributed mostly by smaller voids. Thus we do not detect simultaneous temperature and lensing imprints from the same set of voids. This is not unexpected: if $\Delta T$ is induced by the ISW effect, it would arise from the decay of the gravitational potential, which is a smoothed version of the density field, while the lensing convergence map comes directly from the projected matter density. (2) When interpreted as the ISW signal, our measured $\Delta T$ is a few times larger than expected from a $\Lambda$CDM model (although not strongly inconsistent statistically with the standard-model prediction); but the amplitude of the lensing $\Delta \kappa$ is a very good match to $\Lambda$CDM. Moreover, the projected void profile from observation is consistent with that from our simulations. For the larger voids that show the tentative ISW signal, there is no indication of an enhanced amplitude for the lensing signal; this is of the order of $\Delta \kappa \sim 10^{-3}$ and well within the statistical errors of the Planck lensing map. (3) Our measurement of the stacked void profile is the first to use CMB lensing data; this is more efficient for voids at high redshift, where measurements of weak galaxy lensing are challenging. The good agreement of void abundances between observation and simulations plus the agreement between the observed and simulated void profiles suggest that the detected CMB lensing signal is robust. Accurate measurement of void profiles may provide valuable information for cosmology and gravity. Dark matter void profiles evolve differently in different cosmologies \citep{Demchenko2016}; in certain type of modified gravity, e.g. those with the chamaeleon screening mechanism, voids are expected to be emptier than their GR counterparts. The dark matter profile of voids can therefore provide powerful test for modified gravity \citep{Clampitt2012, Lam2015, Cai2015, Barreira2015}. Our measurement suggests that it is possible to do this with CMB lensing. (4) When repeating the same analyses, removing voids of lower statistical significance gives a null detection in lensing, but the measured $\Delta T$ becomes more strongly non-zero. The amplitude of $\Delta T$ and its significance are both similar to those reported in \citet{Granett08} for voids of this strength. The crucial (and only) factor leading to this result is the selection of voids that are $3\sigma$ deviations in terms of Poisson fluctuations, as in \citet{Granett08}. The level of the temperature signal remains puzzling: for large voids ($r_{\rm v}\approx 125\Mpc$), we find it to be about 20 times the $\Lambda$CDM prediction (albeit with a large uncertainty), which is a larger discrepancy than claimed by \citet{Granett08}. Conversely, there is a {\it positive\/} temperature deviation for voids with $r_{\rm v}\ls 60\Mpc$, which is qualitatively incompatible with our simulations. Such gross discrepancies are not seen in our larger sample of DR12 voids, nor do we see a boosted signal in the lensing by voids (with or without $3\sigma$ thresholding). It therefore seems unlikely that this anomalous temperature result can really be taken as evidence that standard gravity is in error. In particular, our measurements of void lensing argue that $\Lambda$CDM is a good match to observation, even though the temperature signal in this rare void subset remains to be better understood. \clearpage | 16 | 9 | 1609.00301 |
1609 | 1609.01307_arXiv.txt | Over a handful of rotation periods, dynamical processes in barred galaxies induce non-axisymmetric structure in dark matter halos. Using n-body simulations of a Milky Way-like barred galaxy, we identify both a trapped dark-matter component, a \emph{shadow} bar, and a strong response wake in the dark-matter distribution that affects the predicted dark-matter detection rates for current experiments. The presence of a baryonic disk together with well-known dynamical processes (e.g. spiral structure and bar instabilities) increase the dark matter density in the disk plane. We find that the magnitude of the combined stellar and shadow bar evolution, when isolated from the effect of the axisymmetric gravitational potential of the disk, accounts for $>$30\% of this overall increase in disk-plane density. This is significantly larger that of previously claimed deviations from the standard halo model. The dark-matter density and kinematic wakes driven by the Milky Way bar increase the detectability of dark matter overall, especially for the experiments with higher $v_{min}$. These astrophysical features increase the detection rate by more than a factor of two when compared to the standard halo model and by a factor of ten for experiments with high minimum recoil energy thresholds. These same features increase (decrease) the annual modulation for low (high) minimum recoil energy experiments. We present physical arguments for why these dynamics are generic for barred galaxies such as the Milky Way rather than contingent on a specific galaxy model. | Introduction} In the currently favored form of weakly interacting massive particle (WIMP) theory (see e.g. \cite{jungman96,bertone05}), dark matter is composed of a single particle with a mass in the range of 10 $\GeV~$, which a number of experiments are working to directly detect \cite{aprile12,aalseth13,bernabei14,agnese15,angloher16,akerib16,agnes16,amole16,tan16,agnese16,undagoitia16}. Direct-detection (DD) experiments seek to measure the weak nuclear recoils during elastic scattering between dark-matter (DM) particles and the nuclei of a target detector. The unambiguous detection of particle dark matter would address fundamental questions about the nature of the Universe, but despite considerable effort being focused on the direct detection of dark matter, a verifiable signal remains elusive. Limits on WIMP properties derived from these nondetections depend on poorly constrained parameters from astrophysics \cite{mccabe10,mccabe11}. The astrophysical uncertainties in the structure of the DM halo have been recently implicated as a possible resolution for the disagreement between experiments with tentative detections (DAMA/LIBRA and CDMS-Si) and the null results from experiments such as LUX and superCDMS \cite{mao13,pillepich14,bozorgnia16,kelso16,sloane16}. \begin{table*} \centering \begin{threeparttable}[b] \caption{Halo Models\label{tab:simparams}.} \begin{tabular}{lccccc} Model Name & Designation\tnote{a}~~ & Radial Profile & Dynamic? & Core? & Rotation? \\ \hline \hline Standard Halo Model & SHM & isothermal & N & N & N \\ Pristine NFW & pNFW & NFW & N & N & N \\ Adiabatically Contracted NFW & acNFW & NFW & Y\tnote{b} & N & N\\ Fiducial Dynamical NFW & fdNFW & NFW & Y & N & N \\ Cored Dynamical NFW & cdNFW & NFW & Y & Y & N \\ Rotating Dynamical NFW & rdNFW & NFW & Y & N & Y \\ Cored Rotating Dynamical NFW & rcdNFW & NFW & Y & Y & Y \\ \hline \end{tabular} \begin{tablenotes} \item[a] Designations are used in Figures, Model Names are used in text. \item[b] Idealized evolution; see text. \end{tablenotes} \end{threeparttable} \end{table*} Several simulation-based studies of Milky Way-like galaxies (e.g. a multicomponent model featuring at a minimum a stellar disk and responsive DM halo) have determined velocity distributions for the DM halo that differ from the so-called standard halo model (SHM), finding that the spherical density and isotropic velocity distribution assumptions underlying the interpretation of most DD experiments are unlikely to be accurate owing to the presence of substructure in the halo \cite{kuhlen10,purcell12,lisanti12}. Another class of studies primarily focus on the difference between DM-only simulations and simulations that include a stellar component \cite{pillepich14,bozorgnia16,kelso16,sloane16}, finding largely the same results. However, little disagreement exists between these studies regarding the expected response for DD experiments, and the underlying dynamical causes have not been thoroughly investigated. For example, these studies have been unable to reach a consensus on the applicability of a Maxwell-Boltzmann (MB) distribution to describe the DM velocity distribution in the Milky Way (MW) near the Sun, and are roughly divided into groups that claim a MB distribution does describe the tail of the DM velocity distribution \cite{kelso16,bozorgnia16}, and those that find that the tail is suppressed relative to a MB distribution \cite{pillepich14,sloane16}. In addition, the `dark disk', an axisymmetric, flattened DM feature roughly on the size scale of the stellar disk observed in some simulations, comprises an additional component for detection \cite{read08,read09,bruch09,purcell09,ling10,pillepich14}, but its existence continues to be debated. However, as we show in a previous work \cite{petersen15}, a dark disk that mimics the appearance of the stellar disk is a natural consequence of the presence of a stellar disk in a DM halo, something that is obviously present in our own galaxy. The dark disk effect may be enhanced further by the disruption of satellites \cite{pillepich14}, which other studies contend may not be a generic result of cosmological simulations \cite{kelso16}. This scenario is qualitatively different from the dark disk described in \cite{petersen15}. Other studies have claimed that the DM density at the Sun's location should differ by less than 15\% from the average over a constant density ellipsoidal shell using high resolution cosmological simulations \cite{vogelsberger09} and that the density distribution is only slightly positively skewed \cite{kamionkowski08}. Yet other studies point out that many open questions remain regarding the presence of substructure near the Sun owing to either intact or destroyed subhalos \cite{read08,read09,kuhlen12,lisanti12,ohare14}. In the face of these conflicting claims, seeking fundamental effects from known Milky Way (MW) causes is a prudent approach to illuminating the information that DM halo models can provide for DD experiments. Galaxies evolve structurally through the interaction of the baryonic matter in their disks with the DM in their halos mediated by resonant gravitational torques. The strongest evolution of this type is likely to occur in barred galaxies (i.e. galaxies with prolate stellar distributions in their central regions with lengths on the order of the disk scale length). The barred nature of the MW was first suggested in the 1960s as an interpretation of observed gas kinematics \cite{devaucouleurs64}, and subsequently confirmed through diverse observations in the ensuing half century (see \cite{gerhard02} for a review). Recent observations have indicated that the bar hosted by our MW galaxy may be significantly longer than previously thought \cite{wegg15}. Although the MW bar is known to have many consequences for observed astrophysical quantities, the bar's effect on the DM distribution has not been considered when characterizing the DM density and velocity distribution function that determines detection rates for DD experiments. In this paper, we present the implications of non-axisymmetric DM density and velocity distribution functions caused by the bar of the MW for DD experiments. We offer a qualitative analysis of recently published studies in an attempt to unify the seemingly disparate results. In a previous work \cite{petersen15}, we demonstrated that particles in the DM halo will be trapped into a shadow bar that resembles the stellar bar---in addition to forming a DM wake visible in both the density and velocity structure of the dark matter halo at radii on the scale of the stellar disk---the first such study that attempts to isolate the DM structure that results from interactions with the stellar bar. The effect of the shadow bar is cumulative with the expected response of an equilibrium galaxy DM halo to the presence of a stellar disk, resulting in a model for the DM halo that does not resemble the SHM. We will see that bar-driven galaxy evolution affects both the DM density and the kinematics at the Earth's location. Using simulations designed to study the mutual dynamical evolution of the baryonic disk and DM halo for a Milky-Way-like galaxy, we characterize the secular evolution of an initially exponential stellar disk and spherically symmetric dark matter halo. We do not consider any satellite debris or stellar streams at the solar circle \cite{freese04,savage06}, although these may be present. Rather, we detail significant differences from the SHM due to the stellar bar of the MW. Similar to previous studies \cite{mao13}, we find that realistic DM distributions in galactic halos can dramatically increase the predicted detection rates for high $v_{min}$ experiments. Moreover, the effects of long-term evolution in a barred galaxy further increases the tension between heavy and light nuclei experiments \cite{frandsen13}. We demonstrate key regimes in which experiments can use the DM halo structure resulting from the MW bar to their advantage. Conversely, \cite{pillepich14} report an improvement in the tension between the heavy and light nuclei experiments if the detection signal were dominated by a DM debris disk from merger events, which has a sharply decreasing velocity tail. It is possible, of course, that the MW also has a DM debris disk from a merger event. This underscores the importance of the actual MW evolutionary history to DM detection predictions and motivates further detailed study. This paper is organized as follows. In section~\ref{sec:fiducial}, we provide the relevant details about the simulations used for this analysis, including a comparison of the simulations to the MW in section~\ref{subsec:milkywaycompare}. We then describe the results in section~\ref{sec:results}, beginning with the density and kinematic features of the simulated galaxy in section~\ref{subsec:features} before detailing the calculation of detection rates in section~\ref{subsec:detectionrates}. We compare to previous findings in section~\ref{subsec:litmod} (including both the SHM and empirical models), then explore the effect of our results for detection rates in DD experiments (sections~\ref{subsec:experiments} and \ref{subsec:annualmod}). Section~\ref{sec:conclusion} provides a broad overview of our results and prospects for future work. | \label{sec:conclusion} The major results of the paper are as follows: \begin{enumerate} \item The density of the DM halo at the solar position varies depending on the Earth's location relative to the stellar bar. Smaller angles relative to the bar as well as a smaller ratio of $R_{\odot}/R_{\rm bar}$ can increase the density relative to a spherical distribution by a factor of 2. \item The DM velocity profile is reshaped by the stellar+shadow bar. The characteristic quadrupole wake in the DM that forms as a response to the stellar bar lags the bar in velocity and, therefore, enhances the detectability of DM when compared to the SHM (adiabatically contracted NFWmodel) by a factor of 3.5 (2) at $\vmin=300~\kms$. At $\vmin=650~\kms$, detectability relative to the SHM is increased by a factor of 10, and up to a factor of 40 for a cored NFW halo model. Enhancements for initially rotating models are approximately equal to the respective non-rotating model (fiducial dynamical NFW and cored NFW). \item A number of recent astrophysical models suggest the importance of the MW evolutionary history to modeling DM detection rates. As detectability depends on $\vmin$ (which is sensitive to the velocity distribution), and we have demonstrated effects on the velocity distribution from known features in the MW, experiments need to move beyond the SHM to compare with other experiments that have different energy thresholds. \item Similarly, annual modulation in the DM signal will have different detectabilities compared to the SHM as a function of $\vmin$. The stellar+shadow bar, when compared to the adiabatically contracted model, {\it reduces} the annual modulation signal for experiments sensitive to high energy thresholds by approximately 20\%, and {\it boosts} the annual modulation signal for experiments sensitive to low energy thresholds by approximately 20\%. \item When compared to the SHM, we expect an enhancement in detectability and annual modulation. We use an adiabatically contracted model that fixes the gravitational potential of the disk to calibrate the importance of dynamical evolution to the DM detection predictions. For example, when we compare our fiducial dynamical NFW model to the adiabatically contracted NFW model at $\vmin=475~\kms$ (the nominal value for superCDMS at $m_\chi=5$ GeV), we expect an enhancement in detectability of 100\%, but an unchanged annual modulation signal. This illustrates the influence of dynamical evolution. \end{enumerate} The results presented in this paper can be succinctly summarized as indicative that the expected rates of observation for DD experiments is strongly sensitive to realistic DM halos. Models that incorporate known physical processes can be used at a minimum to determine astrophysics-related constraints on DM $m_\chi$ and $\sigma_\chi$. While the literature now has no shortage of simulations touting different halo velocity distributions, the field is still not able to accurately create a MW analogue that accounts for evolutionary history. Acknowledging this fact, in this paper we study the effects of simple dynamical models, implemented through n-body simulations, on DD experiments. We stress that the effects presented in this paper are generic results of the gravitational interaction between the stellar disk and the DM halo. The power in these inferences is a motivation for marrying DD experiments with realistic astronomy. Astronomically realistic models will provide realistic constraints with more power to discriminate between WIMP hypotheses. The change relative to the SHM affect primarily lower $m_\chi$ values. This owes to the low $v_{\rm min}$ values implied by $m_\chi>20$ GeV, allowing experiments to probe nearly the entire $g(v_{\rm min})$ space. In contrast, if $m_\chi<10$ GeV, the discrepancy between our fiducial model and the SHM will be large: $v_{\rm min}$ is in the tail of the $g(v_{\rm min})$ distribution, where we have demonstrated $\left(\Delta g(v_{\rm min})\right)/g(v_{\rm min})$ changes rapidly. The results presented here are by no means an exhaustive parameter search, nor a best-fit MW model. However, the MW is a disk galaxy with a moderate bar. The features induced in the DM distribution by dynamical evolution in our simulations realistically represent those expected in the MW and will obtain generally for any disk galaxy. The density enhancements and velocity asymmetries will have clear impacts on the sensitivities of the various direct-detection experiments and are likely to make the tensions between upper limits and tentative detections stronger and more interesting. Future iterations of direct detection experiments, such as superCDMS (at SNOLAB) \cite{agnese15}, LUX-ZEPLIN \cite{akerib15}, and XENON1T \cite{aprile14}, will build upon the constraints from previous studies. Halo models that accurately account for known dynamical effects in the MW are necessary for meaningful hypothesis testing. Finally, directional detectors will enable a detailed study of the kinematic signature at the solar position. Early efforts may be able to detect a bias in the tangential and radial velocity peaks, as in Figure~\ref{fig:correlatedvel}, which may even prove a discriminating factor for determining the halo profile. This hints at the possibility of DM astronomy in the future. | 16 | 9 | 1609.01307 |
1609 | 1609.03905_arXiv.txt | In the blooming field of exoplanetary science, NASA's \textit{Kepler Space Telescope} has revolutionized our understanding of exoplanets. \textit{Kepler}'s very precise and long-duration photometry is ideal for detecting planetary transits around Sun-like stars. The forthcoming \textit{Transiting Exoplanet Survey Satellite (TESS)} is expected to continue \textit{Kepler}'s legacy. Along with transits, the Doppler technique remains an invaluable tool for discovering planets. The next generation of spectrographs, such as \textit{G-CLEF}, promise precision radial velocity measurements. In this paper, we explore the possibility of detecting planets around hypervelocity and runaway stars, which should host a very compact system as consequence of their turbulent origin. We find that the probability of a multi-planetary transit is $10^{-3}\lesssim P\lesssim 10^{-1}$. We therefore need to observe $\sim 10-1000$ high-velocity stars to spot a transit. However, even if transits are rare around runaway and hypervelocity stars, the chances of detecting such planets using radial velocity surveys is high. We predict that the European \textit{Gaia} satellite, along with \textit{TESS} and the new-generation spectrographs \textit{G-CLEF} and \textit{ESPRESSO}, will spot planetary systems orbiting high-velocity stars. | Discoveries of exoplanets have proliferated in the past decade primarily due to observations with the Doppler technique and transits. Thanks to the high precision achieved with today's spectrographs, Doppler spectroscopy allows for the determination of a planet's minimum mass based upon the shift of stellar absorption lines. \citet{cum08} analysed eight year's worth of radial velocity measurements for nearly $600$ FGKM stars. The fundamental observational quantity is the stellar velocity amplitude induced by the planet \citep{cum04,cum08}. \citet{cum08} showed that $17$-$20$\% of stars have gas giant planets within $20$ AU. \citet{may11} reported the results from an eight year survey using the HARPS spectrograph. They conclude that greater than half of solar-type stars harbour a planet with a period of $\leq 100$ days. Furthermore, they find that $\sim 14$\% of solar-type stars host a planet with mass greater than $50$ M$_{\bigoplus}$. Doppler observations are of vital importance and continue to help in discovering new planets (e.g. \citet{dai16}). However, today transits dominate the search for exoplanets. A transit is the passage of a smaller body in front of a larger body, such as when an exoplanet passes in front of its host star thus producing a drop in brightness \citep{win10}. Several surveys have been dedicated to transits detections, the most important and fruitful of which is NASA's \textit{Kepler Space Telescope}, which has revolutionized exoplanetary science \citep{bor10}. \textit{Kepler}'s original purpose was to determine the frequency and characteristics of planets and planetary systems in the habitable zone around FGKM stars. However, Kepler's very precise and long-duration photometry is ideal for detecting systems with multiple transiting planets \citep{lis11}. NASA's next major exoplanet mission scheduled for launch in 2017 is the \textit{Transiting Exoplanet Survey Satellite (TESS)}. \textit{TESS} is expected to monitor several hundred thousand Sun-like stars for transiting planets across nearly the entire sky using four wide-field cameras \citep{ric15}. \textit{TESS} aims to combine the strengths of wide-field surveys with the fine photometric precision and long intervals of \textit{Kepler}, but compared to Kepler will examine stars that are generally brighter by $3$ mag over a solid angle that is larger by a factor of $\sim 400$ \citep{sul15}. In this paper, we explore the possibility of detecting planets around high-velocity stars using both the Doppler technique and transits. High-velocity stars are most often Galactic halo stars with high peculiar motions, usually divided in two different categories, runaway stars (RSs) and hypervelocity stars (HVSs). RSs are historically defined as Galactic young halo stars with peculiar motions higher than $40$ km s$^{-1}$, which are thought to have travelled to the halo from their birthplace. RSs are produced in binary systems thanks to dynamical multi-body interactions or due to the velocity kick from a supernova explosion \citep{sil11}. HVSs, on the other hand, are stars escaping the Galaxy. \citet{hil88} was the first to predict the existence of HVSs, while \citet{brw05} discovered the first HVS in the outer halo. Hills' mechanism involves the tidal breakup of a binary passing close to a massive Black Hole (BH) \citep*{gil06,gil07,brw15,frl16}. Other mechanisms have, also, been proposed to explain the existence of HVSs, as the interaction of a massive binary black hole with a single star \citep{yut03}, or the interaction of star clusters and BHs \citep*{cap15,fra16,fck16}. Observations of high velocity and hypervelocity objects have usually been limited to high-mass, early-type, stars, due to observational bias \citep*{brw14}. However, recently observers have begun investigating low-mass HVSs candidates \citep*{lii12,pal14,fav15}. The European Space Agency (ESA) satellite \textit{Gaia} is expected to measure proper motions with an unprecedented precision, providing a larger and less biased sample ($\sim 100$ new HVSs in a catalogue of $\sim 10^9$ stars). Moreover, \textit{Gaia}'s sensitivity is good enough to search for multi-planet systems around massive stars and evolved stars and reveal their architecture and three-dimensional orbits \citep{cas08,wif15}. Furthermore, the detection of a planet around a HVS or runaway star will provide valuable information on the survivability of planets in extreme environments \citep*{gin12}. Thus, in this paper we look at the likelihood of finding transits around such high-velocity stars. In Section 2 we discuss our approach to calculating transits including a discussion on the code we used. In Section 3 we explore various possibilities and discuss our outcomes. In Section 4 we explain additional difficulties that are inherent in observing multi-planet transits. In Section 5 we discuss the Doppler technique. We conclude with a discussion and implications for future observations in Section 6. | \label{sec:con} In this paper we have computed the likelihoods of finding exoplanets around high-velocity stars. We considered different stellar masses $M_*$, number of planets $N_P$, mean planetary inclinations $\sigma_i$ and eccentricities $\sigma_e$. We found that the geometrical probability of detecting a transit has generally an increasing trend with $\sigma_e$ as consequence of Eq.\ref{eqn:probab}, and decreases with $\sigma_i$. This indicates that the transit method may detect all of the planets in multi-planet systems if the planetary orbits are nearly lined up with the star \citep{lis11,bra16}. On the other hand, having a larger number of planets around less massive stars reduces the probability of spotting a transit. We considered a semi-major axis in the range $0.015 \le a/\mathrm{AU} \le 0.5$ since the planetary system must be compact to be retained after strong gravitational encounters \citep{gin12}. The joint probability of a multi-planetary transit is $10^{-3}\lesssim P\lesssim 10^{-1}$ and depends upon the semi-major axis range with $p_T\propto a^{-1}$. If we assume $\eta\sim 1$, Eq.\ref{eqn:numbp} predicts that we need to observe $\sim 10-1000$ stars to spot a transit. \textit{TESS} is expected to spot transiting planets across nearly the entire sky by monitoring several hundred thousand Sun-like stars \citep{sul15}. In particular, \textit{TESS} should be able to find transits around hypervelocity and runaway stars. For HVSs, the \textit{Gaia} satellite is expected to find $\sim 100$ such stars \citep{ken14,brw15,deb15}. If we assume that each HVS hosts a compact planetary system, Eq.\ref{eqn:numbp} predicts that at least one transit could be spotted. Even if computations lead to larger geometric probabilities, a transit must take into account the relative flux decrement $(R_p/R_*)^2$ \citep{win10}. If we consider a Jupiter-sized planet and a HVS of typical mass $3$ M$_{\odot}$, the flux decrement is proportional to $k\propto (R_J/R_*)^2\sim 0.1\%$. If the star's mass is $\lesssim 1$ M$_{\odot}$, $k\sim 1\%$. While the geometrical probability favors heavier stars, the relative flux decrement indicates that the S/N ratio is larger in the case of small stars. Assuming $\sigma\sim 1$\%, a Jupiter-sized planet transiting a $1$ M$_{\odot}$ star have $\gtrsim 90$\% of probability to be observed. \textit{Gaia} will spot $\sim 100$ HVSs within few kpc from the Sun. Transits are likely to be observed around such HVSs with $\gtrsim 90$\% probability if they host planets. Along with transits, the Doppler technique is an important tool for finding planets around high-velocity stars. Equation \ref{eqn:krad} indicates that the stellar velocity amplitude induced by the planet is $K\sim 10$-$10^2$ m s$^{-1}$, with larger values for FGKM stars. The detection threshold depends on the number and duration of the observations and on the noise sources \citep{cum08}. Generally the detection threshold is $\gtrsim 1$ m s$^{-1}$, and depends on the duration of the RV survey and on the planet's orbital period and eccentricity \citep{cum04}. The typical radial velocity accuracy ($\sigma \sim 1$ m s$^{-1}$) of modern spectrograph yield $> 50$\% probability of measuring the Doppler shift caused by a hot-Jupiter, whereas Earth-sized planets require long survey durations and large number of observations. In general, compact planetary systems with large planetary minimum masses around low-mass stars can generate a high Doppler signal, that can be measured with $>50$\% probability by the next-generation spectrographs. As discussed, transits are particularly suited for spotting Jupiter-sized exoplanets around low-mass high-velocity stars. On the other hand, RV surveys are able to observe not only massive planets, but also Earth-sized planets orbiting massive high-velocity stars provided that the duration of the survey is large enough and that the Doppler signal is measured several times. Thus, a combination of Doppler spectrographs, such as \textit{G-CLEF} and \textit{ESPRESSO}, working together with \textit{TESS} will hopefully lead to the discovery of planets around high-velocity stars and consequently result in new understandings of planetary formation, evolution, and survivability. | 16 | 9 | 1609.03905 |
1609 | 1609.04008_arXiv.txt | We have resolved the scatter-broadened image of PSR~B0329+54 and detected substructure within it. These results are not influenced by any extended structure of a source but instead are directly attributed to the interstellar medium. We obtained these results at 324~MHz with the ground-space interferometer RadioAstron which included the space radio telescope (SRT), ground-based Westerbork Synthesis Radio Telescope and 64-m Kalyazin Radio Telescope on baseline projections up to 330,000~km in 2013~November~22 and 2014~January~1~to~2. At short 15,000 to 35,000~km ground-space baseline projections the visibility amplitude decreases with baseline length providing a direct measurement of the size of the scattering disk of $4.8\pm0.8$~mas. At longer baselines no visibility detections from the scattering disk would be expected. However, significant detections were obtained with visibility amplitudes of~3~to~5\% of the maximum scattered around a mean and approximately constant up to 330,000~km. These visibilities reflect substructure from scattering in the interstellar medium and offer a new probe of ionized interstellar material. The size of the diffraction spot near Earth is $17,000\pm3,000$~km. With the assumption of turbulent irregularities in the plasma of the interstellar medium, we estimate that the effective scattering screen is located $0.6\pm0.1$ of the distance from Earth toward the pulsar. | All images of radio sources from outside our solar system are influenced by scattering in the interstellar medium (ISM) of our Galaxy. Determining the properties of the scattering is essential for studying the characteristics of the ISM and for a proper interpretation of astronomical radio observations. Pulsars are an almost ideal type of a celestial source for such studies. They are almost point-like so that the results of the study are not influenced by the structure of the celestial source but are almost completely attributable to the influence of the plasma turbulence of the ISM. Scattering of the pulsar signal in the ISM results in angular broadening of the pulsar image, temporal broadening of the pulses, modulation or scintillation of their intensities and, through diffraction patterns, distortion of their radio spectra. Scattering effects of the ISM have been the subject of studies by several authors. Theoretical studies were made by, e.g., \citet{prokhorov1975, rickett1977,goodman1989, narayan1989, gwinn1998}. Observational studies were made with the ground-based VLBI for extragalactic radio sources \citep[e.g.,][]{ojha2006,lazio2008,pus2013,PK2016}, the galactic center Sgr~A$^*$~\citep[e.g.,][]{Lo1998,Bower2006,gwinn2014,PK2016}, and pulsars \citep[e.g.,][]{kondratiev2007,gwinn1993,desai1992,bartel1985}. PSR~B0329+54 is the brightest pulsar below 1~GHz in the northern hemisphere. With a galactic longitude of $145^\circ$ and latitude of $-1\fdg2$, and at a parallax distance of $1.03^{+0.13}_{-0.12}$~kpc \citep{brisken2002}, the pulsar is located just at the outer edge of the Orion spiral arm. The scattering disk remained unresolved. An early upper limit of the angular size at 2.3~GHz is $\theta_\mathrm{scat}<1$~mas \citep{bartel1985}. Extending baselines into space, the first ground-space VLBI observations of a pulsar were made with the VLBI Space Observatory Program (VSOP) by \cite{yangalov2001}. The observed pulsar was PSR~B0329+54. With the VSOP observations its scattering disk was not resolved. The upper limit of the angular size at 1.7~GHz was determined to be $\theta_\mathrm{scat}<2$~mas \citep{yangalov2001}. However, the VSOP observations were also done at an insufficiently low frequency and with baselines only up to about 25,000~km. Ground-space VLBI with RadioAstron allows observations at a frequency as low as 324~MHz, where scattering effects are expected to be much stronger and with baselines about 15 times longer. Since the scattering size increases as $\nu^{-2}$, the size of $\theta_\mathrm{scat}<2$~mas at 1.7~GHz and the size of $\theta_\mathrm{scat} < 1$~mas at 2.3~GHz from the ground-ground VLBI observations correspond to a size of $<50$~mas at 324~MHz. At this frequency RadioAstron has an angular resolution of about 1~mas at its longest baselines. Therefore observations with RadioAstron have the potential to resolve PSR~B0329+54's scattering disk and perhaps reveal hitherto unknown structure in the ISM. Here we report on such investigations with RadioAstron with baselines up to 330,000~km at a frequency of 324~MHz. This paper is the second one in a series. The first paper \citep[Paper I,][]{gwinn+2016} gives a mathematical description of the functions obtained from the interferometer observations and leading to an evaluation of the scattering structure in the interstellar medium along the line of sight to PSR~B0329+54. In this second paper we focus on the visibility magnitude as a function of projected baseline length and on a comparison between angular and temporal broadening. This will allow us to draw conclusions about the size of the scatter-broadened image of PSR~B0329+54 as well as the characteristics of the diffraction spot near Earth and the distance of the scattering screen. | Here we summarize our observations and results and give our conclusions. \begin {enumerate} \item We made VLBI observations of PSR B0329+54 with RadioAstron at 324 MHz on projected baselines up to 330,000~km or 350~M$\lambda$. Our goal was to investigate scattering properties of the ISM which affect radio observations of all celestial sources. While the results of such observations are in general influenced by the convolution of source structure with the scattering processes, pulsars are virtually point-like sources and signatures in the observational results can be directly related to the ISM scattering properties. \item Visibility function at short ground-ground baselines manifests a single bright spike in delay-rate space that vanishes on long space-ground baselines. Thus, the scattering disk of PSR B0329+54 was completely resolved on ground-space baselines of 15,000 to 30,000 km. The FWHM of the angular diameter is $4.8\pm0.8$~mas at 324~Hz. \item The diffractive length scale or size of the diffraction spot near Earth is $17,000\pm3,000$~km. \item With the assumption of turbulent and large-scale irregularities in the plasma, the effective scattering screen is located at $d/D=0.6\pm0.1$ or somewhat more than half of the distance from Earth to the pulsar. \item At longer projected baselines, up to 330,000 km, significant visibility amplitudes were detected, although none were expected from the scattering disk. They are scattered around a mean which stays approximately constant up to the longest baselines. This result indicates that substructure was discovered in the scatter-broadened image of PSR~B0329+54. \end{enumerate} | 16 | 9 | 1609.04008 |
1609 | 1609.08967_arXiv.txt | {I calculate the rate of WIMP capture and annihilation in the Earth in the non-relativistic effective theory of dark matter-nucleon interactions.~Neglecting operator interference, I consider all Galilean invariant interaction operators that can arise from the exchange of a heavy particle of spin less than or equal to one when WIMPs have spin 0, 1/2 or 1.~I compute position and shape of the expected resonances in the mass - capture rate plane and show that Iron is not the most important element in the capture process for many currently ignored interaction operators.~I compare these predictions with the recent results of an Earth WIMP analysis of IceCube in the 86-string configuration and set limits on all isoscalar and isovector coupling constants of the effective theory of dark matter-nucleon interactions.~For certain interaction operators and for a dark matter particle mass of about 50 GeV, I find that these limits are stronger than those I have previously derived in an analysis of the solar WIMP search performed at IceCube in the 79-string configuration.} | \label{sec:intro} Understanding the nature of dark matter is an increasingly important research question in Astroparticle Physics~\cite{Bertone:2010at}.~The search for a first unambiguous non-gravitational signal of dark matter is currently pursued through a variety of complementary approaches~\cite{Bertone:2004pz}.~In the standard paradigm of Weakly Interacting Massive Particles (WIMPs) as a dark matter candidate, WIMPs can be detected via scattering by nuclei in underground laboratories (direct detection), through their annihilation or decay in space (indirect detection), or through WIMP production at particle accelerators such as the Large Hadron Collider (LHC)~\cite{Jungman:1995df,Bergstrom:2000pn,Catena:2013pka}.~At the interface of WIMP direct and indirect detection is the search for energetic neutrinos from the annihilation of WIMPs captured in the Sun or Earth via scattering by nuclei~\cite{Silk:1985ax}, or self-interactions~\cite{Zentner:2009is,Catena:2016ckl}. Crossing the Sun or Earth, WIMPs might lose energy via local interactions, and scatter from gravitationally unbound to gravitationally bound orbits.~In this scenario, WIMPs are expected to accumulate at the Sun's or Earth's centre through subsequent scattering events.~The accumulation of WIMPs at the centre of a celestial body leads to an increase in the local WIMP density.~As a result, WIMPs eventually annihilate at a potentially observable rate, producing Standard Model particles, and in particular neutrinos.~Neutrino observatories such as IceCube, Super-Kamiokande, ANTARES, BAKSAN, and Baikal are currently testing this hypothesis~\cite{Aartsen:2016exj,Choi:2015ara,Adrian-Martinez:2016gti,Boliev:2013ai,Avrorin:2014swy}.~In this study, I primarily focus on the capture and annihilation of WIMPs in the Earth.~For a recent solar WIMP analysis of neutrino telescopes in effective theories see~\cite{Catena:2015iea,Blumenthal:2014cwa,Liang:2013dsa,Guo:2013ypa}. The first pioneering studies of WIMP capture and annihilation in the Earth by Freese \cite{Freese:1985qw} and others~\cite{Krauss:1985aaa,Gaisser:1986ha,Gould:1987ir} assumed the Earth to be in free space.~Corrections due to the Sun's gravitational field~\cite{Gould:1987ww}, WIMP diffusion in the solar system~\cite{Gould:1991rc}, solar depletion~\cite{Gould:1999je,Lundberg:2004dn}, and WIMP weak scattering in the Sun~\cite{Sivertsson:2012qj} have subsequently been studied in detail.~It has been found that the free space approximation implies a relative error on the capture rate of at most 35\%, and for specific WIMP masses only.~For WIMP masses close to the mass of an element in the Earth, the relative error on the capture rate tends to zero~\cite{Sivertsson:2012qj}.~So far, the expected neutrino flux from WIMP annihilation in the Earth has been computed for the standard spin-independent dark-matter nucleon interaction only.~Here, I assume the Earth to be in free space and extend previous calculations to virtually arbitrary WIMP-nucleon interactions. In this work, I compute the rate of WIMP capture and annihilation in the Earth in the non-relativistic effective theory of dark matter-nucleon interactions, formulated in~\cite{Chang:2009yt,Fan:2010gt,Fitzpatrick:2012ix,Fitzpatrick:2012ib} and developed in~\cite{Fornengo:2011sz,Menendez:2012tm,Cirigliano:2012pq,Anand:2013yka,DelNobile:2013sia,Klos:2013rwa,Peter:2013aha,Hill:2013hoa,Catena:2014uqa,Catena:2014hla,Catena:2014epa,Gluscevic:2014vga,Panci:2014gga,Vietze:2014vsa,Barello:2014uda,Catena:2015uua,Schneck:2015eqa,Dent:2015zpa,Catena:2015vpa,Kavanagh:2015jma,D'Eramo:2016atc,Catena:2016hoj,Kahlhoefer:2016eds}.~Neglecting operator interference, the theory includes all Galilean invariant dark matter-nucleon interaction operators that can arise from the exchange of a heavy particle of spin less than or equal to one for WIMPs of spin 0, 1/2 and 1.~For spin 1 WIMPs, two interactions not considered here can arise if operator interference is not negligible~\cite{Dent:2015zpa}.~I compute WIMP capture and annihilation rates considering eleven elements in the Earth's mantle and core, and using nuclear response functions obtained in~\cite{Catena:2015uha} for $^{16}$O, $^{23}$Na, $^{24}$Mg, $^{27}$Al, $^{28}$Si, $^{32}$S, $^{40}$Ca, $^{56}$Fe, and $^{58}$Ni and in this work for $^{31}$P and $^{52}$Cr.~I compare my calculations with the 90\% CL upper limits on the WIMP annihilation rate found in a recent WIMP analysis of IceCube in the 86-string configuration~\cite{Aartsen:2016fep}.~Through this comparison, I set limits on the isoscalar and isovector coupling constants of the non-relativistic effective theory of dark matter-nucleon interactions.~For certain interaction operators and for a dark matter particle mass of about 50 GeV, these limits are stronger than those I have previously found~\cite{Catena:2015iea} in an analysis of the solar WIMP search performed at IceCube in the 79-string configuration~\cite{Aartsen:2012kia}. The paper is organised as follows.~In Sec.~\ref{sec:earth} I introduce the theoretical framework used to calculate the rate of WIMP capture and annihilation in the Earth.~I perform this calculation in Sec.~\ref{sec:results}, where using data from IceCube in the 79 and 86-string configuration~\cite{Aartsen:2012kia,Aartsen:2016fep}, Super-Kamiokande~\cite{Choi:2015ara} and LUX~\cite{Akerib:2013tjd}, I set 90\% CL upper limits on the coupling constants of the effective theory of Sec.~\ref{sec:earth}.~I conclude in Sec.~\ref{sec:conclusions}.~Appendix~\ref{sec:appDM} contains key equations, while in Appendix~\ref{sec:nuc} I describe the nuclear shell model calculation through which I derive the $^{31}$P and $^{52}$Cr nuclear response functions.~Finally, I collect in Appendix~\ref{app:figures} figures for capture rates and exclusion limits relative to interaction operators which for brevity are not discussed in the body of the paper. | \label{sec:conclusions} I have studied the capture and annihilation of WIMP dark matter in the Earth in the effective theory of dark matter-nucleon interactions.~It is the first time that the neutrino signal from WIMP annihilation in the Earth's interior is investigated in this general theoretical framework. Computing the rate of WIMP capture in the Earth I have used nuclear response functions derived through numerical shell model calculations partly in~\cite{Catena:2015uha}, and partly in this work (i.e.~for $^{31}$P and $^{52}$Cr).~For all operators and coupling constants in the effective theory, I have computed the position and shape of the predicted resonances in the corresponding WIMP mass - capture rate plane.~I have found that Iron is not the most important element in the capture process for many interaction operators.~The number of resonances, and their relative high also drastically depend on the interaction operator in analysis.~A variety of factors are relevant in this calculation, ranging from the dependence on the momentum transfer of the WIMP-nucleon interaction to the Earth's composition and associated nuclear physics inputs. Next, I have calculated the rate of WIMP annihilation in the Earth in the effective theory of dark matter-nucleon interactions.~I have compared this prediction with the 90\% CL upper limits on the same rate from a WIMP analysis of IceCube in the 86-string configuration~\cite{Aartsen:2016fep}.~Through this comparison, I have set 90\% CL upper limits on all isoscalar and isovector coupling constants in Eq.~(\ref{eq:H_chiT}).~For comparison, I have also derived limits on the same coupling constants by demanding that the predicted neutrino flux from WIMP annihilation in the Sun is not larger than the corresponding 90\% CL upper limit from observations performed at Super-Kamiokande~\cite{Choi:2015ara}.~For WIMPs with a mass of about 50 GeV, I find that present Earth WIMP searches at IceCube in the 86-string configuration place comparable or even stronger constraints on the strength of the $\hat{\mathcal{O}}_1$, $\hat{\mathcal{O}}_3$, $\hat{\mathcal{O}}_{11}$, $\hat{\mathcal{O}}_{12}$ and $\hat{\mathcal{O}}_{15}$ interactions than current searches for solar WIMPs at neutrino telescopes in general.~This is in particular true for interaction operators that can generate a large nuclear response for WIMP-Iron scattering. | 16 | 9 | 1609.08967 |
1609 | 1609.03072_arXiv.txt | Air-Cherenkov telescopes have mapped the Galactic plane at TeV energies. Here we evaluate the prospects for detecting the neutrino emission from sources in the Galactic plane assuming that the highest energy photons originate from the decay of pions, which yields a straightforward prediction for the neutrino flux from the decay of the associated production of charged pions. Four promising sources are identified based on having a large flux and a flat spectrum. We subsequently evaluate the probability of their identification above the atmospheric neutrino background in IceCube data as a function of time. We show that observing them over the twenty-year lifetime of the instrumentation is likely, and that some should be observable at the $3\,\sigma$ level with six years of data. In the absence of positive results, we derive constraints on the spectral index and cut-off energy of the sources, assuming a hadronic acceleration mechanism. | Introduction} The IceCube experiment has discovered a flux of high-energy neutrinos from extragalactic sources with an energy density similar to that observed for gamma rays ~\cite{halzen:2016}. The observation underscores the important role of sources accelerating protons that produce similar energy in photons and neutrinos, which are the decay products of neutral and charged pions, respectively. The present data cannot exclude a subdominant flux of Galactic origin in the IceCube data~\cite{Taylor:2014hya,Gaggero:2015xza,Ahlers:2015moa,Palladino:2016zoe}. Unidentified sources \cite{Fox:2013oza}, Fermi bubbles~\cite{Taylor:2014hya,Lunardini:2011br,Lunardini:2013gva}, and Sagittarius~A$^*$~\cite{Bai:2014kba} have been reviewed as potential Galactic sources. However, the general conclusion is that these sources can account for a fraction of the events detected. Specifically, the possibility that the hot spot close to the Galactic Center (GC) is produced by a single point source with a flux normalization of $6~\times~10^{-8}~\rm{GeV}~\rm{cm}^{-2}~\rm{s}^{-1}$ has been excluded~\cite{Adrian-Martinez:2014wzf,Gonzalez-Garcia:2013iha}. In a map of the northern Galactic plane obtained with Milagro data, six promising neutrino sources were identified in Refs.~\cite{Halzen:2008zj,GonzalezGarcia:2009jc}. The IceCube Collaboration has carried out extensive searches for point and extended sources in Ref.~\cite{Aartsen:2014cva}, reporting evidence with a significance of $2.5~\sigma$, when the six Milagro sources are considered together~\cite{Aartsen:2014cva}. In the previous study, Ref.~\cite{Gonzalez-Garcia:2013iha}, the authors revisited the prospects for observing the three confirmed Milagro sources and re-evaluated the probability and constraints in light of the low-energy cut-off reported by the Milagro collaboration \cite{Abdo:2012jg,Smith:2010yn}. The authors concluded that more than 10 years of running IceCube is necessary to yield a discovery at the level of $3\sigma$. In the case of the source MGRO J1908+06, evidence at $3\sigma$ could be obtained in seven years assuming values of the spectral index and the cut-off energy that are in good agreement with the best fit reported in~\cite{Abdo:2012jg}. In this paper, we will update the theoretical predictions using the observation and flux measurements reported by HAWC, ARGO-YBJ, and air Cherenkov telescopes (ACT) VERITAS and HESS. Most importantly, with a detector superior to Milagro, the HAWC experiment has confirmed only four of the six sources~\cite{Abeysekara:2015qba,hawc_gamma}: MGRO J1908+06, MGRO J1852+01, MGRO J2031+41, and MGRO J2019+37. For these, we will construct a gamma ray spectrum based on all information available and evaluate the neutrino flux. Subsequently, we will compute the number of signal and background events as well as the p-value for observing the sources as a function of time. Finally, we will determine exclusion limits on a flux of hadronic origin in the absence of an observation. Our main results can be summarized as follows: \begin{itemize} \item MGRO J1908+06: Although historically classified as a pulsar wind nebula (PWN) and currently as an unidentified source, its large size and hard spectrum in TeV photons suggest that it may be a supernova remnant (SNR). SNRs are suspected to be the sources of the highest energy cosmic rays in the Galaxy. We re-evaluate the probability of observing the source using the flux reported by HESS and anticipate a $3\sigma$ observation in about 10 years of IceCube data. However, the answer depends on the actual threshold of the specific analysis. By increasing the energy threshold, IceCube has the potential to observe MGRO J1908+06 at the some statistical level with only six years of data. A lack of observation in 15 years of IceCube data will indicate that MGRO J1908+06 is not a cosmic-ray accelerator. \item MGRO J1852+01: In the original Milagro map of the TeV sky, this source missed the statistical threshold for candidate sources. It has now been conclusively observed by HAWC and is a potential neutrino source considering its relatively large flux. Since the proper study of spectrum and extension of the source have not been performed by HAWC, we have studied the neutrino flux under different assumptions for the source's extension and spectrum. We find that IceCube should see this source in 5 years of data provided that the source is not extended. However, if the source is extended, 15 years is required to reach a significant level of observation. \item MGRO J2031+41: Due to the uncertainties associated with the origin of the flux of the Cygnus cocoon and $\gamma$-Cygni, a complete picture of this source is missing. Its extension and other TeV emissions in its vicinity have made it difficult for ACT experiments like VERITAS to measure the TeV flux from this source. Although previous studies indicated that observing the source would be challenging~\cite{Gonzalez-Garcia:2013iha}, using recent ARGO-YBJ and Fermi data, we argue that neutrino observations at the level of $3\sigma$ may be possible in 10 years of IceCube data. \item MGRO J2019+37: We present an update on the neutrino observation from this source based on the spectrum measured by VERITAS, which has provided up to now the most precise measurement for the spectrum of the source up to 30 TeV. We show that IceCube is likely to observe the source in 15 years. This source is currently classified as a PWN. Thus, the detection of neutrinos from this region could point towards the production mechanism of neutrinos in a PWN as described in Ref. \cite{Lemoine:2014ala}. \end{itemize} | Conclusions } The highest energy survey of the Galactic plane has been performed by Milagro. This survey has identified bright sources in the nearby Cygnus star-forming region and in the inner part the Galaxy. Initially, the sources showed the expected behavior of PeVatrons. PeVatrons are the sources of cosmic rays in the "knee" region of the cosmic-ray spectrum that are expected to be sources of pionic gamma rays whose spectrum extends to several hundreds of TeV without a cut-off. Gamma rays from the decay of neutral pions are inevitably accompanied by neutrinos with a flux that is calculable. In this paper, we re-evaluated the probability of observing four promising Milagro sources in IceCube. We used the updated information from air-Cherenkov and air-shower array experiments to estimate the flux of neutrinos. The prospects for observing these sources in IceCube is highly entangled with discrepancies in the detailed fluxes and morphologies measured by different experiments. Moreover, the uncertainty of the nature of these sources makes it more difficult to understand the observed spectrum. Different spectra and morphology of the sources correspond to different production mechanisms. It should be noted that the discrepancy between measurements may arise from the difference in angular resolution between air-shower arrays and air-Cherenkov telescopes as well as from the range of energies in which they operate. Future results from HAWC will help resolve these discrepancies and reveal more information about the sources. If the gamma rays are hadronic in origin, observation of an accompanying neutrino flux is likely over the lifetime of the IceCube experiment. Evidence from IceCube of neutrinos associated with these sources will greatly help in unraveling the nature of the sources. \clearpage \begin{figure}[!t] \center \begin{tabular}{rl} \includegraphics[width=0.48\textwidth]{milagro2.pdf} &% \includegraphics[width=0.48\textwidth]{MuonsSource2.pdf} \end{tabular} \caption{\label{fig:sources_spectra_second} {\it \underline{Left panel:}} We show in purple the data by HESS~\cite{Aharonian:2009je}, in red the one from VERITAS~\cite{Aliu:2014rha}, and in cyan the one from HAWC~\cite{Abeysekara:2015qba}. In blue we show the previous flux measurements by Milagro~\cite{Abdo:2007ad,Abdo:2009ku}, while the solid orange line and the shaded orange area show the best fit and the $1\sigma$ band as reported in Ref.~\cite{Smith:2010yn} by Milagro. The dotted area is the ARGO-YBJ $1\sigma$ band~\cite{ARGO-YBJ:2012goa}. With green lines we show the spectra obtained considering $\alpha_\gamma=2$ and fixing the normalization to the best fit reported in Table~\ref{tab:sources_fit}, where we also allowed the cut-off energy to vary: $E_{\rm cut, \gamma} =30,~300, and~800$~TeV (short-dashed, solid, and long-dashed lines, in green). {\it \underline{Right panel:}} We show the corresponding number of events for these spectra. The gray band encodes the uncertainty on the cut-off energy. With the black (gold dashed) line, we show the background from atmospheric neutrinos for extended (point-like) sources. } \end{figure} \begin{figure}[!t] \centering \begin{tabular}{rl} \includegraphics[width=0.48\textwidth]{SS_milagro2_v2.pdf} & \includegraphics[width=0.48\textwidth]{energy_threshold.pdf} \end{tabular} \caption{\label{fig:sources_pvalues_second} {\it \underline{Left panel:}} p-values as a function of time, from 4 years to 20 years. The spectra have been fixed, as shown in Fig.~\ref{fig:sources_spectra_second}. The gray band encodes the uncertainty due to different values of $E_{cut,\gamma}$, and morphology, see Table.~\ref{tab:r_bin}. For the green lines we have considered the case of extended source. {\it \underline{Right panel:}} Dependence of the p-value on the energy threshold $E_{\nu}^{th}$. } \end{figure} \begin{figure}[!t] \begin{tabular}{rl} \includegraphics[width=0.48\textwidth]{milagro4.pdf} & \includegraphics[width=0.48\textwidth]{MuonsSource4.pdf} \end{tabular} \caption{\label{fig:sources_spectra_forth} {\it \underline{Left panel:}} We show in blue the value on the flux reported by the Milagro collaboration~\cite{abdo}, which assumed an $E^{-2.6}$ spectrum. With green lines we show the spectra obtained considering $\alpha_\gamma=2$ and fixing the normalization to the best fit reported in Table~\ref{tab:sources_fit}, where we also allowed the cut-off energy to vary: $E_{\rm cut, \gamma} =30,~300, and~800$~TeV (short-dashed, solid, and long-dashed lines, in green). {\it \underline{Right panel:}} Number of events for the spectra reported with green and blue lines in the left panel. The gray band encodes the uncertainty on the cut-off energy. With the black (gold dashed) line, we show the background from atmospheric neutrinos for extended (point-like) sources. } \end{figure} \begin{figure}[!t] \centering \begin{tabular}{rl} \includegraphics[width=0.48\textwidth]{SS_milagro4.pdf} & \includegraphics[width=0.48\textwidth]{SS_milagro4_point.pdf} \end{tabular} \caption{\label{fig:sources_pvalues_forth} {\it \underline{Left panel:}} p-values as a function of time, from 4 years to 20 years. The spectra have been fixed, as shown in Fig.~\ref{fig:sources_spectra_forth}. The gray band encodes the uncertainty due to different values of $E_{cut,\gamma}$. We assume the source to be extended. {\it \underline{Right panel:}} We assume the source to be point-like. } \end{figure} \begin{figure}[!t] \includegraphics[width=0.85\textwidth,height=8cm]{milagro3.pdf} \vspace{0.5cm} \\ \hspace{-1cm} \includegraphics[width=0.48\textwidth]{MuonsSource3.pdf} \includegraphics[width=0.48\textwidth]{SS_milagro3.pdf} \caption{\label{fig:sources_spectra_third} {\it \underline{Upper panel:}} The black points show the data reported by ARGO-YBJ in Ref.~\cite{Argo:2014tqa}, while the dotted region is the one reported in Ref.~\cite{Bartoli:2012tj}. The previous flux measurements by Milagro are shown in blue~\cite{Abdo:2007ad,Abdo:2009ku}, while the orange/yellow area denotes the the power-law model/the power-law model with cut-off as reported in Ref.~\cite{Abdo:2012jg} by Milagro. With the purple band we report the measurements by MAGIC~\cite{Albert:2008yk}. We report in red and grey the results from the VERITAS detector~\cite{A.WeinsteinfortheVERITAS:2014iwa}. With green/magenta lines we show the spectra obtained fixing the parameters to the best fit reported in Table~\ref{tab:sources_fit} for the case without/with Fermi data. In the case without Fermi data, we also allowed the cut-off energy to vary: $E_{\rm cut, \gamma} =30,~300, and~800$~TeV (short-dashed, solid, and long-dashed lines, in green. {\it \underline{Lower panel, left:}} Number of events for the spectra reported with green and magenta lines in the upper panel. The gray band encodes the uncertainty on the cut-off energy. With black lines, we show the background from atmospheric neutrinos. {\it \underline{Lower panel, right:}} p-values as a function of time, from 4 years to 20 years. } \end{figure} \begin{figure}[!t] \includegraphics[width=0.48\textwidth]{milagro1.pdf} \\ \begin{tabular}{rl} \includegraphics[width=0.48\textwidth]{MuonsSource1.pdf} & \includegraphics[width=0.48\textwidth]{SS_milagro1.pdf} \\ \end{tabular} \caption{\label{fig:sources_spectra_first} {\it \underline{Upper panel:}} With red points, we report the VERITAS data~\cite{Aliu:2014xra}. With blue lines, we report the previous flux measurements by Milagro~\cite{Abdo:2007ad,Abdo:2009ku}, while the continuous orange line and the shaded orange area represent the best fit and $1\sigma$ band~\cite{Abdo:2012jg} as reported by Milagro. The 90\% C.L. upper limits from ARGO-YBJ are shown in black~\cite{Bartoli:2012tj}, and the inferred CASA-MIA bound~\cite{Beacom:2007yu} is shown with a black star. With green lines we show the spectra obtained fixing the parameters to the best fit reported in Table~\ref{tab:sources_fit}, where we also allowed the cut-off energy to vary: $E_{\rm cut, \gamma} =30,~300, and~800$~TeV (short-dashed, solid, and long-dashed lines, in green). {\it \underline{Lower panel, left:}} Number of events for the spectra reported with green and magenta lines in the left panel. The gray band encodes the uncertainty on the cut-off energy. With the black (gold dashed) line, we show the background from atmospheric neutrinos for extended (point-like) sources. {\it \underline{Lower panel, right:}} p-values as a function of time, from 4 years to 20 years. } \end{figure} \begin{figure}[!t] \centering \begin{tabular}{rl} \includegraphics[width=0.48\textwidth]{CL_Source_2.pdf} & \includegraphics[width=0.48\textwidth]{CL_Source_4.pdf} \vspace{0.5cm} \\ \includegraphics[width=0.48\textwidth]{CL_Source_3.pdf} & \includegraphics[width=0.48\textwidth]{CL_Source_1.pdf} \end{tabular} \caption{\label{fig:sources_fixedNorm_2} {\it \underline{Upper panel:}} Values of $\alpha_\gamma$ and $E_{cut, \gamma}$ excluded at 95\% (solid) and 99\%~C.L. (dot-dashed) with 15 years of IceCube running with its 86-string configuration. The normalization has been fixed to the best fit reported by HESS~\cite{Aharonian:2009je} (left) and Milagro~2007 (right). We have assumed extended sources. With horizontal lines we denote the values $E_{\rm cut, \gamma} =30,~300, and~800$~TeV (short-dashed, solid, and long-dashed lines, in green). The purple region (left) denotes the values of $\alpha_\gamma$ reported by HESS. The blue line (right) denotes the value of $\alpha_\gamma$ considered by Milagro. {\it \underline{Lower panel:}} The normalization has been fixed to the best fit reported by ARGO-YBJ without Fermi-LAT~\cite{Bartoli:2012tj} (left) and VERITAS~\cite{Aliu:2014rha} (right). We have assumed extended sources. The gray/magenta region (left) denotes the values of $\alpha_\gamma$ reported by ARGO-YBJ without/with Fermi data. The red region (right) denotes the values of $\alpha_\gamma$ reported by VERITAS. } \end{figure} \clearpage | 16 | 9 | 1609.03072 |
1609 | 1609.03591_arXiv.txt | We report the detection of stellar eclipses in the LP 661-13 system. We present the discovery and characterization of this system, including high resolution spectroscopic radial velocities and a photometric solution spanning two observing seasons. LP 661-13 is a low mass binary system with an orbital period of $4.7043512^{+0.0000013}_{-0.0000010}$ days at a distance of $24.9 \pm 1.3$ parsecs. LP 661-13A is a $0.30795 \pm 0.00084$ $M_\odot$ star while LP 661-13B is a $0.19400 \pm 0.00034$ $M_\odot$ star. The radius of each component is $0.3226 \pm 0.0033$ $R_\odot$ and $0.2174 \pm 0.0023$ $R_\odot$, respectively. We detect out of eclipse modulations at a period slightly shorter than the orbital period, implying that at least one of the components is not rotating synchronously. We find that each component is slightly inflated compared to stellar models, and that this cannot be reconciled through age or metallicity effects. As a nearby eclipsing binary system where both components are near or below the full-convection limit, LP 661-13 will be a valuable test of models for the structure of cool dwarf stars. | The M dwarf spectral sequence spans a large range of mass, from 0.6 M$_{\odot}$ at the earliest spectral types down to the main sequence turn off at approximately 0.08 M$_{\odot}$. This mass range spans important transitions in the physical structure of the interior of these stars. Notably, these stars transition to the fully convective regime midway through the spectral sequence, at $0.35 M_{\odot}$ \citep{structure_convection}. These transitions must be accurately captured in stellar models and their effect on the equations of stellar structure must ultimately be reflected in the temperatures and radii of these stars. However, the fundamental properties of low-mass stars remain a significant challenge for stellar structure models, particularly below 0.35 solar masses \citep{Torres_Structure,New_Baraffe_Models_2015}. Testing the mass-radius relation for low-mass stars is traditionally done through the study of eclipsing binary (EB) systems. Precise radial velocity measurements taken throughout the orbit are sensitive to the component masses of the system, while measurements of the eclipse depths and shapes are sensitive to the radii of the eclipsing stars. The best-observed, detached, double-lined EBs can provide measurements accurate at the $1\%$ level, allowing these systems to become strong tests of current stellar models \citep{Torres_2010_Review}. One of the closest and most well studied eclipsing binary systems is CM Draconis (CM Dra). CM Dra is an eclipsing M dwarf binary at a distance of 14.5 parsecs from the Sun \citep{CM_Dra_discovery,Lacy77}. CM Dra also contains a white dwarf at a wide separation from the M dwarf pair. As instrumentation and modeling have improved, the masses and radii of the two M dwarfs are now measured at the $0.5\%$ level \citep{Metcalfe96,Morales09}. Both stars in the CM Dra eclipsing binary are spectral type dM4.5 with masses of 0.23 and 0.21 solar masses and radii of 0.25 and 0.24 solar radii and orbit in a 1.7 day orbit \citep{Morales09}. The radii of these stars are inflated at the $5\%-7\%$ level \citep{Morales09}, and this remains a problem even with the latest stellar models \citep{New_Baraffe_Models_2015}. This problem is not restricted to a handful of systems. Taking only the most well-measured eclipsing binary systems in aggregate, low-mass stars tend to be \emph{inflated} in radius and \emph{cooler} in temperature than stellar models predict \citep{Torres_Structure}. The number of low mass stars with stellar radii measured through interferometry is low \citep{interferometry}, and often do not have any direct means to measure a precise mass. High precision measurements of stellar masses and radii for individual stars are obtained through measurements of eclipsing binary light curves and radial velocities (RVs). Photodynamical analyses of recently discovered triple systems (such as KOI-126, \citealt{carter2011} and Kepler-16 \citealt{Kepler16}) allow even more accurate physical parameters measurements than classical eclipsing binaries due to the presence of eclipses between all three members of the system. However, since the eclipse probability and the probability of detection are strong functions of orbital separation, these systems tend to be dominated by close-in binaries. This makes them more susceptible to the effects of tidal forces between the stars and makes them likely to be tidally locked, preventing the stars from spinning down over their lifetime. This effect makes it more likely for these systems to be magnetically active and to remain significantly magnetically active over their main-sequence lifetimes. If magnetic activity can significantly affect the interior structure of low-mass stars, then this can create an observational bias in the observed radii of these stars, especially if these effects are not accounted for in models. One way around this problem is to search for low-mass eclipsing binary systems that have sufficiently long periods that both stars essentially evolve as ``single" stars, and collect enough high-quality data to constrain their physical parameters to sufficient accuracy to test existing stellar models. While difficult, several such systems have recently been discovered and characterized. The Kepler-16 system consists of a 41 day period eclipsing binary system orbited by a planet in a 229 day orbit, all exhibiting mutual occultations of each other \citep{Kepler16}. The presence of both stellar occultations and planetary transits in this unique system allowed for extremely precise physical parameters of this system to be measured. The secondary star in the system is a 0.202 M$_\odot$, 0.226 R$_\odot$ star with a mass and radius measured with sub-1\% precision. Our group has discovered a 41-day eclipsing binary, LSPM J1112+7626, consisting of two M-dwarfs. These stars have masses of $M_1 = 0.395$ M$_\odot$ and $M_2 = 0.275$ M$_\odot$ and radii of $R_1 = 0.382$ R$_\odot$ and $R_2 = 0.300$ R$_\odot$ \citep{Irwin_41day}. The masses of each component are measured with 0.5\% precision and the radii with 1.5\% precision. In addition to these long period systems, several other eclipsing binary systems with low-mass stellar components have recently been found with periods in the 5-20 day range \citep{2013ApJ...768..127S,2014A&A...572A..50G,2015MNRAS.451.2263Z}. Some of the components in these systems show radii that are consistent with stellar models, while some show significant radius inflation. Assessing the cause of the radius inflation phenomenon in low mass stars requires the discovery of sufficient numbers of these systems with sufficiently different physical characteristics (orbital separation, metallicity, etc) as well as accurate determinations of the mass and radius of each component. MEarth is an ongoing photometric survey of mid-to-late M dwarfs in the solar neighborhood (Distance, $D \lesssim 33$ pc), looking for low mass rocky planets whose periods may extend into the habitable zone of their star \citep{Nutzman,Berta_2013,jonathan_cool_stars}. The MEarth-North array in Arizona has been in operation since 2008, and a copy located in Cerro Tololo, Chile has been in operation since early 2014. By virtue of being designed to be sensitive to small planets transiting these stars, MEarth is also highly sensitive to eclipsing binary systems. Here we present the discovery of an eclipsing binary system revealed during the commissioning phase of the MEarth South array. This system shows out of eclipse modulations due to star spots which change between observing seasons. Through long-term out of eclipse monitoring, we are able to assess the impact that transient starspots have on our ability to measure the radii of each component, which in turn allows us to more reliably probe the physical parameters of this system and assess our errors. We utilize multiple eclipse measurements with the MEarth telescopes as well as radial velocity (RV) measurements in order to constrain the masses and radii of both components to high accuracy and test existing stellar models. In section 2, we detail the MEarth-South array, the discovery, and the follow-up observations of this system. In section 3, we present a joint analysis of the RV and photometric data and constrain the physical parameters of the system. In section 4, we discuss the implications of these measurements in regards to existing theoretical stellar models. | We present here the discovery and analysis of the eclipsing M dwarf - M dwarf binary LP 661-13. We have collected 2 years of eclipse data and precise radial velocity measurements of both components in order to obtain accurate, model-independent measurements of their masses and radii. We find that LP 661-13A is a $0.30795 \pm 0.00084$ $M_\odot$ star with a $0.3226 \pm 0.0033$ $R_\odot$ radius while LP661-13B is a $0.19400 \pm 0.00034$ $M_\odot$ star with a $0.2174 \pm 0.0023$ $R_\odot$ radius. Both components are slightly inflated in radius when compared to stellar models. However, the radius sum (which is much better constrained) is significantly (4 $\sigma$) inflated when compared to the expected radius sum from stellar models. Because the orbit of the system is circularized, it is unlikely that this inflation can be explained through youth. Metallicity is also insufficient to explain the total radius inflation we observe. In the future, the most straightforward way to improve the measurements of this system is to continue out of eclipse monitoring and obtain additional eclipse observations. Because we have observed some spot evolution between observing seasons, additional evolution will help to break the degeneracies between starspot coverage and inferred stellar radii in the model and provide better constraints on the fundamental parameters of this system. Additionally, we can potentially probe the origin of the radius inflation by investigating the marginal X-ray activity as seen by ROSAT and attempt to measure its surface magnetic field. Eclipse measurements in other photometric bandpasses will allow us, with the trigonometric parallax distance we have in hand, to measure the effective temperatures of each component as well, which will serve as another test of stellar models. LP 661-13 represents another low-mass stellar test case measured with high accuracy and will be a useful benchmark for current and future stellar models. LP 661-13 is positioned equatorially on the sky and therefore is a good object for further study from both northern and southern facilities. | 16 | 9 | 1609.03591 |
1609 | 1609.03558_arXiv.txt | Turbulence is a key ingredient for the evolution of the intracluster medium, whose properties can be predicted with high resolution numerical simulations. We present initial results on the generation of solenoidal and compressive turbulence in the intracluster medium during the formation of a small-size cluster using highly resolved, non-radiative cosmological simulations, with a refined monitoring in time. In this first of a series of papers, we closely look at one simulated cluster whose formation was distinguished by a merger around $z \sim 0.3$. We separate laminar gas motions, turbulence and shocks with dedicated filtering strategies and distinguish the solenoidal and compressive components of the gas flows using Hodge-Helmholtz decomposition. Solenoidal turbulence dominates the dissipation of turbulent motions ($\sim 95\%$) in the central cluster volume at all epochs. The dissipation via compressive modes is found to be more important ($\sim 30 \%$ of the total) only at large radii ($\geq 0.5 ~r_{\rm vir}$) and close to merger events. We show that enstrophy (vorticity squared) is good proxy of solenoidal turbulence. All terms ruling the evolution of enstrophy (i.e. baroclinic, compressive, stretching and advective terms) are found to be significant, but in amounts that vary with time and location. Two important trends for the growth of enstrophy in our simulation are identified: first, enstrophy is continuously accreted into the cluster from the outside, and most of that accreted enstrophy is generated near the outer accretion shocks by baroclinic and compressive processes. Second, in the cluster interior vortex stretching is dominant, although the other terms also contribute substantially. | \label{sec:intro} The rarefied media in galaxy clusters (ICMs) are highly dynamic and likely to be turbulent, with strong motions on many scales that can significantly influence a wide range of ICM physical processes \citep[e.g.,][]{2006PhPl...13e6501S,su06,bl07,2011MmSAI..82..588J}. These motions may be driven by processes originating on galactic scales (e.g., star burst winds, AGN outflows and bubbles, \citep[e.g.,][]{2009ApJ...694.1317O,2010MNRAS.407.1277M,2012ApJ...750..166M,2012ApJ...746...94G}), possibly ICM-based magneto-thermal instabilities \citep[e.g.,][]{2011MNRAS.410.2446K,zu13}, but especially by cluster-scale processes associated with cluster formation out of cosmological, large-scale structure \citep[e.g.,][]{do05,va06,ry08,lau09,va11turbo,zu11,miniati14,sc14}. The resulting ICM driving motions on scales that range up to at least 100s of kpc will generally include weak-to-moderately-strong shocks and hydrodynamic shear, both of which are expected to lead to turbulent motions that cascade downwards towards dissipation scales. The solenoidal motions will stretch and fold structures, so are primarily responsible for amplifying and tangling the ICM magnetic field \citep[e.g.,][]{pjr15,bm15}. The compressive turbulence component will, itself, produce weak shocks that can, in turn, generate solenoidal motions \citep[e.g.,][]{pjr15}. Both compressive and solenoidal turbulent components may accelerate cosmic rays through second-order Fermi processes \citep[e.g.,][]{2003ApJ...584..190F,bb05,bl07,2016MNRAS.458.2584B}. Several previous simulation efforts have measured the energy ratio between compressive and solenoidal motions in the ICM, finding a predominance of solenoidal motions \citep[e.g.,][ ]{ry08,iapichino11,va14mhd}. Interplay between the turbulence and shocks may be important in other respects, as well. For instance, turbulent amplification of magnetic fields by shocks and associated second-order Fermi acceleration leading to radio relic emission has been explored in several recent studies \citep[][]{ib12,do16,ji16,fujita15,fujita16, do16}. The relative contributions from solenoidal and compressive turbulent components will depend on the manner in which the turbulence is generated \citep{fed10,pjr15} and its intensity \citep{vazq94}. In the ICM, each of these conditions is likely to vary significantly in both space and time. The present work is motivated particular by the primary need to establish {\it when, where, how and at what level} the two turbulence components are produced and what is their relation to cluster formation dynamics. Here we focus on the turbulence itself, postponing its applications to subsequent works. We focus on turbulence generation, both solenoidal and compressive, and its connections to local ICM dynamical conditions. This complements previous simulation studies that have examined the global energetics of ICM turbulence evolved during cluster formation, including its association with major merger activity. Particularly when issues such as magnetic field amplification and ICM dissipative processes, including cosmic-ray acceleration, are involved and when their dependences on local conditions are important \citep[e.g.][]{su06,bj14}, it can be essential to separate solenoidal from compressive turbulent motions. For example, in recent work \citet{miniati15} showed that the cluster-wide ICM compressive turbulence component is likely to have a steep (Burgers-law-like) spectrum, greatly reducing the power available for cosmic ray acceleration compared to a Kraichnan-like spectrum unless that power can cascade to very small scales, where it can more efficiently transfer energy to the cosmic rays. In order to establish and evaluate the physical roles of turbulence it is essential to separate truly turbulent, uncorrelated flows from correlated, large scale bulk motions and shocks. Uncorrelated flows cascade energy and vorticity to small scales where they work to amplify magnetic fields and dissipate into heat and nonthermal particle energy. Coherent flows, on the other hand, carry signatures of global dynamical events, but are less directly connected to dissipation and magnetic field development. Power spectra and structure functions constructed from simulation cluster-wide velocity fields typically suggest outer coherence scales $\sim 1$ Mpc \citep[e.g.,][]{vbk09,miniati14}. While these scales correctly capture dominant, energy containing processes for the entire cluster, they do not, as emphasized above, necessarily discriminate against non-random, so, non-turbulent motions. They also span highly inhomogeneous, often stratified volumes whose motions on moderate to small scales are often too separated to be well connected causally when local driving conditions vary abruptly in response to nonspherical accretion or interactions (including mergers) with halos. So the ability of such global statistics to represent turbulent motions on the scales where they are most influential is limited. In that context, a more ``local'' approach seems better motivated. One strategy of this kind was suggested by \citet{va12filter} and \citet{va14mhd}. We will follow this strategy here in order to understand more clearly the generation, evolution and dissipation of the solenoidal and compressive turbulent motions produced during cluster formation. The following section outlines our simulations. Section 3 provides a summary of the several analysis tools we employ in this work, while Section 4 presents results of these analyses applied to a selected cluster simulation. Section 5 provides a brief summary and conclusion. | Understanding the dissipation of turbulent energy is key to understand the heating of the plasma, the acceleration of cosmic rays in the ICM, as well as the growth of intracluster magnetic fields \citep[e.g.][]{su06,bj14,miniati15}. While current X-ray line spectroscopy can provide only upper limits on the chaotic motion velocities in relatively bright cluster cores \citep[e.g.,][]{2011MNRAS.410.1797S, 2015A&A...575A..38P}\footnote{The Hitomi satellite in its short life did successfully measure velocity profiles for the Perseus cluster \citep{hitomi}}, future X-ray satellites with superior spectral resolution (e.g. {\small ATHENA}) should be able eventually to detect directly the driving-scale turbulent motions in the ICMs of multiple clusters \citep[][]{2013ApJ...777..137N,2013arXiv1306.2322E,2013MNRAS.435.3111Z,zu16b}. In the meantime, turbulent motions in the ICM induce moderate pressure fluctuations that may be detected in X-rays \citep[e.g.,][]{sc04,2012MNRAS.421..726S,2012MNRAS.421.1123C,2014A&A...569A..67G,2014Natur.515...85Z}, or through the S-Z effect \citep[e.g.,][]{Khatri16}. Numerical simulations of the ICM are fundamental to assessing the real impact of ICM turbulence on all the above. In this work, we focused on the analysis of the connection between accretion-driven shock waves and turbulent motions in the ICM. In particular, we explored both the local and the statistical causal connections between shocks and the emergence of solenoidal and compressive turbulent motions during a simulated cluster lifetime. Our main conclusions from this study can be summarized as follows: \begin{itemize} \item Gas flows in the ICM are characterised by a turbulent behaviour across a wide range of scales, roughly consistent with a Kolmogorov-like model. However, these flows are mixed with larger-scale regular (correlated) velocity components for scales $\geq 0.1-1~\rm Mpc$ and are punctuated by small-scale velocity perturbations due to shocks, which makes it difficult to isolate accurately uncorrelated turbulent fluctuations of the flow at most scales (Sec.~\ref{subsec:results_maps}). \item Using Hodge-Helmoltz decomposition within domains distributed across our simulated cluster, we measure dominant solenoidal velocity fields everywhere within the cluster and at most epochs (with the exception of high redshift epochs, when the cluster is still forming and is far from a virialised state). The solenoidal component makes $\geq 50-80 \%$ of the amplitude of the total velocity field at most epochs and scales (Sec.~\ref{subsec:results_maps}). \item The kinetic energy dissipation rate of the small-scale velocity field is a powerful tool to measure the ratio of compressive and solenoidal motions in a nearly scale-independent way. The dissipation in compressive modes only accounts for a few percent of the total turbulent dissipation rate in the central $\sim \rm ~Mpc^3$ volume. This can increase to about $\sim 15\%$ in the central $\rm Mpc^3$ during major merger events, and to $\sim 30 \%$ in cluster outskirts (Sec.~\ref{subsec:results_maps}). \item Vorticity and enstrophy are trustworthy proxies of the dominant solenoidal turbulent component. In particular, the volume-integrated dissipation rate of solenoidal turbulence and of enstrophy are very well correlated in the $0 \leq z \leq 1$ redshift range, and they show remarkably similar spatial patterns (Sec.~\ref{subsec:enstanal}). \item For the first time, we apply the Navier-Stokes formalism to analyse in detail how enstrophy evolves in the simulated ICM, by decomposing its growth rate into advective, stretching, compressive and baroclinic terms (Sec.~\ref{subsec:enstanal}). \item At accretion shocks baroclinic generation of enstrophy along with enstrophy enhancement during flow compression are the most important source terms of enstrophy. In cluster interiors vortex stretching dominates the growth of enstrophy, although advective concentration of enstrophy and, especially during mergers, enstrophy enhancement through compression can be comparable. Merger shocks largely seed the enstrophy enhanced by vortex stretching and advective concentration in the cluster interior (Sec.~\ref{subsec:enstanal}). \end{itemize} The study of this first cluster of the ISC sample showed how rich is the complexity of simulated ICM turbulence, even in this rather restricted physical setup. Our analysis suggests that a careful combination of filtering techniques is mandatory to identify all major components of the turbulent energy budget reliably, and to give them a physical meaning as a function of scale. Through the extensive analysis of our full ISC sample in planned follow-up work it will be possible to generalise the results obtained for this first cluster in a more robust statistical way. | 16 | 9 | 1609.03558 |
1609 | 1609.03843_arXiv.txt | We present the study of a sample of nine QSO fields, with damped-Ly$\alpha$ (DLA) or sub-DLA systems at $z\sim0.6$, observed with the X-Shooter spectrograph at the Very Large Telescope. By suitably positioning the X-Shooter slit based on high spatial resolution images of HST/ACS we are able to detect absorbing galaxies in 7 out of 9 fields ($\sim$ 78\% success rate) at impact parameters from 10 to 30 kpc. In 5 out of 7 fields the absorbing galaxies are confirmed via detection of multiple emission lines at the redshift of DLAs where only 1 out of 5 also emits a faint continuum. In 2 out of these 5 fields we detect a second galaxy at the DLA redshift. Extinction corrected star formation rates (SFR) of these DLA-galaxies, estimated using their H$\alpha$ fluxes, are in the range 0.3--6.7 M$_\odot$ yr$^{-1}$. The emission metallicities of these five DLA-galaxies are estimated to be from 0.2 to 0.9 Z$_\odot$. Based on the Voigt profile fits to absorption lines we find the metallicity of the absorbing neutral gas to be in a range of 0.05--0.6 Z$_\odot$. The two remaining DLA-galaxies are quiescent galaxies with SFR $<$ 0.4 M$_\odot$ yr$^{-1}$ (3$\sigma$) presenting continuum emission but weak or no emission lines. Using X-Shooter spectrum we estimate i-band absolute magnitude of $-19.5\pm0.2$ for both these DLA-galaxies that indicates they are sub-L$^\star$ galaxies. Comparing our results with that of other surveys in the literature we find a possible redshift evolution of the SFR of DLA-galaxies. | Inter-galactic Medium (IGM) gas accretion and galactic outflows are processes invoked in large scale structure simulations to regulate the galaxy growth \citep{Springel05,Sijacki07,Booth09,Oppenheimer10,Haas13,Vogelsberger14_1,Schaye15}. Active galactic nuclei (AGN) and supernovae winds are the processes which control the galaxy growth in respectively, bright and faint end of the luminosity function (LF) \citep{van-de-Voort11,Puchwein13}. While outflows are observationally ubiquitous in star forming galaxies over cosmic time \citep{Lehnert96,Heckman00,Martin05,Weiner09,Nestor11,Martin12} there are only few examples of galactic inflow observations \citep{Rubin12,Bouche13}. Currently there have been successful progresses in implementing the feedback processes in models of galaxy formation and evolution \citep{Ceverino09,Piontek11,Dalla-Vecchia12,Hopkins-p12,Simpson-c15}. However the true nature of such processes remain poorly understood which is mainly due to the lack of stringent observational constraint on the distribution of gas and metals around galaxies. The circumgalactic medium (CGM) at the interface between the IGM and galaxies extends from approximately 20 -- 300 kpc around galaxies. The CGM is a complicated site of entwined gas from the IGM accretion and galactic outflows through which the baryon exchange between the IGM and galaxies occurs \citep{Suresh15}. Hence, it provides a suitable site to study modes of gas accretion \citep{Keres05,Keres09,Stewart11}, galactic outflows and chemical enrichment of the IGM \citep{Oppenheimer06,Simcoe06,Ryan-Weber09,Wiersma10,Steidel10,Tumlinson11,Dodorico13,Shull14}. Absorption systems in the spectra of background QSOs are among the best tools to study CGM of galaxies over a wide range of redshifts and physical conditions. In particular a subclass of QSO absorbers with Hydrogen column densities of $\log$ (N (\HI) [cm$^{-2}$]) $\gtrsim$ 17.0 called the Lyman limit systems (LLS) likely traces the cool ($T \sim10^4$ K) phase of CGM \citep{Tytler82,Sargent89,Prochaska99,Songaila10}. There exist signatures of bimodality in the metallicity distribution of such systems that may represent galactic outflows in the metal rich branch and cold accretion in the metal poor branch \citep[submitted]{Lehner13,Quiret16}. However, accurate metallicity determinations are difficult for the lower N(\HI) LLS due to uncertainties in N(\HI) measurements while they fall in the flat part of the curve of growth and ionization correction. Two subclasses of higher column density LLS with $\log$ (N (\HI) [cm$^{-2}$]) $\geq$ 20.3 and 19.0 $\leq$ $\log$ (N (\HI) [cm$^{-2}$]) $<$ 20.3 are known as Damped Lyman-$\alpha$ absorbers (DLA) and sub-DLAs, respectively \citep{Peroux01,Wolfe05}. Owing to such high column density of \HI\ they produce Lorentzian wings in their Lyman-$\alpha$ absorption profiles and are dominantly neutral due to self shielding \citep{Wolfe86,Petitjean92,Dessauges-Zavadsky03,Meiring07}. Optical spectroscopic surveys of QSOs have demonstrated that these absorbers dominate the mass density of neutral gas in the universe \citep{Wolfe05,Prochaska05,Noterdaeme09dla,Noterdaeme12dla,Zafar13,Crighton15,Sanchez-Ramirez16}. Therefore, they form the main neutral gas reservoir for the formation of stars at high redshift. % Better constraint on the physical state of the CGM or ISM can be obtained where the absorbing galaxies are detected in emission \citep[e.g.,][]{Moller98a,Moller02,Chen05,Bouche07_simple,Fynbo10,Rao11,Peroux11a,Krogager12,Bouche12,Bouche13,Schroetter15,Kacprzak15}. \citet{Rao11} studied a large sample of absorbing galaxies, primarily identified based on photometric redshifts, to demonstrate that there exists a statistically significant anti-correlation between the N(\HI) and the impact parameter \citep[see also][for similar studies]{,Moller98a,Christensen07,Monier09,Krogager12}. \citet{Peroux11a} used VLT/SINFONI to search for host galaxies of DLAs and sub-DLAs at $z\sim1$ and found them to be faint galaxies with low SFR of the order of a few M$_\odot$ yr$^{-1}$. Benefiting from 3D observations they extracted morpho-kinematic properties of such galaxies that allowed unraveling the nature of absorbing gas to be outflows or extensions of a rotating disk \citep[also see][]{Schroetter15}. In a similar study \citet{Bouche13} could trace the low metallicity inflowing gas seen in absorption at $\sim26$ kpc from the host galaxy at $z\sim$2.3. Higher rates of star formation ($\gtrsim10$ M$_\odot$ yr$^{-1}$) but with smaller detection rates have been measured for DLA host galaxies at $z\sim2$ \citep{Noterdaeme12,Peroux12,Fynbo13}. There exists further evidences that indicate SFR of DLA-galaxies decreases from $z\sim2$ to $z\sim1$ \citep{Bouche12}. However, such results are limited by the low success rates in detection of DLA-galaxies which is usually attributed to the faint nature of DLA-galaxies near the position of bright background QSOs. The identification of a substantially large sample of DLAs at low redshifts is difficult because the Lyman-$\alpha$ still lies in the UV wavelengths and the incidence of DLAs is low. As shown by \citet{Rao06}, strong \MgII\ absorbers are unbiased tracers of DLAs, and can be used to obtain higher detection rates of DLAs in the UV. Their HST-UV surveys of QSOs with strong \MgII\ absorbers led to the identification of 41 DLAs and 81 sub-DLAs at redshifts $0.1<z<1.65$. Subsequently, the absorbing galaxies associated with a sample of 80 of them with redshifts $z<1$ were identified using variety of techniques \citep{Rao11}. More recently, \citet{Turnshek15} used the Advanced Camera for Surveys (ACS) HRC-P200L prism aboard the Hubble Space Telescope (HST) to obtain UV spectra of QSOs with strong \MgII\ absorbers. They found $\approx35$ high-probability DLAs. The dispersion of this prism is non-linear and extremely low at red wavelengths. Therefore, in addition to the dispersed UV light of the QSO which enables detection of DLAs, the spectra of galaxies in the field show little dispersion because they are predominantly red, thus producing deep images of galaxies in the field. Subtracting QSOs from these HST/ACS images produces residual frames in which faint objects appear at impact parameters as small as $b = 2$ kpc ($\sim0.5''$). These data provide direct information on galaxies including sky positions, impact parameters, sizes and morphologies when possible. In this paper we describe the results of a VLT/X-shooter study aimed at identifying the galaxies responsible for DLAs or sub-DLAs in a subset of these QSO fields at $z \sim0.6$. This paper is organized as following. In Section (2) we describe our sample, the strategy of the observations and X-Shooter data reduction. In Section (3) we demonstrate the identification of the absorbing galaxies in the X-Shooter spectra and provide the flux measurements for individual objects. In Section (4) we present the X-Shooter QSO absorption line analysis of our DLAs and sub DLAs. Results are presented in Section (5) and we conclude in section (6). \section[sample]{observations and data reduction} \subsection[]{QSO subtraction of HST/ACS prism images} To identify the absorbing galaxy candidates from HST/ACS high resolution images we carried out an accurate Point Spread Function (PSF) subtraction of QSOs. This procedure is applied on the reduced and sky background subtracted prism images of the QSOs where the fluxes are normalized with respect to the head of the PSF. Details of HST/ACS observations and data reduction of this dataset were described in \citet{Turnshek15}. The quasar light is removed by subtracting a model of the telescope's PSF. We generate such a model PSF by making a stacked image out of QSOs in our sample. A median combined image is made out of all QSOs after they are co-aligned based on a cross-correlation technique. That is, for a given pixel in the PSF model, its flux is the median of all the fluxes in the input images for that pixel. In subtracting the quasar light, the PSF model is shifted to align with the quasar and scaled in flux so that the residual is flat just outside the quasar core. As a result the QSO core is well removed. The dispersed part of the spectrum, forming the PSF's tail, is not well matched by the median model and shows a larger variation from one quasar to another. The differences in the QSOs’ UV spectra, both continuum and lines, cause residuals in the tail that could obscure a compact galaxy lying on the dispersion axis. Fig. \ref{fig_FC_1} presents the results of PSF subtraction on HST/ACS observations. North and East are respectively towards up and left for all panels. Name of QSO fields are provided at the top left corner of each panel. Absorbing galaxy candidates falling in the slit and the QSO are marked with respectively circles and dashed squares. Slit width and length is 1$''$ and 11$''$, respectively and slit position angles (PA) are drawn to match our X-Shooter observations. Detected galaxies in our X-Shooter observations are marked with arrows where longer red ones correspond to galaxies associated with absorbers and shorter black ones are other galaxies. (see section \ref{dla_emission}). In summary we have targeted 9 QSO fields having DLAs at $z\sim0.6$ with X-Shooter to cover 25 absorbing galaxy candidates. The position angle of one of the slits in the field of 0958+0549 has been mis-aligned (see Fig. \ref{fig_FC_1}) and hence we have missed G3. As a result number of targeted galaxies is 24 that means 2.7 galaxy candidates per field. \subsection[]{X-Shooter observation and data reduction}\label{xsh_obs} \begin{figure*} \centering \hspace*{-1cm} \vbox{ \hbox{ \includegraphics[width=0.55\hsize,bb=0 0 541 379,clip=,angle=0]{tim_images_cut/0218_cut.ps} \includegraphics[width=0.55\hsize,bb=0 0 541 379,clip=,angle=0]{tim_images_cut/0957_cut.ps} } \hbox{ \includegraphics[width=0.55\hsize,bb=0 0 541 379,clip=,angle=0]{tim_images_cut/0958_cut.ps} \includegraphics[width=0.55\hsize,bb=0 0 541 379,clip=,angle=0]{tim_images_cut/1012_cut.ps} } \hbox{ \includegraphics[width=0.55\hsize,bb=0 0 541 379,clip=,angle=0]{tim_images_cut/1138_cut.ps} \includegraphics[width=0.55\hsize,bb=0 0 541 379,clip=,angle=0]{tim_images_cut/1204_cut.ps} } } \caption{Slit configurations for 9 QSO fields observed with X-Shooter. Panels are HST/ACS slitless prism images where QSOs PSF are subtracted. Different X-Shooter slit orientations used for observations are entitled ``Slit1`` and ``Slit2``. Circles and dashed square mark the sky position of candidate galaxies and the QSO of each field, respectively. North and East are towards up and left. The solid bar below the direction arrow in each plot is a 1$''$ scale. Longer (vertical or horizontal, in red) and shorter (inclined, in black) arrows in each panel mark detected galaxies associated with and unrelated to DLAs, respectively.} \vskip -.25cm \begin{picture}(0,0)(0,0) \put( -260,630){\bf \large 0218$-$0832} \put( 25,630){\bf \large 0957$-$0807} \put( -260,435){\bf \large 0958$+$0549} \put( 25,435){\bf \large 1012$+$0739} \put( -260,240){\bf \large 1138$+$0139} \put( 25,240){\bf \large 1204$+$0953} \thicklines \put( -160,512){\textcolor{red}{\vector(0,1){25}}} \put( -120,600){\vector(-1,-1){10}} \put( -120,520){\vector(-1,1){10}} \put( 183,555){\vector(-1,-1){10}} \put( -95,382){\textcolor{red}{\vector(-1,0){25}}} \put( 150,335){\textcolor{red}{\vector(1,0){25}}} \put( 185,305){\textcolor{red}{\vector(1,0){25}}} \put( 200,295){\scriptsize G3} \put( -110,230){\textcolor{red}{\vector(0,-1){25}}} \put( 80,193){\textcolor{red}{\vector(1,0){25}}} \put( 115,130){\vector(1,1){10}} \end{picture} \label{fig_FC_1} \end{figure*} \begin{figure*} \addtocounter{figure}{-1} \centering \hspace*{-1cm} \vbox{ \hbox{ \includegraphics[width=0.55\hsize,bb=0 0 541 379,clip=,angle=0]{tim_images_cut/1217_cut.ps} \includegraphics[width=0.55\hsize,bb=0 0 541 379,clip=,angle=0]{tim_images_cut/1357_cut.ps} } \hbox{ \includegraphics[width=0.55\hsize,bb=0 0 541 379,clip=,angle=0]{tim_images_cut/1515_cut.ps} } } \caption{continued.} \begin{picture}(0,0)(0,0) \put( -260,410){\bf \large 1217$+$0500} \put( 25,410){\bf \large 1357$+$0525} \put( -260,210){\bf \large 1515$+$0410} \thicklines \put( -165,295){\textcolor{red}{\vector(0,1){25}}} \put( -170,125){\textcolor{red}{\vector(0,1){25}}} \end{picture} \label{fig_FC_2} \end{figure*} We observed the DLA-galaxy candidates using VLT/X-Shooter \citep{Vernet11} at the European Southern Observatory (ESO) in service mode [programme 092.A-0690(A), Peroux, PI]. The X-Shooter spectrograph covers a wavelength range of 0.3 $\mu$m to 2.3 $\mu$m at medium resolution in a simultaneous use of three arms in UVB, VIS and NIR. Therefore, not only does it cover the interesting nebular emission lines it also covers the UV absorption lines in the spectrum of the QSO that are dominantly in the UVB arm. Having known the position of the DLA-galaxy candidates with respect to the QSOs we have observed all our targets by setting specified slit PAs. Furthermore, to have a robust sky subtraction, specially in NIR, the nodding mode was used following an ABBA scheme. We have not used the default nod lengths (2.5$''$) of the X-Shooter as it may smear out the signal from our targeted faint galaxies by the negative trace of the bright QSOs while subtracting the sky off images. Instead, knowing the impact parameter of candidate galaxies, we fine tune the nodding length using the ``GenericOffset`` template. Table \ref{log_XSH} presents a summary of the details of our X-Shooter observations. Slit widths of 1.0$''$,0.9$''$ and 0.9$''$ are used for respectively UVB, VIS and NIR arms of X-Shooter throughout of our observations. This choice of slit widths results in formal spectral resolutions of 5100, 8800 and 5600 for the UVB, VIS and NIR respectively. We made use of the X-Shooter common Pipeline Library (CPL) \citep{Goldoni06} release 6.5.1\footnote{http://www.eso.org/sci/facilities/paranal/instruments/xshooter/doc/} for reducing the science raw images and produce the final 2D spectra. All our science data are taken in \textsc{nodding} mode and hence we follow a standard procedure of data reduction as follows. We first compute an initial guess for the wavelength solution and position of the center and edges of the orders. Then we trace the accurate position of the center of each order and follow this step by generating the master flat frame out of five individual lamp flat exposures. Next we find a 2D wavelength solution and modify it by applying a flexure correction to correct for the shifts that can be of the order of the size of a pixel. Finally, having generated the required calibration tables we reduce each pair of science frames to obtain the flat-fielded, wavelength calibrated and sky subtracted 2D spectrum. For the flux calibration of each science frame we have chosen the standard star from the X-Shooter archives which has the minimum airmass difference, observed within a maximum 7 days time gap from that science frame. However, we note that recalibration using the standard stars of the same nights results in fluxes consistent within 5\%. To extract the 1D flux of the QSO or other objects from a final reduced 2D frame we carry on a spectral point spread function (SPSF) subtraction through the following steps: (1) dividing the 2D image in chunks of sizes $\sim$ 200 pixels along the wavelength; (2) estimating the mean spatial profile of each chunk by averaging the flux along the wavelength at each spatial pixel; (3) modeling this profile using a Moffat function to estimate the $\sigma$ and central position of each chunk; (4) fitting these parameters using a low order polynomial to extract their values for each pixel (wavelength bin). Having done so we know the center of the QSO ($y_\lambda$) and its spatial broadening parameter ($\sigma_\lambda$) at each wavelength ($\lambda$). (5) Now, we find the emission profile, P$_\lambda$($y$), at each $\lambda$. We model P$_\lambda$($y$) with a Moffat, to obtain the amplitude, while keeping $y_\lambda$ and $\sigma_\lambda$ fixed based on what obtained at stage (4). We further integrate over the best fitted P$_\lambda$($y$) to calculate the total flux at each $\lambda$. Therefore, we simultaneously find the 1D and 2D spectrum of the QSO or other bright sources. We then subtract the 2D modeled spectrum to obtain the residual image. We include masks in cases where we expect emission lines from the galaxy candidate at low impact parameters (b $\lesssim1.5''$). In such cases we mask a region of $\sim100$ \kms\ around the expected wavelengths of nebular emission lines and interpolate the parameters at these ranges using both sides to find the QSO flux. QSO's spectra obtained from the SPSF subtraction are used to study the UVB absorption lines and the residual images are searched for the expected DLA galaxies. % \begin{table*} \small \caption{Log of the VLT/X-Shooter observations.} \begin{tabular}{lcccccccccccc} \hline QSO/field (I)& $z_{\rm em}$ (II) & $z_{\rm abs}$ (III)& cumulative nodding (IV) & Position angle (V) & N$_{\rm gal}$ (VI) & N$_{\rm suit}\times 2\times$ EXPTIME (VII) \\ \hline 0218$-$0832(1) & 1.218 & 0.5899 & +3.0,$-$3.0 & 4 & 2 & 2$\times$2$\times$1200 \\ 0218$-$0832(2) & 1.218 & 0.5899 & +3.0,$-$6.0 & 82 & 2 & 1$\times$2$\times$1200 \\ 0957$+$0807(1) & 0.870 & 0.6975 & +3.5,$-$7.0 & 160 & 2 & 2$\times$2$\times$1440 \\ 0957$+$0807(2) & 0.870 & 0.6975 & +2.0,$-$3.0 & 175 & 2 & 2$\times$2$\times$1200 \\ 0958$+$0549(1) & 0.730 & 0.6557 & +2.0,$-$3.0 & 151 & 2 & 2$\times$2$\times$1200 \\ 0958$+$0549(2) & 0.730 & 0.6557 & +2.0,$-$5.0 & 19 & 1 & 2$\times$2$\times$1200 \\ 1012$+$0739 & 1.030 & 0.6164 & +3.0,$-$6.0 & 43 & 2 & 2$\times$2$\times$1200 \\ 1138$+$0139 & 1.042 & 0.6130 & $-$2.5,+6.0 & 126 & 3 & 1$\times$2$\times$1200 \\ 1204$+$0953 & 1.276 & 0.6401 & $-$3.0,+6.0 & 15 & 3 & 2$\times$2$\times$1200 \\ 1217$+$0500 & 0.632 & 0.5413 & +3.0,$-$6.0 & 87 & 2 & 1$\times$2$\times$1200 \\ 1357$+$0525(1) & 0.740 & 0.6327 & +3.5,$-$6.0 & 51 & 1 & 3$\times$2$\times$1200 \\ 1357$+$0525(2) & 0.740 & 0.6327 & +3.0,$-$6.0 & 177 & 2 & 2$\times$2$\times$1200 \\ 1515$+$0410 & 1.272 & 0.5592 & +3.5,$-$6.0 & 86 & 2 & 2$\times$2$\times$1200 \\ \hline \end{tabular}% \begin{flushleft} (I) QSO field: (1) and (2) define the same QSO field with different configurations of the slit position; (II) redshift of the QSO; (III) redshift of the DLA; (IV) cumulative nodding length from blind offset in arcsec; (V) position angle of the slit measured from North of East in degrees; (VI) Number of covered galaxy candidates in the X-Shooter slit; (VII) number of suitable exposures multiplied by the exposure time in second of individuals (a factor of 2 included for all as all exposures are in nodding mode). \end{flushleft} \label{log_XSH} \end{table*} \begin{figure*} \centering \vbox{ \includegraphics[width=0.25\hsize,bb=206 -1 404 792,clip=,angle=90]{s_0218_gal_1.ps} \includegraphics[width=0.50\hsize,bb=107 -1 503 792,clip=,angle=90]{1515_gal_1.ps} } \caption{{\it top:} part of the X-Shooter spectrum of the DLA-galaxy towards J0218$-$0832. Black and red lines are rest frame spectra of the DLA-galaxy (binned over 100 \kms), at $z_{\rm G2}=0.5895$, and an early type galaxy with similar absorption lines, respectively. Grey rectangles demonstrate the wavelength regions that are affected by telluric sky absorption lines. We have corrected the spectrum for the telluric absorption features using \textsc{Molecfit} \citep{Smette15,Kausch15}. \CaII\ H\&K, H$\delta$ and H9 absorption features are detected and labeled. {\it middle and bottom:} same as {\it top} but for the DLA-galaxy towards J1515$+$0410 at $z_{\rm G2}=0.5580$. \CaII\ H\&K, H$\delta$ and H$\gamma$ are tentatively and 4000 \AA\ break and NaD are clearly detected.} \begin{picture}(0,0)(0,0) \put( 130,438){\bf \large 0218$-$0832 (G2)} \put( 130,312){\bf \large 1515$+$0410 (G2)} \put( 130,185){\bf \large 1515$+$0410 (G2)} \end{picture} \label{fig_cont_0218_1515} \end{figure*} \begin{figure*} \centering \vbox{ \includegraphics[width=0.25\hsize,bb=206 -1 404 792,clip=,angle=90]{0958_gal.ps} \includegraphics[width=0.25\hsize,bb=206 -1 404 792,clip=,angle=90]{1012_gal_2.ps} \includegraphics[width=0.25\hsize,bb=206 -1 404 792,clip=,angle=90]{1138_gal.ps} \includegraphics[width=0.25\hsize,bb=206 -1 404 792,clip=,angle=90]{1204_gal.ps} \includegraphics[width=0.25\hsize,bb=206 -1 404 792,clip=,angle=90]{1217_gal.ps} } \caption{The emission lines from the DLA-galaxies detected by X-Shooter (black histogram) and the best fit Gaussian models (red continuous lines). Standard deviation of the flux ($\pm1\sigma$) for non-detected emission lines are presented with long-dashed lines. The vertical lines in panels with non-detected emission mark the expected position of the expected line. The QSO fields and DLA-galaxies are marked in top-left of each panel.} \begin{picture}(400,400)(0,0) \put( 160,445){\bf Observed wavelength (\AA)} \end{picture} \begin{picture}(0,0)(0,0) \put( -420,1070){\bf \large 0958$+$0549 (G2)} \put( -420,944){\bf \large 1012$+$0739 (G2)} \put( -420,817){\bf \large 1138$+$0139 (G2)} \put( -420,690){\bf \large 1204$+$0953 (G1)} \put( -420,563){\bf \large 1217$+$0500 (G1)} \end{picture} \label{fig_emit_all} \end{figure*} We have detected wavelength offsets of $\sim$ 10 -- 20 \kms\ in the VIS arm of the X-Shooter for some of our exposures. Such systematic errors in the wavelength calibration of the X-Shooter have been already reported by number of authors \citep[e.g.][]{Noterdaeme12}. To quantify and correct such shifts we use the sky absorption lines obtained from Ultraviolet and Visual Echelle Spectrograph on VLT (VLT/UVES) as a reference and compare it with our X-Shooter sky lines. To do so we use a cross-correlation technique described in \citet{Rahmani13}. We cross-correlate the sky absorption lines in the VIS arm of the X-Shooter with that of VLT/UVES to measure and correct the offset for each spectrum. The shortcomings of the X-Shooter flux calibration has been reported in the literature \citep[][]{Fynbo10,Schonebeck14,Japelj15}. Some of the potential problems include different slit widths and sky condition between the object and the standard star, slitloss and atmospheric dispersion. For example \citet[][]{Fynbo10} found that their QSOs fluxes were on average 30\% higher than that of SDSS fiber spectra while the spectral shapes were the same. Hence, they rescaled their fluxes to that of SDSS. However, \citet{Peroux13} \citep[see also][]{Fynbo11} found their X-Shooter fluxes to be consistent with that of SDSS. As QSOs are known to show flux variations over different time scales, such a comparison should be carried out very carefully. Our observing strategy is to cover the HST/ACS DLA-galaxy candidates within the X-Shooter slit. As we have multiple candidates in the field of each QSO we have always applied a blind offset from sky position of QSOs. As a result 8 out of 9 QSOs are partially covered by the X-Shooter slit and in one case (J1217$+$0500) we have not covered the QSO. Hence, while nodding we cover only part of QSOs fluxes. Therefore, rescaling the fluxes based on a comparison between X-Shooter and SDSS fluxes might be inappropriate. The possible systematic errors in our measured fluxes which are dominated by the slitloss, depend on the observing conditions mainly the airmass and the parallactic angle. Hence, a comparison of measured fluxes of the same object between different OBs with different airmass and parallactic angles provides hints towards understanding the possible error budget in the flux measurements. We observe that the fluxes of the emission lines do not change more than 20\% between OBs. Therefore, we expect the error budget introduced by the slitloss will not exceed 20\% of our estimated fluxes. In practice to obtain the flux errors for the emission lines we find that fluctuations of the flux adjacent to the emission lines dominates the Poisson noise. This is mainly due to the residuals from subtracting the QSO spectrum. Thus, in modeling the emission lines we make use of the error estimated from the fluctuations of the flux adjacent to the emission lines. However, to understand the effect of using Poisson noise we re-fitted the emission profiles of some objects using Poisson noise. We find the total fluxes are consistent with our reported values while the estimated errors based on Poisson noise are smaller. Therefore, statistical flux errors based on the fluctuations of the background flux are conservative. \section[]{DLA-galaxy candidates in X-Shooter}\label{dla_emission} In this section we present the results of our search for DLA-galaxies using X-Shooter spectra. For galaxies detected in emission we recover the emission profiles and measure the total fluxes. To do so, we first integrate the flux over the spatial axis to obtain the emission profile. Then we model the resulting profile with a Gaussian function and integrate it to calculate the total flux. We estimate the 1$\sigma$ error of the total flux using the fluctuation of the background level close to the emission line. In cases of non-detection we estimate the 3$\sigma$ upper limits using the fluctuation of the flux at the expected wavelength position of the emission line. We estimate the SFR of galaxies based on their H$\alpha$ flux and using a \citet{Kennicutt98} relation which is corrected to a \citet{Chabrier03} initial mass function. Details of the detected galaxies and non-detections are summarized in Table \ref{xsh_flux}.% \subsection{0218$-$0832}\label{dlagal_0218} There are four DLA-galaxy candidates in this QSO field. We have chosen two different observing configurations ``1`` and ``2`` to cover all four targets. In configuration ``1`` we detect the nebular emission lines from two galaxies at $z$ = 2.441 and 1.613 (indicated by black arrows in panel 0218$-0832$ of Fig. \ref{fig_FC_1}). These two galaxies are at higher redshifts than the QSO at $z_{\rm em}$ = 1.218 and are not seen in absorption in the quasar spectrum. Interestingly, both of these emitters are at the expected sky positions of G1 and G3 obtained from HST/ACS PSF-subtracted image. The fluxes of different nebular lines from these two high-$z$ galaxies are measured and quoted in Table \ref{xsh_flux}. For G3 where the H$\alpha$ line is also detected we further measure a SFR=4.8$\pm$0.4 M$_\odot$ yr$^{-1}$. The observed wavelength for Ly$\alpha$ line at the redshift of G1, $z = $ 2.441, is at $\lambda_{\rm obs}\sim4183$ \AA\ which falls in the UVB arm of X-Shooter. To measure the possible Ly$\alpha$ flux of this galaxy we masked a $\pm$150 \kms\ region centered on 4183 \AA\ and subtracted the QSO spectrum by applying a SPSF. No Ly$\alpha$ emission was detected for G2 and we put a stringent 3$\sigma$ upper limit of 2.4$\times 10^{-17}$ erg s$^{-1}$ cm$^{-2}$ on the Ly$\alpha$ flux. In configuration ``2`` we find a continuum emitting object with no emission line at $\sim2''$ from the QSO in our X-Shooter 2D image. This object is at the expected position of G2. A careful inspection of G2 in HST/ACS image indicates two very closely spaced galaxies, probably merging, surrounded in an extended lower surface brightness halo. We do not find any signature from G4 in configuration ``2`` of our X-Shooter spectrum. The 1D spectrum of G2 was extracted from the final reduced 2D frame of X-Shooter following the SPSF method described in section \ref{xsh_obs}. Top panel of Fig. \ref{fig_cont_0218_1515} presents a portion of the spectrum of G2. To increase the SNR we have binned the spectrum in wavelength bins of 100 \kms. By cross-correlating the spectrum of this DLA-galaxy with the spectrum of a typical early type galaxy (overplotted in red) we obtained $z_{\rm G2}=0.5895$. In Fig. \ref{fig_cont_0218_1515} the wavelength scale is converted to the rest frame corresponding to $z_{\rm G2}$. \CaII\ H\&K absorption features and two Hydrogen Balmer absorption lines are marked. However, the SNR of this spectrum is not good enough for modeling the continuum and absorption lines. Integrating the X-Shooter 1D spectrum of G2 having convolved it with SDSS i-band filter we find m$^i_{\rm ab}$ = 21.7$\pm$0.2 where the error was estimated based on 30\% variation of the flux level. Similarly by shifting the spectrum to its rest frame, based on $z_{\rm G2}$, we find the rest frame r-band and i-band absolute magnitudes of $-20.1\pm0.2$ and $-21.4\pm0.2$ which indicate the host DLA is probably a sub-L$^\star$ galaxy. \subsection{0957$-$0807} There are four DLA-galaxy candidates in the field of this QSO at impact parameters $<2.7''$. We have observed this field with two slit configurations to suitably cover all the DLA-galaxy candidates. We note that G1 and G2 are low surface brightness galaxies that are not visible in Fig. \ref{fig_FC_1} due to the chosen contrast. The brightest candidate in this field is G3 that looks like a bulgeless late-type disk galaxy. In the case of configuration ``1``, where G1 and G2 are covered, neither expected DLA-galaxies nor any other emitters are detected in our X-Shooter spectra. One of the exposures in this field has got a PA = 169$^{\circ}$ which differs from what was requested. We do not detect any emitter in this exposure as well. In case of configuration ``2`` we do not detect emission lines from the expected DLA-galaxies but we do find multiple emission lines at $z = 0.9289$ at the expected position of G3 which is at a higher redshift than QSO J0957$-$0807 at $z_{\rm em}$ = 0.870. To be complete we have quoted total fluxes of the detected emission lines associated with this galaxy in Table \ref{xsh_flux}. \subsection{0958$+$0549} As shown in Fig. \ref{fig_FC_1} three DLA-galaxy candidates at $b \lesssim3.0''$ are targeted in this QSO field using two slit configurations, though further objects are visible at larger impact parameters. Unfortunately, G3 fell out of slit in configuration ``2`` due to an oversight in PA calculation. Therefore, only G1 and G2 are expected to fall into slit in configuration ``1``. As can be inferred from Fig. \ref{fig_FC_1}, G2 presents an extended nature having brighter knots in both up and down sides. Multiple emission lines are detected in the X-Shooter 2D spectrum at the position of G2 having $z_{\rm em}$ = 0.655 that matches with that of DLA absorption lines. However, we do not detect emission lines at the expected position of G1. We also inspected X-Shooter configuration ``2`` 2D spectrum but found no emitter. G2 has the highest H$\alpha$ flux amongst the detected DLA-galaxies in our sample (see Table \ref{xsh_flux}). Moreover, from the 2D spectrum we notice the extended nature of emission lines from G2. Such an extended feature is more pronounced in [OIII]5008 emission line, as presented in Fig. \ref{fig_0958_2d}, where the spectrum reaches its highest SNR. The [OIII]5008 appears in two cores at a projected separation of $1.4''$ ($\sim$ 10 kpc) that are connected by a faint diffuse emission. The 1D spectra of emission lines also show a strong dominant component along with a weaker secondary one (see Fig. \ref{fig_emit_all}). We model the [\OII], H$\beta$ and [\OIII] emission lines using 2-component Gaussian profiles where the $z$ and $\sigma$ of each component is same for all the emission lines. Based on such a fit we find a velocity separation of $\sim$ 100 \kms\ between the two cores where the fainter one contributes $\sim$ 20\% to the total flux. In Table \ref{xsh_flux} we have quoted the total flux for [OIII]5008 and other emission lines without decomposing the two emitters. Estimates of the SFR for each core is not possible as all the H$\alpha$ flux is concentrated in a single core. \begin{figure} \centering \vspace*{-2cm} \includegraphics[width=1.1\hsize,bb=30 97 580 693,clip=,angle=0]{oiii5007_0958.ps} \vspace*{-2cm} \caption{The [OIII]5008 emission of G2 towards 0958$+$0549. The wavelength range is from 8279.6 to 8292.6 \AA\ and the spatial extent is $\sim4.3''$ ($\sim30$ kpc).} \label{fig_0958_2d} \end{figure} \begin{landscape} \begin{table} \centering \caption{Flux measurements of detected galaxies in our X-Shooter data (in unit of 10$^{-17}$ erg s$^{-1}$ cm$^{-2}$). \textbf{Highlighted} DLA-galaxies are those detected in emission at the expected redshift of DLAs and {\it italics} are those detected only in continuum. The 3$\sigma$ flux upper limits are also given in case of non-detections.} \begin{tabular}{lccccccccccccccccccccc} \hline QSO/config & $z_{\rm abs}$& DLA-galaxy& $z_{\rm gal}^{a}$ & [OII] & [OII] & H$\beta$ & [OIII]& [OIII] & [NII] & [NII] & H$\alpha$ & E(B$-$V)& SFR$^{\star}$ & SFR$_{\rm cor}^{\star\star}$\\ &&(impact parameter$^\dagger$)&&3727&3729&&4960&5008&6549&6585&&&(M$_\odot$ yr$^{-1}$)&(M$_\odot$ yr$^{-1}$)\\ \hline \multirow{ 2}{*}{0218$-$0832$^1$} & & G1 ($0.7''$:5.8 kpc) & 2.441 & $<$6.3 & $<$5.1 & $<$4.5 & 1.8$\pm$0.3 & 5.3$\pm$0.6 & --- & --- & --- &---& --- & ---\\ &0.5899$$ & G3 ($3.8''$:32.4 kpc) & 1.613 & $<$1.8 & $<$1.8 & $<$9.6 & 3.7$\pm$0.5 & 10.2$\pm$0.7 & $<$1.5 & $<$2.7 & 6.2$\pm$0.5 &--- & 4.8$\pm$0.4 & 4.8$\pm$0.4 \\ \\ \multirow{ 2}{*}{0218$-$0832$^2$} & & \textit{\textbf{G2 ( 2.2$''$:14.8 kpc)}} & 0.5895$^{b}$ & \boldmath$<$\textbf{\textit{2.0}} & \boldmath$<$\textbf{\textit{2.1}} & \boldmath$<$\textbf{\textit{1.3}} & \boldmath$<$\textbf{\textit{2.1}} & \boldmath$<$\textbf{\textit{1.7}} & \boldmath$<$\textbf{\textit{6.4} } & \boldmath$<$\textbf{\textit{3.9}} & \boldmath$<$\textbf{\textit{2.8}} & ---&\boldmath$<$\textbf{\textit{0.18}} & ---\\ &0.5899$$ & G4 ($3.3''$:22.1 kpc) & --- & $<$1.8 & $<$2.4 & $<$1.2 & $<$1.8 & $<$1.5 & $<$3.3 & $<$2.4 & $<$1.8 & --- &$<$0.12 & --- \\ \\ \multirow{ 2}{*}{0957$+$0807$^1$} & & G1 (1.1$''$:8.3 kpc) & --- & $<$2.1 & $<$2.1 & $<$1.5 & $<$1.8 & $<$2.1 & $<$2.4 & $<$3.9 & $<$2.7& --- & $<$0.26 & ---\\ &0.6975 & G2 ($1.0''$:7.6 kpc) & --- & $<$2.7 & $<$2.1 & $<$1.5 & $<$2.7 & $<$1.5 & $<$3.3 & $<$3.3 & $<$4.5 & --- & $<$0.43 & --- \\ \\ \multirow{ 2}{*}{0957$+$0807$^2$} & & G3 (2.3$''$:16.9 kpc) & 0.9289 & 6.0$\pm$2.3$^c$ & 6.0$\pm$2.3$^c$ & $<$8.3 & $<$7.0 & $<$2.4 & $<$3.0 & 6.1$\pm$4.0 & 11.8$\pm$2.4 & --- & 2.3$\pm$0.5 & 2.3$\pm$0.5 \\ &0.6975 & G4 (2.7$''$:21.3 kpc) & --- & $<$2.7 & $<$4.8 & $<$1.5 & $<$3.6 & $<$1.8 & $<$3.6 & $<$3.0 & $<$3.3& ---& $<$0.32 & --- \\ \\ \multirow{ 2}{*}{0958$+$0549$^1$} & & G1 (1.1$''$:7.7 kpc) & --- & $<$3.3& $<$3.3 & $<$1.8 & $<$1.5 & $<$3.3 & $<$10.0 & $<$8.6 & $<$8.1 & ---& $<$0.67 & ---\\ &0.6557 & {\bf G2 (2.9$''$:20.4 kpc) } & \textbf{0.6547} & \boldmath$2.7\pm0.8$ & \boldmath$5.5\pm0.8$ & \boldmath$5.3\pm0.3$ & \boldmath$4.3\pm0.4$ & \boldmath$14.5\pm0.4$ & \boldmath$<9.3$ & \boldmath$<5.4$ & \boldmath$13.0\pm0.8$ & \boldmath$-0.15\pm0.08$& \boldmath$1.07\pm0.07$ & \boldmath$1.07\pm0.07$\\ \\ \\ \multirow{ 3}{*}{1012$+$0739} & & G1 (0.9$''$: 6.3 kpc) & --- & $<$1.7 & --- & $<$2.4& --- & $<$1.2 & $<$2.3 & $<$2.5 & $<$1.7 & ---& $<$0.12 & --- \\ &0.6164 & {\bf G2 (4.3$''$:29.8 kpc) } & \textbf{0.6155} & \boldmath$<1.8$ &\boldmath$<1.7$ & \boldmath$0.66\pm0.33$& \boldmath$<1.7$ & \boldmath$<1.2$ & \boldmath$<3.7$ & \boldmath$2.3\pm1.7$ & \boldmath$7.2\pm1.6$& \boldmath$1.28\pm0.53$& \boldmath$0.5\pm0.1$ & \boldmath$6.7\pm1.3$ \\ & & {\bf G3}(7.2$''$:50.0 kpc) & 0.6162 & 2.6$\pm$0.8& 3.9$\pm$0.8 & 1.9$\pm$0.4 & 1.3$\pm$0.2 &5.1$\pm$0.3 & $<$3.0 & $<$1.7 & 3.9$\pm$0.5& $-0.33\pm0.24$& 0.28$\pm$0.04 & 0.28$\pm$0.04\\ \\ \multirow{ 3}{*}{1138$+$0139} & & { G1} (1.0$''$: 7.0 kpc) & --- & $<$1.5 & $<$1.2 & $<$1.2 & $<$1.5 & $<$1.2 & $<$2.4 & $<$2.4 & $<$2.7& --- &$<$0.19 & --- \\ &0.6130 & {\bf G2 (1.8$''$:12.4 kpc)} & \textbf{0.6122} & \boldmath$4.5\pm1.1$ & \boldmath$6.4\pm1.1$& \boldmath$2.0\pm0.4$ & \boldmath$1.2\pm0.5$ & \boldmath$5.0\pm0.5$ & \boldmath$<2.7$ & \boldmath$<3.6$ & \boldmath$9.3\pm1.1$ & \boldmath$0.46\pm0.22$ & \boldmath$0.65\pm0.08$ & \boldmath$1.7\pm0.2$ \\ & & G3 (2.2$''$:15.2 kpc) & --- & $<$1.8 & $<$1.8 & $<$1.2 & $<$1.5 & $<$1.2 & $<$2.4 & $<$2.4 & $<$1.8 & --- &$<$0.13 & ---\\ \\ \multirow{ 3}{*}{1204$+$0953} & & {\bf G1 (1.5$''$:10.0 kpc)} & \textbf{0.6395} & \boldmath$1.4\pm1.0$ & \boldmath$1.8\pm1.0$ & \boldmath$2.2\pm1.0$ & \boldmath$<1.9$ & \boldmath$<2.1$ & \boldmath$<10.3$ & \boldmath$2.1\pm0.8$& \boldmath$8.9\pm3.5$ & \boldmath$0.33\pm0.58$& \boldmath$0.69\pm0.27$ &\boldmath$1.3\pm0.5$ \\ &0.6401 & G2 (2.7$''$:23.0 kpc) & 1.2776 & 1.9$\pm$0.2 & 3.1$\pm$0.2 & 2.1$\pm$1.0 & $<$6.3 & 11.9$\pm$1.5 & $<$0.9 & $<$0.9 & 8.5$\pm$0.3 & 0.33$\pm$0.46 & 3.7$\pm$0.1 & 7.2$\pm$0.3\\ & & G3 (3.8$''$:27.2 kpc) & --- & $<$2.4 & $<$2.4 & $<$1.5 & $<$1.2 & $<$1.5 & $<$3.3 & $<$2.4 & $<$3.6 & ---& $<$0.28 & --- \\ \\ \multirow{ 2}{*}{1217$+$0500} & & {\bf G1 (2.3$''$:14.7 kpc)} & \textbf{0.5405} & \boldmath$3.8\pm1.4$ & \boldmath$6.5\pm1.4$ & \boldmath$3.7\pm0.6$ & \boldmath$1.3\pm0.4$ & \boldmath$6.9\pm0.5$ & \boldmath$<7.3$ & \boldmath$<9.3$ & \boldmath$8.6\pm3.0$ & \boldmath$-0.21\pm0.37$& \boldmath$0.44\pm0.15$ & \boldmath$0.44\pm0.15$ \\ &0.5413 & G2 (2.4$''$:15.1 kpc) & --- & $<$7.5 & $<$7.2 & $<$2.1 & $<$2.4 & $<$3.0 & $<$7.8 & $<$9.9 & $<$7.2 & --- & $<$0.37 & --- \\ \\ 1357$+$0525$^1$ &0.6327 & G1 (1.8$''$:12.7 kpc) & --- & $<$3.3 & $<$3.0 & $<$2.4 & $<$2.1 & $<$1.2 & $<$3.0 & $<$4.5 & $<$5.7 & --- & $<$0.43 & --- \\ \\ 1357$+$0525$^2$ &0.6327 & G2 (2.6$''$:17.7 kpc) & --- & $<$3.3 & $<$3.0 & $<$1.5 & $<$1.8 & $<$1.8 & $<$3.6 & $<$7.2 & $<$5.7 & --- & $<$0.43 & --- \\ \\ \multirow{ 2}{*}{1515$+$0410} & & G1 (0.9$''$:5.9 kpc) & --- & $<$1.8 & $<$1.8 & $<$1.2 & $<$0.9 & $<$1.2 & $<$6.9 & $<$3.6 & $<$3.6 & ---& $<$0.20& ---\\ &0.5592 & \textbf{\textit{G2 (1.6$''$:10.2 kpc)}} & 0.5580$^{b}$ & \boldmath$<$\textbf{\textit{2.6}} & \boldmath$<$\textit{\textbf{2.5}} & \boldmath$<$\textit{\textbf{1.6}} & \boldmath$<$\textit{\textbf{1.6}} & \boldmath$<$\textit{\textbf{1.2}} & \boldmath$<$\textit{\textbf{12.3}} & \boldmath$<$\textit{\textbf{5.2}} & \boldmath$<$\textit{\textbf{6.6}} & --- & \boldmath$<$\textit{\textbf{0.37 }} & ---\\ \\ \hline \end{tabular}% \begin{flushleft} $^\dagger$ Impact parameters (in kpc) are calculated based on redshifts of the detected galaxies. For non-detections we assume the galaxy candidates are at the redshift of the DLAs. \\ $^{\star}$ SFR estimated from H$\alpha$ flux assuming a \citet{Kennicutt98} conversion corrected to a \citet{Chabrier03} initial mass function.\\ $^{\star\star}$ SFR corrected for dust extinction in the host galaxy.\\ $^{a}$ average redshift of those emission lines in the VIS arm\\ $^{b}$ absorption redshifts of the DLA-galaxies obtained from cross-correlation analyses (see Section \ref{dlagal_0218} and \ref{dlagal_1515})\\ $^c$ Separate fluxes of the two [OII] lines can not be measured. \\ \end{flushleft} \label{xsh_flux} \end{table} \end{landscape} \subsection{1012$+$0739} The HST/ACS image of this QSO presents a crowd with several possible DLA-galaxy candidates. However, a single X-Shooter slit configuration was chosen to cover the two closest objects, called G1 and G2, at impact parameters of respectively $0.9''$ and $4.3''$. Given the selected configuration we incidentally cover a third object in the slit at an impact parameter of $7.2''$ which is dubbed G3. No emission is detected at the expected position of G1 in our X-Shooter spectrum. G2 is detected in H$\alpha$ emission with high significance at a redshift matching $z_{\rm abs}$ along with a faint continuum. The continuum emission from G2 is visible in the entire range from $\lambda_{rest} \sim$ 4300 -- 12600 \AA\ but does not have a high enough SNR to enable the analysis of galaxy absorption lines. The H$\alpha$ emission from G2 with $\sigma=146$ \kms\ is the broadest line associated to DLA-galaxies in our sample. Apart from a tentative H$\beta$ and an [\NII]$\lambda$6585 none of other emission lines are significantly detected from G2. Interestingly, multiple emission lines are detected from G3 at $z=0.6162$ that also matches with the redshift of the sub-DLA. A velocity separation of $\sim140$ \kms\ persists between G2 and G3. The impact parameter of G2 is $\sim$ 20 kpc smaller than that of G3. Hence, we will consider G2 as the galaxy responsible for the sub-DLA in this sightline.% Due to a high clustering possibility of DLA-galaxies in this field we inspected the SDSS database of this QSO sightline. Two objects presented on the left side of HST/ACS image are detected as a faint elongated one having m$_i \sim$ 21 mag with a photometric redshift $z_{\rm phot}$ = 0.58$\pm$0.21 at $\sim7.7''$ distance to the QSO. There further exists a faint, m$_i \sim$ 21 mag, object at $12''$ south-east of this QSO (not covered in HST/ACS image) with $z_{\rm phot}$ = 0.57$\pm$0.10. Photometric redshifts of both of these faint objects are consistent with $z_{\rm abs}$ of the associated sub-DLA. Therefore, a clustering of faint galaxies may exists in this field where further deep spectroscopic observations will be required to confirm or infirm this hypothesis. \begin{figure*} \centering \vspace*{-0.6cm} \includegraphics[width=0.95\hsize,bb=18 17 594 773,clip=,angle=0]{dfit_0218_abs.ps} \vskip 0.5cm \caption{ Metal absorption lines towards J0218$-$0832. The zero velocity is set at the redshift of the DLA-galaxy, $z_{\rm G2}=0.5895$. Vertical dashed lines denote the velocity of different absorbing components. The histogram and continuous lines present the observed normalized QSO flux and the best obtained Voigt profile model, respectively. The error spectrum is shown as dashed line and the residual of the fit ([data]-[model]) as the dashed-dotted line (both shifted with a constant offset of 0.45 for panels not covering zero flux on y-axis). Similar figures for the remaining absorbers in our sample are presented in Appendix \ref{app_abs_profile}. } \begin{picture}(0,0)(0,0) \put( -260,360){\rotatebox{90}{\large Normalized Flux}} \put( -30,59){\large Velocity (\kms)} \end{picture} \label{fig_0218_abs} \end{figure*} \subsection{1138$+$0139} There are three DLA-galaxy candidates towards this QSO sightline in the HST/ACS image at impact parameters $<2.2''$. All three candidates show some UV fluxes, though at different levels. A single X-Shooter configuration covers these 3 candidates within the slit and partially the QSO. However, the nebular emission lines are detected only from the candidate at an impact parameter of $1.8''$ (12 kpc). Inspecting the rest of the X-Shooter 2D spectrum we do not find any other emitter. Therefore, we estimate the emission line fluxes for G2 while for G1 and G3 we calculate 3$\sigma$ upper limits from non-detections. Fig. \ref{fig_emit_all} presents detected emission profiles of the nebular lines. The [OIII]4960 line is contaminated by the residuals from a sky emission line since it is close to a sky emission line. Hence, to model the emission profile we fix the $\sigma$ and the $z_{\rm em}$ to that obtained from [OIII]5008. The total measured flux of the [OIII]4960 is significant at a level lower than 3$\sigma$ (see Table \ref{xsh_flux}). This QSO sightline has another strong \MgII\ absorber at $z = $ 0.7658 \citep{Quider11}. Using QSO spectrum obtained from our X-Shooter observations we estimate the rest frame equivalent widths (EW) of \MgII$\lambda$2796 and \FeII$\lambda$2600 to be respectively 0.93$\pm$0.01 \AA\ and 0.34$\pm$0.01 \AA. \MgII\ systems at such EWs have high chances to be associated with sub-DLAs or DLAs \citep{Rao00,Rao06}. However, none of the expected emission lines at $z = $ 0.7658 are detected in our X-Shooter spectrum. \subsection{1204$+$0953} The PSF subtracted HST/ACS image of this field presents several objects at impact parameters of $<6''$. We have chosen a single slit configuration for this field in which we cover three of such galaxy candidates and partially the QSO. Multiple nebular emission lines are detected in X-Shooter 2D spectrum at impact parameters of $1.5''$ and $2.7''$. The redshift of the former, $z$ = 0.6395, matches that of the DLA but the latter, $z$ = 1.2776, is $\sim$ 450 \kms\ higher than that of the emission redshift of the QSO. Furthermore, no absorption signature is detected to be associated with this high redshift galaxy. Hence, most likely this galaxy is in the background with respect to the QSO. As no emission line is detected at impact parameter of $3.8''$ as expected from G3 we only place upper limits for fluxes of nebular emission lines assuming the same redshift as that of the DLA. We also notice the detection of [NII]6585 emission line from the DLA-galaxy in this sightline at more than 2$\sigma$ significance. Two further \MgII\ absorption systems are detected towards this QSO at $z \sim$ 0.830 and 0.837 with rest-frame EW$^{2796}$ of respectively 0.14$\pm$0.01 \AA\ and 0.46$\pm$0.03 \AA. The expected N(\HI) for such \MgII\ absorbers are in the range of LLS but lower than that of sub-DLAs \citep{Rao06}. We inspected the 2D spectrum searching for nebular emission lines at such redshifts. A weak emission line was detected at an observed wavelength of $\lambda \sim$ 9146.7 \AA\ and an impact parameter of $7.8''$. In case of assuming it is [OIII]5008 such a line is at $z = 0.8263$ or $\sim$ 635 \kms\ with respect to the weaker \MgII\ absorber with the impact parameter of 60 kpc. However, further deeper observations are required to conform this hypothesis. \subsection{1217$+$0500} Two DLA-galaxy candidates are present in the HST/ACS image of this field at impact parameters of $2.3''$ and $2.4''$. We are able to cover both of them in the slit using a single X-Shooter slit configuration at the cost of not covering the QSO. Hence, in the case of this sightline we neither have details of the metal absorption lines nor measure the absorption metallicity. We detect nebular emission lines from one of the candidates at an impact parameter of $2.3''$ at the redshift of the DLA. Apart from the detected emission lines tabulated in Table \ref{xsh_flux} we also detect with a high significance [SIII]9533 from this DLA-galaxy. No emission line is detected for G2 at the DLA's redshift or other redshifts. Hence, upper limits were estimated for the fluxes of this galaxy assuming it lies at the redshift of this sub-DLA. % \subsection{1357$+$0525}\label{1357_em} Our original analysis of this prism field suggested that there were two DLA-galaxy candidates towards this QSO at impact parameters of $1.8''$ and $2.6''$. G1 looked like a disk galaxy with a bright bulge in its center. Two different X-Shooter configurations were used to observe G1 and G2 in two different slits (Table \ref{log_XSH}). However, no significant continuum or emission lines were detected from any objects at these locations. Hence, we have used the observations at the positions of G1 and G2 to estimate upper limits from non-detections of nebular emission lines.% Our inability to detect apparently bright optical flux at the position of G1 is inconsistent with our original interpretation of this field, namely that, in addition to providing a dispersed image of a blue QSO, the prism image also provides a deep image of mildly-dispersed optical light from galaxies in the field (e.g., Section 2.1). Instead, as the X-Shooter observation showed, there is no optical light at the position of G1. The above two observational results indicate that the light seen in the prism image is UV light. Only if we assume that the object is a Lyman-$\alpha$ emitter (LAE) do the two above observations make sense; it explains why an object was not seen with X-Shooter, despite the fact that it appeared to be the brightest galaxy candidate in our sample. This leads to the interpretation that the significant flux seen in the prism image is actually due to a Ly$\alpha$ emitter (LAE) at the DLA redshift ($z=0.6327$). If so, the LAE has a total extended Ly$\alpha$ emission flux of $\sim 7.6 \times 10^{-14}$ ergs cm$^{-2}$ s$^{-1}$, which translates to a total Ly$\alpha$ luminosity of $\sim 1.3 \times 10^{44}$ ergs s$^{-1}$. It has an impact parameter of $\sim 0.81''$ ($\sim 5.5$ kpc) relative to the QSO sightline and an oval-like extent of $\sim 22$ kpc by $\sim 6$ kpc. It has a bright central region $\sim 1$ kpc across which contains $\sim 24$\% of the total measured flux. These characteristics are similar to the many high-redshift LAE found in the literature \citep[e.g.][and references therein]{Bielby16}, but this example is rather remarkable since it has a low redshift and is identified with a DLA absorber. However, as we do not have a spectrum of this object we do not count it as a confirmed detection. \subsection{1515$+$0410}\label{dlagal_1515} There are two DLA-galaxy candidates at impact parameters of $0.9''$ and $1.6''$ in the field of this QSO. A high level of UV flux was detected at the core of G2 in the PSF subtracted HST/ACS image. In a single X-Shooter configuration we are able to cover both galaxy candidates and partially the QSO in the slit. No emitting object were detected at the expected position of G1. However, at the expected position of G2, a continuum flux (with no emission line) is detected over an observed wavelength range from 5500 \AA\ to 21000 \AA. The extracted 1D spectrum of G2 represents several absorption features (see middle and bottom panels of Fig. \ref{fig_cont_0218_1515}). We cross-correlate the spectrum of G2 with a typical early type galaxy to obtain $z_{\rm G2}=0.5580$. The absorption features and the break match those of a typical early type galaxy redshifted at the position of the DLA. The \CaII\ H\&K, H$\delta$ and \MgI$\lambda$5170 in particular are detected and labeled in Fig. \ref{fig_cont_0218_1515}. However, the SNR of this spectrum is not enough for modeling the continuum and absorption lines. Therefore, we confirm G2 to be the galaxy responsible for the sub-DLA absorption in this QSO sightline at an impact parameter of 10 kpc. In the absence of H$\alpha$ we put a stringent 3$\sigma$ upper limit of 0.4 M$_\odot$ yr$^{-1}$ on the SFR of this quenched galaxy. Integrating the X-Shooter spectrum of this galaxy when convolved with the SDSS i-band filter we find a m$^i_{\rm AB}$ = 21.4$\pm$0.2 mag. We further estimate the SDSS ``r`` and ``i`` band absolute magnitudes to be respectively $-$19.2$\pm$0.2 and $-$19.5$\pm$0.2 mag which given the uncertainties are similar to those estimated for G2 towards 0218$-$0832. This finding is in line with their rest frame flux levels (see Fig. \ref{fig_cont_0218_1515}). | We have studied X-Shooter observations of a set of 9 QSO fields. These QSOs are selected based on HST/ACS observations that indicate they have a DLA or sub-DLA in their spectra at $<z>\sim0.6$. A PSF subtraction of high spatial resolution prism images obtained from these HST/ACS observations provide us with 24 candidate DLA-galaxies in 9 fields at low impact parameters ($b\gtrsim0.7''$ or 6.0 kpc at $z\sim0.6$). The main goal of our X-Shooter observations is to confirm such candidates and study galaxies responsible for DLA and sub-DLA systems. The wide wavelength range covered by the X-Shooter (0.3 -- 2.3 $\mu$m) allows us to search for several nebular emission lines from DLA-galaxy candidates while simultaneously obtaining QSOs' spectra to study the associated metal absorption lines. We detect a total number of 12 galaxies in our X-Shooter observations. Four out of 12 detections are high redshift line emitters unrelated to \HI\ absorbers. The remaining 8 detected galaxies are related to DLAs in 7 QSO fields at impact parameters from 10 to 50 kpc. Having detected galaxies responsible for very strong \HI\ absorption in 7 out of 9 QSO fields we reach a high detection rate of 78\%. The high rate in detection of DLA-galaxies obtained in this work indicates that a high spatial resolution imaging of QSOs followed by deep spectroscopic observations is an efficient strategy in finding galaxies responsible for DLAs and sub-DLAs. In 2 of the fields (0957+0807 and 1357+0525) we could not detect the absorbing galaxies in our X-Shooter data. We postulate the following two reasons for such non-detections: (I) they are not detected because they lie in the PSF of HST/ACS images. This can happen for absorbing galaxies at (Ia) $b\lesssim0.5''$ or (Ib) in the tail of the PSF where the QSO subtraction leaves larger residuals. (II) They have fainter emission lines or continuum than can be observed with our X-Shooter observations. We ignore the possibility of the absorbing galaxies to be at larger distances than that covered by HST/ACS field of view as the two non-detections have high N(\HI) of 10$^{21.44}$ and 10$^{20.81}$ cm$^{-2}$ and hence are unlikely to lie at $b\gtrsim20$ kpc ($\sim3''$) (see Fig. \ref{fig_b_nhi}). While the high N(\HI) of these two systems favors the reason (Ia) for the non-detections, still the small sample size prevents us to draw a concrete conclusion. We estimate the extinction corrected SFR of the 5 DLA-galaxies detected in emission, to be in the range of 0.4 to 6.7 M$_\odot$yr$^{-1}$ with a median of 1.07 M$_\odot$yr$^{-1}$. We also discover two sub-L$^\star$ (M$_i\sim-19.5$ mag) DLA-galaxies detected only in continuum with 3$\sigma$ SFR upper limits of 0.2 and 0.4 M$_\odot$yr$^{-1}$. Overall, such findings demonstrate that the average population of DLA-galaxies at $z\sim0.6$ are low luminosity dwarf galaxies at maximum impact parameter of 30 kpc. A comparison between our results and those reported in the \citep{Peroux12} but also $z\sim1$ indicate possible signatures of redshift evolution of SFR of DLA-galaxies.% In two of the QSO fields we detect two galaxies matching the absorption redshift. In particular in the case of 0958+0549 the appearing morphology resembles a merging configuration of two galaxies connected via a faint emitting stream. The absorption in such cases might be produced by the disrupted gas streams resulting from strong gravitational interactions of galaxies. Further 3D morpho-kinematic integral field unit (IFU) observations using instrument like Multi Unit Spectroscopic Explorer (MUSE) on VLT of such systems can unravel the physics of these galaxy absorbers. | 16 | 9 | 1609.03843 |
1609 | 1609.00184_arXiv.txt | We describe the development and implementation of a light-weight, fully autonomous 2-axis pointing and stabilization system designed for balloon-borne astronomical payloads. The system is developed using off-the-shelf components such as Arduino Uno controller, HMC 5883L magnetometer, MPU-9150 Inertial Measurement Unit (IMU) and iWave GPS receiver unit. It is a compact and rugged system which can also be used to take images/video in a moving vehicle, or in areal photography. The system performance is evaluated from the ground, as well as in conditions simulated to imitate the actual flight by using a tethered launch. | High-altitude balloon platforms are an economical alternative to space missions for testing instruments as well as for specific classes of observations, particularly those that require a rapid response, such as comets or other transients. We have developed a number of payloads which operate in the near ultraviolet (NUV: 200--400 nm) which we have flown on high-altitude balloons\cite{Safonova16}. We are limited to payloads under 6 kg for regulatory reasons and this constrains our payload size. Our first experiments were of atmospheric lines\cite{Sreejith16} where the pointing stability is less important, but we do plan to observe astronomical sources for which a pointing mechanism is required. Light balloons are an exceptionally challenging platform for accurate pointing because the platform itself is in constant motion, sometimes with violent jerks and rotations. Most pointing systems for scientific balloon experiments to date have been designed for the use on large balloons with payload weights of a tonne, or more. Such systems include SPIDER\cite{Crill08}, BETTII\cite{Rizzo}, BOOMERANG\cite{Crill}, BLAST\cite{Pascale} or BLAST-Pol\cite{Fissel}. The accuracy of pointing of these systems varies from several arcminutes to few arcseconds. For example, the pointing system in SPIDER has an accuracy of $1^{\circ}$ and in BLAST Pol of $30^{\as}$; accuracy increasing with weight and complexity of the system. In a broad sense, pointing systems for high-altitude balloons consist of four parts: 1. attitude sensors (ASs), 2. actuators, 3. attitude control system, and 4. mechanical structure. In this work, we describe the design and realization of a low-cost light-weight 2-axis correction pointing and stabilization system intended for use in small balloon flights, built completely using off-the-shelf components. The primary challenge in this development is that its weight must be under 1 kg, given the total mass constraint of 6 kg. We plan to use this pointing system with other instruments that we are developing. The immediate requirements for accurate pointing come from a light-weight (650 gm) compact star-sensor camera {\it StarSense} with an accuracy of $30^{"}$ and $10^{\circ}$ field of view (FOV) \cite{Mayuresh}, and a wide-field compact ($15 \times 15\times 35$ cm) NUV imager \cite{Ambily}. Both these instruments are developed for use in small balloon payloads, as well as in nanosatellites or CubeSats, and will have a test flight in November 2016 \cite{Sreejith2016}. | We have designed and developed a low-cost light-weight, closed-loop pointing and stabilization platform for use in balloon-borne astronomical payloads. This system was build completely from off-the-shelf components: an MPU-9150 IMU, a HMC5883L magnetometer, an Arduino controller and a SiRF StarIII GSC3f GPS receiver unit. The system performance was checked on the ground and in tethered flights with satisfactory results. The system can point to an accuracy of $\pm 0.28^{\circ}$ and track objects from the ground with an accuracy of $\pm 0.13^{\circ}$. The performance in the tethered flights was poorer ($0.40^{\circ}$ in best conditions), largely because of strong winds at low altitudes. However, the stability of pointing was still within $\sim 1.6^{\circ}$ even in worst conditions. Such winds are not present in the stratosphere\cite{Manchanda}, where payloads are known to be stable at float\cite{ritareport}, and we expect pointing accuracy and stability of our system to be similar to those on the ground. We are exploring several avenues to further improve the system performance including using better sensors and servomotors with finer steps. We have developed a star-senso\cite{Mayuresh}r with a resolution of $30^{\prime\prime}$ which we will patch into the pointing system. We plan to have a high-altitude floating balloon flight in November 2016 with an imager and a spectrograph where this pointing system will be put to use\cite{{Ambily},{Sreejith2016}}. | 16 | 9 | 1609.00184 |
1609 | 1609.02148_arXiv.txt | We report a new ultra-faint stellar system found in Dark Energy Camera data from the first observing run of the Magellanic Satellites Survey (\maglites). \iauname (Pictor II or Pic II) is a low surface brightness ($\mu = 28.5^{+1}_{-1} \magn \asec^{-2}$ within its half-light radius) resolved overdensity of old and metal-poor stars located at a heliocentric distance of $45^{+5}_{-4} \kpc$. The physical size ($r_{1/2} = 46^{+15}_{-11} \pc$) and low luminosity ($M_V = -3.2^{+0.4}_{-0.5} \magn$) of this satellite are consistent with the locus of spectroscopically confirmed ultra-faint galaxies. \picII is located $11.3^{+3.1}_{-0.9} \kpc$ from the Large Magellanic Cloud (LMC), and comparisons with simulation results in the literature suggest that this satellite was likely accreted with the LMC. The close proximity of \picII to the LMC also makes it the most likely ultra-faint galaxy candidate to still be gravitationally bound to the LMC. | \label{sec:intro} The standard cosmological model generically predicts the formation of structure over a wide range of mass scales from galaxy clusters to ultra-faint galaxies. The Local Group offers a unique environment to search for evidence of hierarchical structure formation on the smallest scales. For decades authors have speculated that some of the smaller Milky Way satellites may have originated with the Large and Small Magellanic Clouds \citep[LMC, SMC; \eg,][]{Lynden-Bell:1976,D'Onghia:2008a,Sales:2011a,Nichols:2011a}. The recent discovery of more than twenty ultra-faint ($M_V \gtrsim -8$) galaxy candidates by wide-area optical surveys including the Dark Energy Survey \citep[DES;][]{Bechtol:2015wya,Koposov:2015cua,Kim:2015c,Drlica-Wagner:2015ufc}, the Survey of the MAgellanic Stellar History \citep[SMASH;][]{martin_2015_hydra_ii}, Pan-STARRS \citep{Laevens:2015a,Laevens:2015b}, and VST ATLAS \citep{Torrealba:2016a,Torrealba:2016b} has renewed interest in identifying faint galactic companions of the Magellanic Clouds. Indeed, 15 of the 17 candidates in the DES footprint are located in the southern half of the surveyed area, near to the Magellanic Clouds. This inhomogeneity in the spatial distribution of satellites allows the DES data alone to exclude an isotropic spatial distribution of Milky Way satellites at the $3\sigma$-level \citep{Drlica-Wagner:2015ufc}. Instead, the observed distribution can be well, though not uniquely, described by an association between several of the new satellites and the Magellanic system. Simple models incorporating DES and SDSS observations predict that the entire sky may contain $\roughly 100$ ultra-faint galaxies with physical properties comparable to the DES satellites and that 20--30\% of these could be spatially associated with the Magellanic Clouds \citep{Drlica-Wagner:2015ufc}. These conclusions are largely supported by detailed simulations \citep{Deason:2015,Wheeler:2015,Yozin:2015,Jethwa:2016,Sales:2016}, which also find evidence for a Magellanic bias in the Milky Way satellite distribution. In addition, the systemic radial velocities of several of the newly discovered satellites may be consistent with the orbit of the Clouds \citep{Koposov:2015b,Walker:2016,Jethwa:2016,Sales:2016}. Since the Magellanic Clouds are likely on their first passage around the Milky Way \citep{Besla:2007,Busha:2011,Kallivayalil:2013}, satellite galaxies that originated with the Clouds would have formed in an environment that was rather different from the one they inhabit today. Comparing these systems to systems that formed around the Milky Way or far from any massive host would test environmental influences on the age, star formation history, and chemical evolution of the smallest galaxies. Furthermore, the existence and properties of satellites of satellites can test the hierarchical structure predictions of $\Lambda$CDM. Two low-luminosity satellites have been recently found around more isolated Local Volume analogs of the Magellanic Clouds: Antlia B around NGC 3109 \citep{Sand:2015} and MADCASH J074238+652501-dw around NGC 2403 \citep{Carlin:2016}. Satellite-host associations are more certain in these cases relative to the Magellanic system, due to the absence of a nearby large galaxy like the Milky Way. However, only the Magellanic Clouds are close enough to efficiently detect and characterize ultra-faint satellites. The \textbf{Mag}ellanic Satel\textbf{Lite}s \textbf{S}urvey (\maglites; PI K. Bechtol) is a NOAO community survey that uses the Dark Energy Camera \citep[DECam;][]{Flaugher:2015} to complete an annulus of contiguous imaging around the periphery of the Magellanic system (\figref{maglites}). In \secref{data} we describe the scope and progress of \maglites. Initial inspection of stellar catalogs assembled from the first \maglites observing run (R1) revealed a resolved stellar overdensity at $(\ra,\dec) = (101\fdg180,-59\fdg897)$, as described in \secref{discovery}. The physical properties of this satellite are similar to known ultra-faint galaxies (\figref{maglites}, right panel), and are detailed in \secref{properties}. In \secref{discussion} we conclude by discussing the possible association between this stellar system and the Magellanic Clouds. This satellite resides in the constellation Pictor, and if it is confirmed to be a dark-matter-dominated galaxy, would be named Pictor~II (Pic~II); otherwise it will be named \maglites 1. Until spectroscopic observations clarify the physical nature of this system, we refer to it as \picII. \begin{figure*} \center \includegraphics[width=0.40\textwidth]{{maglites_allms}.pdf} \includegraphics[width=0.49\textwidth]{{size_luminosity_spectroscopic_status_maglites_r1}.pdf} \caption{ \textit{Left}: Orthonormal projection of the southern celestial hemisphere showing the HI density of the Magellanic Stream in gray scale \citep{Nidever:2010}. Over-plotted are the footprints of DES (black), \maglites (green), and SMASH (blue hexagons representing individual DECam pointings). The location of \picII is shown with a gold star. Other candidate and confirmed Milky Way satellite galaxies are marked with triangles. The distant LMC star clusters NGC~1841, Reticulum, and ESO~121-SC03 are marked with black crosses. \textit{Right}: Absolute visual magnitude ($M_V$) versus azimuthally averaged physical half-light radius ($r_{1/2}$) for dwarf galaxies (solid red triangles), globular clusters (black crosses), and recently discovered systems lacking spectroscopic measurements (open red triangles). The black dashed lines indicate contours of constant surface brightness ($\mu$; average within the half-light radius). \picII is marked by a gold star. } \label{fig:maglites} \end{figure*} | \label{sec:discussion} The low luminosity ($M_V = -3.2$) and large physical size ($r_{1/2} = 46 \pc$) of \picII are consistent with the population of dark-matter-dominated Milky Way satellite galaxies (\figref{maglites}). Specifically, \picII possesses structural properties similar to the recently confirmed dwarf galaxies Reticulum~II and Horologium~I \citep{Bechtol:2015wya,Koposov:2015cua}. While stellar kinematic data are necessary to measure the dark matter content of \picII and assign a definitive classification, the \maglites photometry suggests that it will likely join the ranks of recently discovered dwarf galaxies. The proximity between \picII and the LMC, $D_{\rm LMC} = 11.3^{+3.1}_{-0.9} \kpc$, is suggestive of a physical association between these two systems. Several studies have shown that the population of old LMC stars extends to radii $>13\kpc$ \citep{Munoz:2006,Majewski:2009,Saha:2010,Balbinot:2015,Mackey:2016}. Additionally, kinematic measurements by \citet{vanderMarel:2014} suggest that the LMC tidal radius is at least $16 \kpc$ and may be as large as $22 \pm 5 \kpc$, which places \picII well within the LMC sphere of influence. The most distant LMC star clusters reside at similar distances: NGC 1841 at $D_{\rm LMC} = 14.9 \kpc$, Reticulum at $D_{\rm LMC} = 11.4 \kpc$, and ESO 121-SC03 $D_{\rm LMC} = 9.7 \kpc$ \citep{Schommer:1992}. If \picII is bound to the LMC it would be expected to have a line-of-sight velocity that is similar to these clusters: $214\kms$, $243\kms$, and $309\kms$, respectively \citep[][and references therein]{Schommer:1992}. Incidentally, \picII is located at a heliocentric distance that is consistent with the plane of the LMC disk \citep[$\roughly 46 \kpc$;][]{vanderMarel:2014}. Several recent studies have used numerical simulations to investigate the evolution of the Magellanic system as it was accreted onto the Milky Way \citep[\ie,][]{Deason:2015,Jethwa:2016}. Using the ELVIS simulations \citep{Garrison-Kimmel:2014}, \citet{Deason:2015} find that $> 40\%$ of satellites galaxies that are currently located at $D_{\rm LMC} < 20 \kpc$ were bound to the LMC before infall into the Milky Way. This fraction increases to $> 65\%$ if the Magellanic group was accreted recently ($\age_{\rm infall} < 2 \Gyr$) and $> 80\%$ when considering only dynamical analogs of the LMC. Based on these results, if \picII originated as a member of the LMC group then it would have a radial velocity that is within $\roughly 150 \kms$ of that of the LMC. \citet{Jethwa:2016} used dedicated simulations to model the dynamics of the Milky Way, LMC, and SMC, and concluded that $30\%$ of the Milky Way's satellite galaxies originated with the LMC. They predict that satellites of the LMC are distributed within $\pm 20\degree$ of the plane of the Magellanic Stream \citep[MS;][]{Nidever:2008} and would be preferentially found at positive MS longitudes in a leading arm of satellites (\figref{sims}). The MS coordinates of \picII, $(L_{\rm MS}, B_{\rm MS}) = 9\fdg58, 11\fdg11$, lie within the preferred region for Magellanic satellites and are well-aligned with the putative plane connecting the LMC, SMC, and the DES-discovered satellites with $B_{\rm MS} < 0\degree$ \citep{Jethwa:2016}. Furthermore, the simulations of \citet{Jethwa:2016} predict that \picII has a line-of-sight velocity in the Galactic standard of rest (GSR) in the range of $15 \kms < v_{\rm GSR} < 175 \kms$ (68\% interval). Taken together, the photometric properties of \picII and recent simulations of the Magellanic system support the hypothesis that \picII is a dwarf galaxy that arrived at the Milky Way as part of the Magellanic system. However, kinematic measurements are required to confirm the past or present relationship between \picII and its massive nearby neighbors. If \picII is confirmed to be a gravitationally bound galactic companion of the LMC, it would be the most direct example of a satellite of a satellite within the Local Group, further supporting the standard cosmological framework of hierarchical structure formation. The fortuitous discovery of \picII in early \maglites data will be followed by more comprehensive searches for satellite galaxies once additional data are collected. \begin{figure} \includegraphics[width=\columnwidth]{jethwa_mc_sims.pdf} \caption{ \label{fig:sims} Phase space coordinates of \picII (gold star) relative to the simulated distribution of LMC satellites from \citet{Jethwa:2016}, represented by colored contours. Recently discovered DES satellites and Hydra II are shown with cyan markers. \textit{Top}: The density of simulated LMC satellites projected onto the sky in MS coordinates. The DES and \maglites footprints are outlined in black and green respectively. \textit{Middle}: The density of simulated LMC satellites with $5\degree < {\rm B_{MS}} < 25\degree$ projected onto the plane of Galactocentric radius and MS longitude. \textit{Bottom}: Distribution of line-of-sight velocities in the Galactocentric standard reference frame for simulated satellites of the LMC. The black dashed line represents the MS longitude of \picII. Figure adapted from \citet{Jethwa:2016}. } \label{fig:association} \end{figure} | 16 | 9 | 1609.02148 |
1609 | 1609.05903_arXiv.txt | We investigate the reliability of mass estimators based on the observable velocity dispersion and half-light radius $R_\mathrm{h}$ for dispersion-supported galaxies. We show how to extend them to flattened systems and provide simple formulae for the mass within an ellipsoid under the assumption the dark matter density and the stellar density are stratified on the same self-similar ellipsoids. We demonstrate explicitly that the spherical mass estimators~\citep{Walker2009,Wolf2010} give accurate values for the mass within the half-light ellipsoid, provided $R_\mathrm{h}$ is replaced by its `circularized' analogue $R_\mathrm{h}\sqrt{1-\epsilon}$. We provide a mathematical justification for this surprisingly simple and effective workaround. It means, for example, that the mass-to-light ratios are valid not just when the light and dark matter are spherically distributed, but also when they are flattened on ellipsoids of the same constant shape. | Accurate estimates of the dark matter content of dwarf spheroidal galaxies (dSphs) are crucial for furthering our understanding of galaxy formation and structure. Calculating reliable mass estimates has historically been an awkward problem as with only line-of-sight (l.o.s.) velocity measurements the mass profile of a spherical galaxy can only be inferred by making an assumption about the degree of velocity anisotropy i.e. the ratio of radial to tangential motion. Through comparisons to solutions of the Jeans equations, it has been shown that the mass contained near the half-light radius of a dispersion-supported galaxy is approximately independent of the velocity anisotropy and the radial profile of the dark and luminous matter and is simply related to the half-light radius $\Rh$ and the luminosity-averaged l.o.s. velocity dispersion $\sqrt{\slosT}$. There exist several different forms for these formulae in the literature \citep{Walker2009,Wolf2010,Amorisco2012,Campbell2016} that may be summarised as \begin{equation} M_\mathrm{sph}(<r_x) = \frac{C_x \slosT\Rh}{G} \label{Eqn::WalkerWolf} \end{equation} where $M_\mathrm{sph}(<r_x)$ is the mass contained within a sphere of radius $r_x$ and $G$ the familiar gravitational constant. $C_x$ is a constant that depends on the choice of radius $r_x$. \cite{Walker2009} proposed that if $r_x=\Rh$ then $C_x=2.5$ based on a simple example of the stellar distribution following a Plummer profile and the dark matter following a cored isothermal profile although this was validated through fuller testing. \cite{Wolf2010} demonstrated that for $r_x\approx\tfrac{4}{3}\Rh$ (approximately the 3D spherical half-light radius for a range of observationally-motivated profiles) that $C_x=4$ reproduced the results from full Jeans analyses and was also shown to be mathematically true under the assumption of a near-flat velocity dispersion profile. Although spherical mass estimators have proved useful for understanding dSphs, they cannot give the full picture as they do not consider the fundamentally aspherical shape of these galaxies. Our aim in this Letter is to find mass estimators equivalent to equation~\eqref{Eqn::WalkerWolf} applicable to flattened systems. We begin by inspecting the validity of the spherical mass estimators and go on to investigate the applicability of the estimator when considering flattened systems in which the dark and light matter are stratified on the same self-similar ellipsoids. We give formulae similar to equation~\eqref{Eqn::WalkerWolf} that may be used when the 3D shape of the system is known. By marginalizing over prior assumptions on the intrinsic shape and alignment, we show how the mass can be estimated when the intrinsic shape and alignment are not known. | This {\it Letter} has answered the question: how should the mass of a flattened, dispersion-supported galaxy like a dwarf spheroidal be estimated? If the galaxy were spherical, then the answer is well-established. Accurate mass estimators depending on the observable half-light radius and the velocity dispersion of the stars have been devised by a number of investigators \citep{Walker2009,Wolf2010,Amorisco2012,Campbell2016}. We have shown how to modify the spherical mass estimators so that they work for flattened systems in which the light and dark matter are stratified on the same concentric self-similar ellipsoids. This represents a limiting case as simulations indicate the dark matter distribution is in fact rounder than the light \citep{Ab10,Ze12} due to baryonic feedback effects, particularly for the more massive dSphs. The modifications require knowledge of the intrinsic shape and alignment of the triaxial figure and reproduce the mass within ellipsoids by deprojecting the half-light radius and line-of-sight velocity dispersion. The resulting mass estimates are independent of details of the radial profile and are as accurate as the corresponding spherical formulae. This would be of little use if we require knowledge of intrinsic properties. However, we have also shown that, when averaging over triaxial configurations that are consistent with the observed ellipticity $\epsilon$, major-axis half-light length $R_\mathrm{h}$ and line-of-sight velocity dispersion, the mass within the half-light ellipsoid is well approximated by the spherical mass estimate using the `circularized' half-light radius of $R_\mathrm{h}\sqrt{1-\epsilon}$. The scatter in the estimate increases with ellipticity but is only $10-20\percent$ for $\epsilon\sim0.6$. In turn, this observation implies that mass-to-light ratios using spherical estimators, together with a luminosity of $L_{1/2}=L/2$, are accurate and insensitive to the flattening of the dSph. This therefore provides a surprisingly simple, flexible and effective way to account for the effects of flattening. \clearpage | 16 | 9 | 1609.05903 |
1609 | 1609.07137_arXiv.txt | The rate at which matter flows into a galactic nucleus during early phases of galaxy evolution can sometimes exceed the Eddington limit of the growing central black hole by several orders of magnitude. We discuss the necessary conditions for the black hole to actually accrete this matter at such a high rate, and consider the observational appearance and detectability of a hyperaccreting black hole. In order to be accreted at a hyper-Eddington rate, the infalling gas must have a sufficiently low angular momentum. Although most of the gas is accreted, a significant fraction accumulates in an optically thick envelope with luminosity $\sim L_\edd$, probably pierced by jets of much higher power. If $\dot M > 10^3 M_\edd$, the envelope spectrum resembles a blackbody with a temperature of a few thousand K, but for lower (but still hyper-Eddington) accretion rates the spectrum becomes a very dilute and hard Wien spectrum. We consider the likelihood of various regimes of hyperaccretion, and discuss its possible observational signatures. | \label{sec:Introduction} Episodes of hyperaccretion --- accretion at rates far exceeding the Eddington limit --- are often invoked to explain the early rapid growth of massive black holes \citep[MBHs:][and references therein]{volonteri05,volonteri15}. Theoretical estimates of mass supply rates available in protogalaxies are certainly compatible with hyperaccretion. For example, the characteristic infall rate of self-gravitating gas in a halo with velocity dispersion $100 \sigma_{100}\kms$, \beq \dot M_{\rm ff} = {\sigma^3\over G} = 240 \sigma_{100}^3 \msun \ {\rm yr}^{-1}, \label{msg} \eeq exceeds the Eddington rate for a $10^6 m_6 \msun$ black hole by a factor $\sim 10^5 \sigma_{100}^3 m_6^{-1}$. Here we have defined the Eddington accretion rate assuming electron scattering opacity and without an overall radiative efficiency, i.e., $\dot M_\edd = L_\edd / c^2$. Observational surveys suggest that the distribution of mass supply rates in active galactic nuclei (AGN), normalized to the Eddington value, is mass-independent, and is a decreasing function of the Eddington ratio but not a steep one, e.g., a power-law with index $\sim -0.65$ \citep{aird12}, which \cite{aird13} suggest to steepen to $\sim -2$ or cut off with an exponential \citep{stanley15} for Eddington ratios larger than unity. If we instead extrapolated these power-laws to high accretion rates we would infer a non-negligible fraction of supercritical AGN, $\sim 10^{-3}$ at $z=1$ and $\sim 10^{-2}$ at $z=2$. These fractions, however, are derived from extrapolation of results for low-redshift ``normal" AGN. For gas-rich protogalaxies at high redshifts, particularly following mergers, or for black holes that have not yet grown to their final masses, the occasional availability of gas at a hypercritical rate is more likely, as the same fraction of $\dot M_{\rm ff}$ represents a higher fraction of $\dot M_\edd$ for a lower-mass black hole. Indeed, large-scale simulations of galaxy assembly suggest that supercritical infall rates are fairly common \citep[e.g.,][]{dubois14}. The availability of gas at a hypercritical rate is a separate question from the acceptance of such gas by the black hole; the latter is the subject of much uncertainty. X-ray binaries with hyper-Eddington mass transfer rates, such as SS 433 \citep{begelman06}, microquasars in outburst \citep{neilsen16} and ultraluminous X-ray sources \citep{poutanen07,middleton15,pinto16}, appear to eject much of the supplied gas before it reaches the black hole. This regulates the accretion luminosity to a moderately supercritical value, as posited in the inflow-outflow models of \cite{shakura73} and \cite{blandford99}. These systems have in common that the mass is supplied through a thin disk, with nearly Keplerian angular momentum. Only after the gas passes the trapping radius, $r_{\rm tr} = (\dot M / \dot M_\edd) r_{\rm g}$ \citep{begelman79}, where $r_{\rm g}= GM/c^2$ is the black hole's gravitational radius, does radiation energy density build up in the flow, thickening the disk and apparently driving the outflow. The mechanism driving the outflow in specific cases is not well understood, but it could be that the combination of strong angular momentum transport, in conjunction with relatively weak photon trapping, allows enough gas to escape at each radius that the radiation remains marginally trapped all the way in to the center. Although the outer geometry of accretion flows in AGN is not well known, it is possible that high angular momentum disk accretion similarly occurs there, so that few if any AGN produce hypercritical luminosities. The situation appears to be different in a small subset of candidate tidal disruption events (TDEs), where debris from the disrupted star falls back toward the black hole at a hypercritical rate and appears to be accreted without difficulty, liberating a hypercritical luminosity and powering jets \citep{zauderer11,cenko12,brown15}. A key difference between this case and the cases with strong mass loss and regulated accretion is that the specific angular momentum of the infalling gas is far below the Keplerian angular momentum at the trapping radius. The radiation is therefore strongly coupled to the gas where the angular momentum is deposited, and as a result it may not be able to drive strong mass loss. To describe this case, \cite{coughlin14} proposed that instead of being blown away, the infalling gas would inflate into a weakly bound envelope which they called a ZEBRA (ZEro-BeRnoulli Accretion flow). Essentially all the matter in such an envelope could be accreted if the gas close to the black hole could be pushed into low binding-energy relativistic orbits before falling in, as in the ``Polish Doughnut" model \citep{jaroszynski80}, or finds a way of venting most of its accretion energy into the centrifugally evacuated funnel around the rotation axis, as seems to happen in the candidate TDEs. In either case, the luminosity leaking out in directions other than the accretion funnel would be limited to roughly $L_\edd$. In this paper we adopt the view that hyperaccretion is bimodal: either most of the gas is blown away and the residual accretion rate is close to Eddington, or the matter is accreted at nearly the rate supplied. The parameter that discriminates between the two cases is the ratio of the specific angular momentum in the supplied gas to its Keplerian value at the trapping radius, which is a function of the mass supply rate, $\dot M$. In section 2 we summarize the parameters that determine the outcome of hyperaccretion, and in section 3 we study the structure and evolution of the ZEBRA envelope expected to develop in the low angular momentum case, expressing our results in terms of the fraction of black hole mass acquired during hyperaccretion episodes. We consider the appearance of highly supercritical black holes in section 4, and discuss the observational consequences in section 5. We summarize our results and conclude in section 6. | We have considered the conditions under which massive black holes in galactic nuclei might accept infalling matter at an extremely super-Eddington rate, and how such hyperaccreting MBHs might be detectable. We adopt a bimodal criterion for hyperaccretion, in which the black hole is able to swallow material at a hypercritical rate only if this matter falls within the radiation trapping radius without first forming a disk. This argument is based on observations of SS 433 and other X-ray binaries undergoing mass transfer at a hypercritical rate, where there is evidence that powerful winds expel most of the supplied mass before it reaches the black hole, in contrast to hyperaccreting TDEs \citep{coughlin14}, where the infalling gas has a very low angular momentum compared to the Keplerian value at the trapping radius and much of this matter seems to reach the black hole. Theoretical arguments suggest that the establishment of a powerful wind requires some process to transfer energy from the inflowing gas to the outflow, the nature of which is not understood \citep{shakura73,blandford99}. In the absence of such a mechanism, there are self-consistent solutions in which the strong outflow is replaced by a gentle circulation or ``breeze" \citep{begelman12a}, in which case supplied gas could accumulate and hyperaccretion would be possible under a wider range of conditions. Thus, our proposed criterion for hyperaccretion represents a conservative view of the process. According to our adopted view, hyperaccretion commences only if matter crossing into the black hole sphere of influence has a small enough specific angular momentum compared to the Keplerian value, typically a few percent or less. While this can be a stringent constraint, it is relaxed considerably for relatively small black holes in protogalactic halos with relatively large velocity dispersions (typically, more massive halos). This condition is most readily met when the growth of the black hole has lagged behind the $M-\sigma$ relation for its host bulge \citep{ferrarese00,gebhardt00,tremaine02}. Nevertheless, the likelihood of meeting the angular momentum condition depends on the outer boundary conditions for the mass supply, which we leave for later investigations. Once this initial angular momentum condition is met, matter begins to accumulate in an envelope at a rate slightly lower than the growth rate of the black hole. As the envelope grows --- in radius as well as mass --- the specific angular momentum in its outer regions increases, which means that the angular momentum constraint for maintaining an episode of hyperaccretion actually weakens with time. The increasing mass of the envelope is needed in order to store the angular momentum left behind when gas is swallowed by the black hole, until some mechanism is able to remove it. Barring such a mechanism, we show that the density distribution in the envelope approaches the slope $\sim - 3/2$ characteristic of free-fall and Bondi accretion. This enables us to estimate the thermal and radiative properties of the envelope, which must radiate at $\sim L_\edd$. While the effective temperature of the outer envelope is typically a few thousand K, the radiation is thermalized only for $\dot M \gta 10^3 \dot M_\edd$; hyperaccreting black holes in this regime would resemble red giants. For lower values of $\dot M / \dot M_\edd$ the the color temperature rapidly increases, until the envelope becomes a hard X-ray source for $\dot M \lta$ a few hundred $\dot M_\edd$. In addition to the isotropic emission from the envelope, we expect hyperaccreting MBHs to produce jets that carry most of the accretion luminosity, which could be orders of magnitude larger than the envelope emission. While the nature of the jet production mechanism is unclear, and may be different in different situations (e.g., magnetic propulsion vs.~driving by radiation pressure), analogy with hyperaccreting TDEs \citep{kara16} suggests that a subset of hyperaccreting MBHs might be most readily detectable though intense, geometrically beamed X-ray emission. Such sources would be rare, however, not only because of beaming but also because we would expect most hyperaccreting MBHs to have relatively small masses, which implies that their lifetimes are short and their numbers relatively small. Our investigation suggests an explanation for why very few if any hyperaccreting MBHs have been identified: truly hyperaccreting sources would not resemble AGN. Either they would be intrinsically X-ray weak because the temperature of the envelope (not a standard accretion disk) is relatively low ($\lta 10^4$ K), or they would be heavily obscured, extremely hard X-ray sources. Given their low masses, envelope emission at $L_\edd$ would be hard to pick out at high redshifts. It is therefore not surprising that standard observational strategies have not detected such sources. Probably, the best hope would be to detect intense X-ray beams from rare sources pointing at us, which could have quasar-like fluxes; this could provide an exciting window into the early growth of supermassive black holes. | 16 | 9 | 1609.07137 |
1609 | 1609.06307_arXiv.txt | We analyze the wavelength-dependent variability of a sample of spectroscopically confirmed active galactic nuclei (AGN) selected from near-UV ($NUV$) variable sources in the \textsl{GALEX} Time Domain Survey that have a large amplitude of optical variability (difference-flux S/N $>$ 3) in the Pan-STARRS1 Medium Deep Survey (PS1 MDS). By matching \textsl{GALEX} and PS1 epochs in 5 bands ($NUV$, $g_{P1}$, $r_{P1}$, $i_{P1}$, $z_{P1}$) in time, and taking their flux difference, we create co-temporal difference-flux spectral energy distributions ($\Delta f$SEDs) using two chosen epochs for each of the 23 objects in our sample on timescales of about a year. We confirm the "bluer-when-brighter" trend reported in previous studies, and measure a median spectral index of the $\Delta f$SEDs of $\alpha_{\lambda}$ = 2.1 that is consistent with an accretion disk spectrum. We further fit the $\Delta f$SEDs of each source with a standard accretion disk model in which the accretion rate changes from one epoch to the other. In our sample, 17 out of 23 ($\sim$74\%) sources are well described by this variable accretion-rate disk model, with a median average characteristic disk temperature $\bar{T}^*$ of $1.2\times 10^5$~K that is consistent with the temperatures expected given the distribution of accretion rates and black hole masses inferred for the sample. Our analysis also shows that the variable accretion rate model is a better fit to the $\Delta f$SEDs than a simple power law. | Active galactic nuclei (AGN) are known to vary across the observable electromagnetic spectrum on timescales ranging from seconds to years. Large time-domain surveys have confirmed variability as a ubiquitous characteristic of AGN \citep[e.g.][]{Vandenberk04,Wilhite05}, ruling out models involving extrinsic factors such as gravitational microlensing, star collisions or multiple supernovae or starbursts near the nucleus \citep[][and references therein]{Kokubo15}. In particular, the origin of UV/optical variability is of great interest for its connection to the AGN central engine, since the UV/optical continuum is thought to arise directly from the accretion disk around a supermassive black hole. A well-established characteristic of AGN UV/optical variability is the "bluer-when-brighter" trend, in which the source is bluer in the bright state than in the faint state \citep{Vandenberk04,Wilhite05,Schmidt12}. (Although a reverse trend was observed for low-luminosity AGN in \citet{2016ApJ...821...86H}). However, the interpretation of the "bluer-when-brighter" trend is still under debate. Whether or not the intrinsic AGN color becomes bluer when it brightens, or if the red color in a low state is caused by contamination from a non-variable component (e.g., host galaxy flux) \citep{Winkler92}, cannot be distinguished due to the difficulties of separating host galaxy light from the AGN. \cite{2009ApJ...698..895K} found that the optical variability of an AGN is well described by a damped random walk (DRW) process, with model-fit timescales that are consistent with the thermal timescale of an accretion disk. A similar conclusion was reached by \cite{2010ApJ...721.1014M} with a sample of $\sim$9000 spectroscopically confirmed quasars in SDSS Stripe 82. Furthermore, \cite{2010ApJ...721.1014M} found that the variability amplitude was anti-correlated with Eddington ratio, implying that the mechanism driving the optical variability was related to the accretion disk. Several prevailing models for the origin of UV/optical variability in the AGN accretion disk include thermal reprocessing of X-ray emission \citep{1991ApJ...371..541K}, changes in the mass accretion rate \citep[e.g.][]{Pereyra06,Li08}, and an extremely inhomogeneous accretion disk \citep{Dexter11,Ruan14}. In an effort to examine these models, several studies have been carried out. For example, \cite{Ruan14} found that a relative spectral variability composite spectrum constructed from SDSS quasars is better fit with an inhomogeneous disk model consisting of multiple zones undergoing independent temperature fluctuations, than simply changes in a steady-state mass accretion rate. While the localized temperature fluctuations may arise from magnetorotational instabilities (MRI) in an accretion flow \citep[see discussions in][]{2009ApJ...704..781H,2012A&A...540A.114J,2013ApJ...778...65J}, it has also been pointed out by \cite{Kokubo15} that the inhomogeneous disk model predicts a weak inter-band correlation, which is contradictory to what is often observed in the SDSS quasar light curves. Kokubo (2015) attributed the successful fitting result in \cite{Ruan14} to the use of a composite difference spectrum, in which the superposition of localized flares at different radii smears out the inter-band flux-flux correlation in individual quasars. On the other hand, \citet[][ hereafter P06]{Pereyra06} presented the first successful result of fitting a composite difference spectrum of SDSS quasars to a standard thin disk model \citep{Shakura73} with changes in accretion rate from one epoch to the next. Recently, many works that try to explain AGN variability have also reached conclusions that are supportive of this variable accretion rate model \citep{Li08,Sakata11}. The model suggests that the "bluer when brighter" trend is caused by intrinsic AGN spectral hardening. The same conclusion was reached in \cite{Wilhite05}, where they found steeper spectral index in the composite flux difference spectrum than that in the composite spectrum. Despite the elegance of this model, there is a discrepancy between the AGN UV-optical variability timescale and the sound crossing and the viscous timescales (by a factor of 10$^3$, see discussion in \autoref{subsec:timescales}) of the accretion disk. Furthermore, \cite{Schmidt12,2014ApJ...783...46K} reported that on the timescales of years, characteristic optical color variability in individual SDSS stripe 82 quasars is larger than the color variability predicted by a steady-state accretion disk with a varying accretion rate. In this paper we continue the investigation of the origin of UV/optical variability in quasars with several improvements. We apply the variable accretion rate model (P06) to the difference-flux spectral energy distributions (SEDs) of individual quasars instead of using a composite quasar difference spectrum, and over a wavelength range broader than previous optical studies ($\approx 1700-9000 \AA$ in observer's frame) thanks to the nearly simultaneous near-UV ($NUV$) observations from \textsl{GALEX}. The sample of AGNs and quasars investigated in this paper is selected from NUV-variability from the \textsl{GALEX} Time Domain Survey (TDS) and is analyzed using broad-band \textsl{GALEX} $NUV$ and Pan-STARRS1 Medium Deep Survey (PS1 MDS) optical $griz$ photometric data. In \autoref{sec:selection}, we describe the sample selection and the spectroscopic data used in this paper. We describe details of the P06 model in \autoref{sec:model}. The creation of $\Delta f$SEDs, the spectral fitting procedures, and the derivation of fundamental AGN parameters are detailed in \autoref{sec:analysis}. We present our result in \autoref{sec:results}. The implications of the results are discussed in \autoref{sec:discuss}. | \label{sec:discuss} \subsection{Emission-line variability} \label{sec:emissionlines} Before masking the bands containing broad emission lines, 11/23 sources in \autoref{tab:results} have large $\chi^2_\nu$ values and are clearly ill-fitted by the change in accretion rate model. One likely explanation is that the the $\Delta f$SEDs of these objects are contaminated by emission line variability. Broad emission lines are known to both vary in flux and in profile \citep[e.g.]{1990ApJ...354..446W,0004-637X-475-1-106}. In extreme cases such as a 'changing-look' AGN that involves the sudden appearance or disappearance of the broad Balmer lines on the timescale of years \citep[e.g.][]{2014ApJ...788...48S}, the broad band flux may experience a dramatic change due to the change in emission line flux. Using XMMLSS\_MOS01-22 as an example, the difference-flux in $g$ band deviates from the model prediction by $\sim$30\% (\autoref{fig:bestfit_good}), which would inevitably result in a poor model fit if we force to fit all bands. In fact, \cite{Wilhite05} also reported that broad emission lines could vary by 30\% as much as the continuum in the spectroscopic variability study of SDSS quasars. It is important to recognize that the amount of flux change contributed by a broad AGN emission line can be non-negligible compared to the change in continuum in broadband. We find that masking the broad-emission-line contaminated bands can improve the poor model fit. Although only 5/11 originally ill-fitted sources have $\chi^2_\nu < $ 3.0 after applying band masks, most of the sources have a qualitatively good model fit even when having $\chi^2_\nu > $ 3.0 (\autoref{fig:bad_fit}). \subsection{COSMOS\_MOS27-07} \label{subsec:exception} While examining the spectra, we found one source, COSMOS\_MOS27-07, that lacks common AGN emission lines even though classified as an AGN from its morphology and light curve. The light curve of COSMOS\_MOS27-07 is shown in \autoref{fig:lc_tde}. COSMOS\_MOS27-07 resides in an early-type host galaxy with colors $u-g = 1.71$ and $g-r = 1.64$. The \textsl{GALEX} TDS catalog reports a maximum variability amplitude ($\Delta m_{max}$) of \textgreater 1.37 mag, with significant variability on timescales of $1-2$ years \citep{Gezari13}. It is unlikely that such amplitude of variability in $NUV$ is caused by core-collapse supernovae since their $NUV$ light curves are powered by expanding shock-heated ejecta, which cool quickly with time. Such variability can be powered by a tidal disruption event (TDE), which features persistent emission in the UV ($\gtrsim$ 1 year). However, the long term plateau in the GALEX light curve is unlike the power-law decay observed in TDEs. Observations in X-ray wavelengths may reveal if this object is a X-ray bright, optically normal galaxy \citep[XBONG,][]{2002astro.ph..3019C}. \subsection{Timescales} \label{subsec:timescales} Due to the time sampling of \textsl{GALEX} TDS, the $\Delta f$SEDs in our active galaxy sample probe a variability timescale of about a year. An assumption in our model is that enough time has elapsed for the the disk to adjust to a different mass accretion rate. However, the timescale associated with mass inflow, the viscous timescale, is on the order of 1000 years assuming a characteristic disk with thickness $h/r = 0.01$, $M = 10^8 M_{\odot}$ and $\alpha = 0.1$ at $r \sim 100 r_g$, where $r_g \equiv 2GM/c^2$ is the gravitational radius. In general, the viscous timescale is given by: \begin{equation} \tau = \left(\frac{h}{r}\right)^{-2}\frac{1}{\alpha \Omega}, \end{equation} where $h/r$ is the thickness of the disk in units of the radius, $\alpha$ the viscosity, and $\Omega$ the angular velocity. To match the timescale we observed, the mass accretion perturbation must originate from a radius close to the UV/optical emitting region or have a much thicker disk ($\tau_{visc} \sim 10$ yr if $h/r = 0.1$). As pointed out in \cite{Pereyra06}, the observed months-to-years variability timescales are more consistent with the sound-speed timescale of an accretion disk. Changes in accretion rate might result in density and pressure perturbations, which can propagate across a proportion of the disk as sound waves. The DRW model proposed by \cite{2009ApJ...698..895K} to describe the optical variability seen in quasar light curves, also shows fits with a characteristic timescale of $\sim$200 days that is consistent with the thermal timescale. Another possibility is that the UV/optical variability is driven by reprocessing of X-ray photons from the inner disk. Under the assumption of a classic thin disk, the UV and optical spectrum arise predominately from spatially separated locations. Therefore, one would expect the time-lag between different bands to be observed as a result of signal propagation. In fact, inter-band correlations were found in short timescale variability in NGC 2617 and NGC 5548 \citep{2014ApJ...788...48S,2015arXiv151005648F} as an evidence supporting the X-ray reprocessing scenario. Following an outburst in NGC 2617, the galaxy has shown disk emission in the UV/optical lagging the X-ray by 2 to 3 days \citep{2014ApJ...788...48S}. This time lag is consistent with the light crossing time from the inner X-ray emitting region to the outer UV/optical emitting region. Although the UV and optical variability seem to correlate well with the X-rays on short ($\sim$ day) timescales, previous studies also reported the presence of pronounced long timescale ($\sim$ year) variability in the optical but not in the X-rays \citep{2009MNRAS.394..427B,2014MNRAS.444.1469M}, suggesting an independent mechanism contributing to the optical variability. Unfortunately, with this study, we can not differentiate between the steady-state spectrum of an X-ray-illuminated source from that of a steady-state accretion disk, since they have the same radial dependence ($r^{-3}$). Simultaneous X-ray and UV/optical observations on the light-crossing timescale ($\sim$ days), not available in this study, would be required to distinguish between these two models. \subsection{Inhomogeneous disk model} The main weakness of the P06 model lies in the discrepancy between the observed continuum variability timescales, and the much longer timescales over which an accretion rate can change globally in the disk (see \autoref{subsec:timescales} for discussion). Inspired by the better agreement with thermal timescales \citep{2009ApJ...698..895K}, \cite{Dexter11} proposed an inhomogeneous disk model in which the disk is separated into multiple zones that each vary on the thermal timescale, with an independent temperature fluctuation conforming to the DRW process. Furthermore, \cite{Dexter11} find that the localized temperature fluctuation model predicts a higher flux at shorter wavelength ($\lambda < 1000$ \AA) than the standard thin disk model, and can better describe the composite HST quasar spectrum from Zhang et al. (1997). \cite{Ruan14} fitted a composite {\it relative} variability spectrum, created by dividing the composite difference spectrum by the composite quasar spectrum, in the wavelength range of $\lambda = 1500 - 6000$ \AA. They found that the composite relative variability spectrum is better described by the thermal fluctuation model than the standard thin disk model, assuming a fixed 5\% increase in the mass accretion rate for the standard thin disk model. Unfortunately, we cannot directly compare our results by fitting a "relative variability'' SED, because of emission line contamination in our broadband photometry. However, the motivation for constructing the relative variability spectrum in \cite{Ruan14} was to get rid of the effect of internal AGN extinction. However, when we measure the spectral indices of the single-epoch optical spectra of our active galaxy sample, we find only 5 out of 23 targets with a spectral index redder than the spectral index of the SDSS quasar composite $\alpha_\lambda \approx$ 1.56 \citep{Vandenberk01} that might indicate the presence of internal extinction. In addition, we suspect that comparing the composite relative variability spectrum to the model difference spectra divided by a low-state disk model, is not a fair comparison, since the low-state disk model, as we discussed in \autoref{sec:results}, is much bluer than the observed composite quasar spectrum. In sum, we don't think that internal extinction is a significant issue for our sample, and we find a good agreement between the P06 model and our variability SEDs. Recent work of \cite{2016arXiv160503185C} shows the revised thermal fluctuation model with radius-dependent characteristic timescales also shows noticeable departures from the prediction of a standard thin disk model at $\lambda \aplt 1000$ \AA. However, since the shortest rest wavelength probed in our sample is $\approx$ 800 \AA, we cannot look for this characteristic signature. Our analysis shows that, on an object by object basis, the P06 model is an appropriate description of the individual $\Delta f$SEDs of the AGNs and quasars. We do not find the need to invoke the thermal fluctuation model in most of the cases. However, the NUV flux in ELASN1\_MOS12-05 and GROTH\_MOS03-13 is indeed higher than predicted by the P06 model, introducing localized thermal fluctuations might be able to explain the flux excess in NUV. | 16 | 9 | 1609.06307 |
1609 | 1609.08630_arXiv.txt | Directionally sensitive dark matter (DM) direct detection experiments present the only way to observe the full three-dimensional velocity distribution of the Milky Way halo local to Earth. In this work we compare methods for extracting information about the local DM velocity distribution from a set of recoil directions and energies in a range of hypothetical directional and non-directional experiments. We compare a model independent empirical parameterisation of the velocity distribution based on an angular discretisation with a model dependent approach which assumes knowledge of the functional form of the distribution. The methods are tested under three distinct halo models which cover a range of possible phase space structures for the local velocity distribution: a smooth Maxwellian halo, a tidal stream and a debris flow. In each case we use simulated directional data to attempt to reconstruct the shape and parameters describing each model as well as the DM particle properties. We find that the empirical parametrisation is able to make accurate unbiased reconstructions of the DM mass and cross section as well as capture features in the underlying velocity distribution in certain directions without any assumptions about its true functional form. We also find that by extracting directionally averaged velocity parameters with this method one can discriminate between halo models with different classes of substructure. | \label{sec:introduction} The search for dark matter (DM) via the measurement of keV-scale nuclear recoils in dedicated low-background underground detectors has a unique and potentially powerful directional signature. The relative motion of the Solar system with respect to the non-rotating DM halo of the Milky Way should give rise to an anisotropic flux of DM particles with a peak incoming direction coinciding with the constellation of Cygnus~\cite{Spergel:1987kx}. This peak direction is typically regarded as a `smoking gun' signal for a particle of Galactic origin, as it is not mimicked by any known cosmic or terrestrial background . As such, the measurement of nuclear recoil directions consistent with this predicted direction is a powerful tool for both the discovery of dark matter~\cite{Copi:1999pw,Morgan:2004ys,Billard:2009mf,Green:2010zm,Mayet:2016zxu} as well as continuing the search at cross sections below the neutrino floor~\cite{Grothaus:2014hja,O'Hare:2015mda}. Additionally, directional detection may be the only way of measuring the full three-dimensional velocity distribution of DM at the Earth's Galactic radius~\cite{Billard:2010jh,Billard:2012qu,Lee:2012pf,O'Hare:2014oxa}. This in turn may give insights into the the process of galaxy formation and the merger history of our own Milky Way. For a recent review of the discovery reach of directional detection experiments see Ref.~\cite{Mayet:2016zxu}. Measuring the direction of nuclear recoils at the keV scale is experimentally challenging. A variety of prototype experiments are currently in operation utilising a range of novel techniques to extract directional information from a nuclear recoil signal (see e.g.,~Refs.~\cite{Cappella:2013,Capparelli:2014lua,Aleksandrov:2016fyr}, as well as Ref.~\cite{Ahlen:2009ev} for a review). One promising approach is to use a gaseous time projection chamber (TPC) at low pressure in order for the track of electrons ionised by a nuclear recoil to be large enough to detect at around $\mathcal{O}(1\,{\rm mm})$ in size. The direction of this recoil can be inferred by drifting the liberated electrons to a time sampled pixelised anode to reconstruct the 3 dimensional orientation of the track. Experiments such as MIMAC~\cite{Riffard:2013psa,Riffard:2016mgw}, DRIFT~\cite{Daw:2011wq,Battat:2014van}, NEWAGE~\cite{Nakamura:2015iza,Miuchi:2010hn}, DMTPC~\cite{Monroe:2011er,Leyton:2016nit} and D3~\cite{Jaegle:2011rn} currently make use of this technology in some variant. Attempts to measure recoil directionality encounter a range of experimental difficulties on top of the usual challenges found in direct detection experiments. The most immediate limitation of gas TPCs is their ability to be scaled to competitive detector masses, with the largest of these prototype experiments currently operating around the 0.1 kg scale~\cite{Santos:2011kf}. There are also challenges that arise in accurately reconstructing the 3-dimensional recoil track. Most notably there is the problem of head-tail recognition - the measurement of the sense of the nuclear recoil (i.e., $+\qhat$ or $-\qhat$) - which has proven to be difficult to achieve~\cite{Billard:2012bk} and has been shown to have a significant impact on the discovery potential of directional experiments~\cite{Green:2007at,Billard:2014ewa}. The expected event rate in direct detection (DD) experiments depends crucially on the astrophysics of the local halo. In particular, a failure to properly account for uncertainties in the DM velocity distribution may lead to biased measurements of the DM mass and cross section from a future signal~\cite{Peter:2011eu}. It will therefore be imperative to include these uncertainties in fits to direct detection data. This can be done by fitting to phenomenological models for the local distribution~\cite{Billard:2010jh,Lee:2012pf,O'Hare:2014oxa}, or by attempting to integrate out the astrophysics dependence of the DM signal so that comparisons can be made between exclusion limits from different experiments in a `halo-independent' way~\cite{Fox:2010bz,Fox:2010bu,Frandsen:2011gi,Gondolo:2012rs,DelNobile:2013cta,Fox:2014kua,Feldstein:2014gza,Anderson:2015xaa,Gelmini:2016pei,Kahlhoefer:2016eds}. Alternatively one can use empirical parametrisations of the speed distribution to account for astrophysical uncertainties, although this may lead to weakened constraints on other DM parameters~\cite{Peter:2011eu,Kavanagh:2013wba,Kavanagh:2013eya}. Here, we extend the use of general parametrisations of the speed distribution to the fitting of the {\it velocity} distribution with directional data.\footnote{We distinguish here between the distribution of the 3-dimensional vector \textit{velocity} $\mathbf{v}$ and the scalar \textit{speed}, given by $v = |\mathbf{v}|$.} Following the formalism introduced in Ref.~\cite{Kavanagh:2015aqa} we will test a binned approach for parametrising the full 3-dimensional local velocity distribution with directional detectors in a model independent way. In this approach the velocity distribution is divided into angular bins, each described by an empirical 1-d speed distribution which does not vary with angle over the bin. The goal of this work is to use mock data to test the accuracy of the reconstucted DM signal using this empirical method compared with model-dependent approaches in both energy only and directionally sensitive direct detection experiments. We compare reconstructions of the DM mass, standard spin-dependent cross section and velocity distribution in three distinct cases: a) when the velocity distribution is known exactly; b) when the general functional form of the distribution is known (as in Refs.~\cite{Billard:2010jh,Lee:2012pf,O'Hare:2014oxa}); and c) when no assumptions are made about the velocity distribution. To begin in Sec.~\ref{sec:Benchmarks} we will review the relevent directional detection theory and list the benchmark particle and astrophysics parameters that we will attempt to reconstruct. In Sec.~\ref{sec:paramrecon} we describe our mock experimental setups, statistical analysis and methods for reconstructing the velocity distribution. In Sec.~\ref{sec:results} we present the results of reconstructions of the DM mass and cross section as well as the shape and parameters of the velocity distribution. We also include results for directional experiments that lack the ability to tell the forward or backward going sense of observed nuclear recoils. Finally, we discuss the implications of these results and summarise in Sec.~\ref{sec:conc}. | \label{sec:conc} We have explored a number of methods for reconstructing the DM velocity distribution from future directional experiments. We have focused in particular on using a general, empirical parametrisation to fit the velocity distribution and compared this with the case where the underlying form of the velocity distribution is known. This allows us to understand whether the two methods lead to different reconstructed parameter values (which may be indicative of biased reconstructions) and how much the constraining power of the experiments changes as we open up the parameter space with a more general fit. Previous works have demonstrated that the DM mass can be recovered from non-directional direct detection experiments without making assumptions about the form of the speed distribution \cite{Peter:2011eu,Kavanagh:2013wba}. As we show in Fig.~\ref{fig:mx-recon}, such astrophysics-independent approaches can be successfully extended to directional experiments. In particular, the use of an approximate, discretised velocity distribution does not spoil the accurate reconstruction of the DM mass. Our empirical parametrisation typically leads to larger uncertainties than when the underlying form of the distribution is known, but we see no evidence of bias. The DM mass reconstructed using the two methods is similar in almost all cases and the true DM mass of 50 GeV is always enclosed within the $95\%$ confidence intervals over a range of halo models. In principle, we should also be able to recover the DM velocity distribution as well as the DM mass. In order to make the fitting procedure tractable, we have discretised the velocity distribution into $N=3$ distinct angular bins. As demonstrated in Sec.~\ref{sec:shape}, looking at the speed distribution $f^k(v)$ within each angular bin may allow us to pick out key features but it is generally difficult to make comparisons with different possible underlying velocity distributions. Instead, we construct confidence intervals for $\vy$ and $\vT$, the average DM velocity parallel and transverse to the direction of the Earth's motion. These measures of the shape of the distribution allow us to distinguish robustly between different underlying halo models. Although a perfect reconstruction of the full velocity distribution is difficult even with large event numbers, we have shown that this model independent approach can be used as a first step in identifying deviations from the assumption of the SHM to point towards the existence of substructures. In principle one could then move to a particular model dependent parametrisation which would be able to measure the substructure more accurately and extract the astrophysically meaningful parameters. We find that with directionality only in a Fluorine experiment, it may be possible to detect or reject the presence of a substantial stream with 95\% confidence. More isotropic features, such as a debris flow, are more difficult to distinguish from the SHM. Adding directionality in a Xenon experiment allows us to break degeneracies in the shape of the velocity distribution and leads to good discrimination between models with and without a stream. The SHM and SHM+DF models remain harder to distinguish using this method, whether the underlying functional form is known or not. In experiments without the ability to determine the sense of the nuclear recoils we see the discretised approach suffer. This is because the $N=3$ binning is effectively reduced to 2 as the forward and backward bins are folded. The result of this is that it becomes impossible to precisely measure the average speed in the direction of the folding due to a degeneracy between positive and negative values. This confirms the results of previous studies~\cite{Green:2007at,Billard:2014ewa} finding that the lack of sense recognition greatly reduces the power of directional experiments. The benchmark examples we have chosen in this work enable us to broadly compare the success of a discretised parametrisation of the DM velocity distribution under a range of scenarios. However the parameter space that describes different classes of substructure, for instance streams, is large. It is unlikely that the conclusions drawn from our benchmark (which includes a rather large stream component) can be extended generally over the range of possible stream speeds and directions. However, we have demonstrated that an empirical parametrisation can accommodate a wide range of underlying velocity distributions without a large loss in sensitivity compared to when the functional form is fixed and known. In this work, we have considered only ideal direct detection experiments. Experimental complications such as finite energy and angular resolution, as well as the possibility of lower-dimensional readouts, will of course affect the reconstruction of the DM parameters in real experiments. We note, however, that the angular binning procedure we have used in the empirical reconstructions may be a natural way to account for finite angular resolution. If the angular resolution (typically in the range 20$\,^{\circ}$-80$\,^{\circ}$ \cite{Billard:2012bk}) is smaller than the binning angle (here, $60^\circ$), the inclusion of these effects should have little impact on the results. It is not yet clear, however, what the optimum binning angle (and therefore the optimum number of bins) would be. In spite of these open questions, the study we have presented here shows that for exploring the full three-dimensional local velocity distribution, which is a primary motivation for directional experiments, one can make significant progress without assumptions about the underlying astrophysics. The method we have presented allows one to combine directional and non-directional experiments in a general way in order to accurately reconstruct the DM mass, identify broad features in the DM velocity distribution and perhaps even distinguish different underlying models for the DM halo. | 16 | 9 | 1609.08630 |
1609 | 1609.01527_arXiv.txt | We use Cycle 21 Hubble Space Telescope (HST) observations and HST archival ACS Treasury observations of 30 Galactic Globular Clusters to characterize two distinct stellar populations. A sophisticated Bayesian technique is employed to simultaneously sample the joint posterior distribution of age, distance, and extinction for each cluster, as well as unique helium values for two populations within each cluster and the relative proportion of those populations. We find the helium differences among the two populations in the clusters fall in the range of $\sim$0.04 to 0.11. Because adequate models varying in CNO are not presently available, we view these spreads as upper limits and present them with statistical rather than observational uncertainties. Evidence supports previous studies suggesting an increase in helium content concurrent with increasing mass of the cluster and also find that the proportion of the first population of stars increases with mass as well. Our results are examined in the context of proposed globular cluster formation scenarios. Additionally, we leverage our Bayesian technique to shed light on inconsistencies between the theoretical models and the observed data. \noindent{\it Keywords}: (Galaxy:) globular clusters: general, (stars:) Hertzsprung-Russell and colour-magnitude diagrams, Galaxy: formation | The historical understanding of globular clusters as simple stellar populations has changed quickly and dramatically in the past decade. Overwhelming evidence has amassed that globular clusters are not simply a singular homogenous assembly of stars; rather, the distinct stellar groupings of chemical characteristics within the clusters are interpreted as multiple discrete populations (e.g.: \citealt{Bedin:2004}, \citealt{Gratton:2004}, \citealt{Carretta:2006}, \citealt{Villanova:2007}, \citealt{Piotto:2007}, \citealt{Piotto:2009}, \citealt{Milone:2009}, \citealt{Milone:2012a}). In order to better understand the history and evolution of the Milky Way, it is vital to understand these clusters and their properties. New data from \cite{Piotto:2015} shows the vast majority of Galactic globular clusters host multiple populations. The immense influx of new data has spurred many new studies, both observational and theoretical, on the characteristics of the clusters (\citealt{Bedin:2004}, \citealt{Gratton:2004}, \citealt{Carretta:2006}, \citealt{Villanova:2007}, \citealt{Piotto:2007}, \citealt{Piotto:2009}, \citealt{Milone:2009}, \citealt{Milone:2012a}, \citealt{Piotto:2015}, \citealt{Nardiello:2015}, among others). Differences in helium abundances and the light elements carbon, nitrogen, and oxygen (CNO) produce visible effects seen in ultraviolet color-magnitude diagrams. However, the source of these varying abundances is still unclear, though several possible scenarios have been suggested. Among the proposed mechanisms are asymptotic giant branch (AGB) stars and fast-rotating massive stars (FRMS), where later generations of stars form out of the enriched ejecta of these stars (\citealt{Cottrell:1981}, \citealt{Gratton:2004}, \citealt{DErcole:2008}, \citealt{Decressin:2007a}). Other scenarios, involving accretion onto proto-planetary disks or very massive stars, have also been put forward as possible origins of abundance variations (\citealt{Bastian:2013}, \citealt{Bastian:2015}, \citealt{Denissenkov:2014}). However, as explored in depth by \cite{Bastian:2015} and \cite{Renzini:2015}, none of the currently proposed scenarios are able to fully explain the wide range of abundance patterns observed. A new, unknown mechanism may yet have to be devised to properly explain the range of observations. Extension of observations to ultraviolet wavelengths has provided great gains in photometric evidence of multiple populations, especially with high precision space-based observations from the Hubble Space Telescope. The particular passbands F275W, F336W, and F438W disentangle multiple populations in globular clusters due to their sensitivity to helium and C, N, and O abundances. Specifically, the F275W filter contains an OH band, the F336W filter contains an NH band, and the F438W filter contains both CN and CH bands. These filters highlight distinctions among different metal contents correlated with helium and thus are able to visually show the chemical discontinuities between populations in color-magnitude space. Thus, a careful and thorough analysis of the color-magnitude diagram (CMD) allows a differentiation of the multiple populations of a cluster and their relative abundances. We use the sensitive ultraviolet photometry from \cite{Piotto:2015} as well as visual photometry from \cite{Sarajedini:2007} to investigate the properties of 30 of the Galactic globular clusters. As presented in \cite{Stenning:2016} and \cite{Wagner-Kaiser:2016}, we employ a sophisticated Bayesian technique to fit theoretical models to the observed data. In the past, isochrones have primarily been fit to observations of star clusters by eye, trying different parameters and nudging isochrones in color-magnitude space until a solution looks good (\citealt{Jeffery:2016}). However, with our approach we are able to numerically determine the set of parameters for clusters using an in-depth grid of theoretical models (\citealt{von-Hippel:2006}, \citealt{De-Gennaro:2009}, \citealt{van-Dyk:2009}, \citealt{Stein:2013}, \citealt{Jeffery:2016}). We apply this method to 30 Galactic globular clusters, each fit with two dominant populations of stars, in order to better understand the relationships and origins of multiple populations. In Section \ref{Data}, we discuss the photometry used to explore the multiple population phenomenon. In Section \ref{Methods}, the Bayesian framework is briefly summarized, with further details in \cite{Stenning:2016}. In Section \ref{Results}, our results are compared to previous studies and the primary results from our investigation are presented. We discuss formation scenarios and examine inconsistencies between the models and data in Section \ref{Discussion} and conclude in Section \ref{Conclusions}. | We analyze 30 Galactic globular clusters that appear to harbor two stellar populations. Our analysis simultaneously obtained an age, distance, and extinction for each cluster. At the same time, a helium value is determined for each of the two populations within the cluster as well as the proportion of stars that belong to each stellar population. From this analysis, we draw the following conclusions: 1. We present the largest sample of 30 uniformly analyzed Galactic globular clusters under the multiple population scenario, using a Bayesian statistical technique. 2. The helium enrichment differences between the two populations of these clusters tends to fall into a narrow range of $\Delta$Y $\sim$ 0.04 to 0.11. In all cases, we expect that our estimates of the helium differences are upper limits. 3. Similar to previous work, we see a strong indication that more massive clusters have a higher spread of helium enrichment than less massive clusters, possibly due to a deeper gravitational potential. Under an AGB or FRMS formation scenario, a deep gravitational potential well would assist the cluster in holding on to enriched material. 4. Correlations are observed between of the fraction of stars in population A with both mass and radial location in the Galaxy. This suggests that environment and orbital history may play a significant role in the multi-faceted picture of factors driving multiple population characteristics in clusters. | 16 | 9 | 1609.01527 |
1609 | 1609.03464_arXiv.txt | We report the results of \textit{XMM-Newton} observations of the Galactic mixed-morphology supernova remnant \G350. Diffuse thermal X-ray emission fills the north-western part of the remnant surrounded by radio shell-like structures. We did not detect any X-ray counterpart of the latter structures, but found several bright blobs within the diffuse emission. The X-ray spectrum of the most part of the remnant can be described by a collisionally-ionized plasma model {\sc vapec} with solar abundances and a temperature of $\approx 0.8$ keV. The solar abundances of plasma indicate that the X-ray emission comes from the shocked interstellar material. The overabundance of Fe was found in some of the bright blobs. We also analysed the brightest point-like X-ray source 1RXS J172653.4$-$382157 projected on the extended emission. Its spectrum is well described by the two-temperature optically thin thermal plasma model {\sc mekal} typical for cataclysmic variable stars. The cataclysmic variable source nature is supported by the presence of a faint ($g\approx21$) optical source with non-stellar spectral energy distribution at the X-ray position of 1RXS J172653.4$-$382157. It was detected with the \textit{XMM-Newton} optical/UV monitor in the $U$ filter and was also found in the archival H$\alpha$ and optical/near-infrared broadband sky survey images. On the other hand, the X-ray spectrum is also described by the power law plus thermal component model typical for a rotation powered pulsar. Therefore, the pulsar interpretation of the source cannot be excluded. For this source, we derived the upper limit for the pulsed fraction of 27 per cent. | Mixed-morphology (MM) supernova remnants (SNRs) are characterized by a shell-like morphology in the radio and a centrally filled thermal emission in X-rays \citep[e.g.,][]{rho1998}. This class represents about 8 per cent of total Galactic SNR population and about 25 per cent of all Galactic SNRs observed in X-rays \citep{rho1998}. Two main scenarios were proposed to explain MM SNRs morphology: the first one is based on effects of thermal conduction processes within the SNR interior \citep[e.g.,][]{cox1999} and the other one -- on a cloudlet evaporation \citep[e.g.,][]{white1991} assuming the SNR shock propagation through a cloudy ISM. A fraction of MM SNRs shows enhanced metal abundances \citep[see, e.g.,][]{lazendic2006} that is not properly addressed by traditional models. Determination of MM SNRs properties, such as temperatures, abundances and densities, is important to improve the models of their formation. \begin{figure*} \begin{minipage}[h]{0.49\linewidth} \center{\includegraphics[width=0.87\linewidth,clip]{rosat58amin.eps}} \end{minipage} \begin{minipage}[h]{0.49\linewidth} \center{\includegraphics[width=0.94\linewidth,clip]{snr_xmm_0.4-7.2kev.eps}} \end{minipage} \caption{\textit{ROSAT} 58 arcmin$\times$58 arcmin \G350 field in 0.1--2.4 keV band (left panel) together with the VLA 1.4 GHz radio contours \citep{gaensler1998}. 31 arcmin$\times$31 arcmin exposure and vignetting corrected QPB-subtracted \textit{XMM-Newton} image in 0.4--7.2 keV band is shown in the right panel (the data of MOS and pn detectors are combined). The respective region is shown in the left panel with the dashed square. Square root brightness scale is used for both images. The X-ray source J1726 is marked by the arrows.} \label{fig:img} \end{figure*} \begin{figure*} \begin{minipage}[h]{0.49\linewidth} \center{\includegraphics[width=0.83\linewidth,clip]{snr0.4-7kev.eps}} \end{minipage} \begin{minipage}[h]{0.49\linewidth} \center{\includegraphics[width=0.83\linewidth,clip]{snr0.7-1.2kev.eps}} \end{minipage} \begin{minipage}[h]{0.49\linewidth} \center{\includegraphics[width=0.83\linewidth,clip]{snr1.8-2.2kev.eps}} \end{minipage} \begin{minipage}[h]{0.49\linewidth} \center{\includegraphics[width=0.83\linewidth,clip]{snr3-7kev.eps}} \end{minipage} \caption{\G350~images in 0.4--7 keV (top left), 0.7--1.2 keV (top right), 1.8--2.2 keV (bottom left) and 3--7 keV (bottom right) energy bands. The data of MOS and pn detectors are combined. The images are exposure and vignetting corrected and QPB-subtracted. Point sources are removed. The pixel size is 20 arcsec. The intensity is given in counts s$^{-1}$ deg$^{-2}$. In the top left panel, regions used for spectral analysis are shown (see text for details).} \label{fig:snr_images} \end{figure*} SNR \G350~was discovered in the radio with Molonglo and Parkes telescopes at 408 MHz and 5 GHz, respectively \citep{caswell1975}. Very Large Array (VLA) observations \citep{gaensler1998} showed complicated morphology of \G350 that consists of three spatially distinct emission regions: the bright north-western and fainter inner and south-eastern arcs (Fig.~\ref{fig:img}, left panel). The whole extent of \G350 in the radio is $\sim 40$ arcmin. In the optical, \citet{stupar2011} found some H$\alpha$ filaments and clumps spatially coinciding with the radio structures. \G350~was observed in X-rays with \textit{ROSAT} and \textit{ASCA}. The multiwavelength data show that the SNR belongs to the MM SNR type (see Fig.~\ref{fig:img}, left panel). The age of the SNR was estimated as $\sim 10^4$ yr assuming the Sedov expansion phase \citep*{helfand1980,clark1975}. Basing on the radio surface brightness-to-diameter relationship, \citet{case1998} estimated the distance $D$ to \G350~to be about 3.7 kpc. A bright X-ray point-like source, 1RXS J172653.4$-$382157 (hereafter J1726), was detected in the SNR field with \textit{ROSAT} (see Fig.~\ref{fig:img}, left panel) and \textit{ASCA}. It may be an associated neutron star (NS) although no radio pulsar was detected within the SNR \citep{kaspi1996}. On the other hand, it may be an unrelated object. To study the SNR and J1726 properties, we performed \textit{XMM-Newton}\footnote{PI Zyuzin, \textit{XMM-Newton}/EPIC, ObsID 0724220101} observations. Here we present results of the analysis of the \G350 and J1726 X-ray emission. The details of observations and imaging are described in Section~\ref{sec:data}. The analysis of the X-ray SNR spectra is presented in Section~\ref{sec:spec}. J1726 is analysed in Section~\ref{sec:j1726}. We discuss results in Section~\ref{sec:discus} and summarise them in Section~\ref{sec:sum}. | \label{sec:discus} \subsection{The remnant} The spectra of the most part of the SNR can be fitted with {\sc vapec+gauss} model except for the region 6 where the additional PL component is required. As can be seen from Table.~\ref{t:best-fit}, temperatures and absorption column densities obtained for different parts of the SNR are consistent with each other within uncertainties. Plasma in regions 1--2 has solar abundances suggesting that the emission comes from the shocked interstellar medium that is typical for MM SNRs \citep{rho1998}. However, the analysis of bright regions 3--5 revealed the overabundance of Fe in these regions which may indicate the presence of ejecta material. Alternatively, the metal enrichment can be provided by the dust destruction \citep[see e.g.][]{shelton2004}. The measured $N_{\rm H}$ value (Table~\ref{t:best-fit}) allows to independently estimate the distance to the remnant. The empirical $N_{\rm H}$--$A_V$ relation \citep{guver2009} results in the optical extinction $A_V\approx3$ assuming $N_{\rm H}=6.5\times 10^{21}$~cm$^{-2}$. Using $A_V$--distance fit from \citet*{drimmel2003}, we got $D\approx3$~kpc that is compatible with $D=3.7$~kpc obtained by \citet{case1998} from the radio surface brightness. For the distance of 3.7 kpc, the SNR radius is about 20 pc. The gas number density can be calculated from the {\sc vapec} model normalization given as \begin{equation} \label{eq:norm} N=\frac{10^{-14}}{4\pi D^{2}_{\rm cm}}\int n_e n_{\rm H}dV\equiv \frac{10^{-14}}{4\pi D^{2}_{\rm cm}}\times {\rm VEM}, \end{equation} where $n_e$ and $n_{\rm H}$ are the electron and hydrogen number densities, respectively, $D_{\rm cm}$ is the distance in centimetres and VEM is the volume emission measure. For solar abundances, $n_e=1.2 n_{\rm H}$ assuming almost complete ionization. The volume of the particular X-ray emitting region was estimated as $V=7.3\times 10^{56}SL_{20}D_{\rm 3.7kpc}^2$ cm$^3$, where $S$ is the area of this region in arcmin$^2$ and $L_{20}$ is its extension along the line of sight in the units of 20 pc. Then \begin{equation} \label{eq:nh} n_{\rm H}=13.6 N^{1/2}(SL_{20})^{-1/2}. \end{equation} The resulting $n_{\rm H}$ values are presented in Table~\ref{t:con} together with the masses of the emitting gas $M_g$ calculated assuming mean atomic weight for solar abundances $\mu=0.604$. The number density distribution seems to be rather uniform though it depends on the volume estimation. The total gas mass is about 15$M_\odot$ if we assume identical extensions of 20 pc for all regions. \begin{table*} \caption{Visual magnitudes, observed and dereddened fluxes of the J1726 possible optical counterpart obtained from different surveys. Dereddening was done with the interstellar absorption $A_V$ = 0.9 assuming 2-T {\sc mekal} X-ray spectral model (see text for details).} \label{t:flux} \begin{center} \begin{tabular}{cccccc} \hline Filter & $\lambda_{\rm eff}$, $\mu$m & MJD & Magnitude & Flux, $\mu$Jy & Dereddened flux, $\mu$Jy\\ \hline $U$(OM) & 0.344 & 56556.83866 & $20.35\pm0.19$ & $11\pm2$ & $41.7\pm7.6$ \\ \hline $u$ & 0.361 & 56566.01030$^a$ & $19.81\pm0.07$ & $17.8\pm1.2$ & $65.8\pm4.4$ \\ & & 56566.01254$^b$ & $19.74\pm0.07$ & $19.0\pm1.2$ & $70.3\pm4.4$ \\ \hline $g$ & 0.468 & 56566.02029$^a$ & $21.04\pm0.05$ & $15.7\pm0.7$ & $42.1\pm1.9$ \\ & & 56566.02224$^b$ & $21.17\pm0.05$ & $13.8\pm0.7$ & $37.0\pm1.9$ \\ \hline $r$ & 0.624 & 56566.02705$^a$ & $20.33\pm0.06$ & $23.7\pm1.3$ & $48.6\pm2.7$ \\ & & 56566.02785$^b$ & $20.51\pm0.08$ & $20.0\pm1.5$ & $41.0\pm3.1$ \\ & & 56149.08459$^a$ & $20.58\pm0.06$ & $18.8\pm1.0$ & $38.6\pm2.1$ \\ & & 56149.08535$^b$ & $20.66\pm0.06$ & $17.5\pm1.0$ & $35.9\pm2.1$ \\ \hline H$\alpha$ & 0.659 & 56149.07406$^a$ & $19.13\pm0.04$ & $55.4\pm2.1$ & $108.5\pm4.3$\\ \hline $i$ & 0.760 & 56149.09070$^a$ & $20.02\pm0.06$ & $25.4\pm1.4$ & $43.6\pm2.4$ \\ & & 56149.09146$^b$ & $19.76\pm0.07$ & $32.1\pm2.0$ & $55.2\pm3.4$ \\ \hline $Z$ & 0.878 & 55725.26322 & $18.97\pm0.09$ & $58.2^{+5.2}_{-4.8}$ & $87.6^{+7.9}_{-7.3}$ \\ $Y$ & 1.021 & 55725.25799 & $19.05\pm0.13$ & $50.3^{+6.5}_{-5.8}$ & $78.3^{+9.0}_{-8.0}$ \\ $J$ & 1.254 & 55309.36337 & $18.55\pm0.13$ & $58.9^{+7.7}_{-6.8}$ & $74.3^{+9.8}_{-8.6}$ \\ $H$ & 1.646 & 55309.35370 & $17.87\pm0.17$ & $73.1^{+12.5}_{-10.7}$ & $84.9^{+14.5}_{-12.4}$ \\ $K_s$ & 2.149 & 55309.35857 & $17.51\pm0.20$ & $66.9^{+13.3}_{-11.1}$ & $73.8^{+14.7}_{-12.2}$ \\ \hline \end{tabular} \begin{tablenotes} \item $^a$ Primary observation. \item $^b$ Duplicate observation. \end{tablenotes} \end{center} \label{fluxes} \end{table*} \begin{table} \caption{The hydrogen number densities $n_{\rm H}$ and masses of the emitting gas $M_g$ for the different SNR regions.} \label{t:con} \begin{center} \begin{tabular}{ccc} \hline Region & $n_{\rm H}$, & $M_g$, \\ & 10$^{-2}L_{20}^{-1/2}$ cm$^{-3}$ & $L_{20}^{1/2}D_{\rm 3.7kpc}^{2}M_\odot$ \\ \hline \multicolumn{3}{c}{}\\ 1 & $4.7^{+0.4}_{-0.3}$ & $6.2^{+0.5}_{-0.4}$\\ \multicolumn{3}{c}{}\\ 2 & $5.9^{+0.3}_{-0.3}$ & $6.5^{+0.3}_{-0.3}$\\ \multicolumn{3}{c}{}\\ 3 & $5.4^{+0.7}_{-0.5}$ & $0.9^{+0.1}_{-0.1}$\\ \multicolumn{3}{c}{}\\ 4+5 & $5.4^{+0.6}_{-0.5}$ & $1.1^{+0.1}_{-0.1}$\\ \multicolumn{3}{c}{}\\ 6 & $5.4^{+1.8}_{-1.2}$ & $0.3^{+0.1}_{-0.1}$\\ \hline \end{tabular} \end{center} \end{table} \G350~has a complicated non-spherically symmetric morphology which is similar to that of the SNR G166.0+4.3 \citep[see e.g.][]{bocchino2009}. Both SNRs show three radio emitting arcs while the X-ray emission fills the part of the volume enclosed within two arcs. The G166.0+4.3 radio morphology was explained \citet{pinealt1987} as follows. The supernova explodes in a moderately dense medium and then the SN shock passes through a low-density cavity (hot tunnel). \citet{gaensler1998} suggested the similar model for \G350. The inner arc is then formed at the boundary of the cavity. This implies that the observed X-ray emission fills the low-density region, in qualitative accordance with the $n_{\rm H}$ values in Table~\ref{t:con}. The spatially-resolved spectral analysis showed the uniform temperature distribution over the remnant. This is consistent with the predictions of the thermal conduction model \citep[e.g.][]{cox1999}. The conduction timescale can be estimated as \begin{equation} t_{\rm cond}\sim 27\frac{n_e}{1\ {\rm cm}^{-3}}\left(\frac{l_T}{10\ {\rm pc}}\right)^2\left(\frac{T}{0.8\ {\rm keV}}\right)^{-2.5}\frac{{\rm ln}\Lambda}{33}\ {\rm kyr}, \end{equation} where $l_T=T/\nabla{T}$ is the scale length of the temperature gradient and ln$\Lambda=29.7+{\rm ln}n_e^{-0.5}(T/0.086\ {\rm keV})$ is the Coulomb logarithm. For $l_T\approx 20$~pc, $t_{\rm cond}\sim 7$~kyr is less than the estimated Sedov age of \G350~\citep{helfand1980,clark1975}. Therefore thermal conduction may play a role in smoothing the temperature distribution. Direct comparison with results obtained by \citet{cox1999} is difficult since their solutions are constructed assuming spherical symmetry which is not the case of \G350. There exist simulations of the X-ray emission for MM SNRs evolving in the non-uniform density medium. For example, \citet{schneiter2006} presented simulations for the SNR 3C~400.2 assuming the SNR evolution in the medium with a jump in the density and taking into account the effects of thermal conduction and interstellar absorption. Their results for the SNR explosion at the denser side or at the density interface show the X-ray emission pattern remarkably similar to that observed in \G350~(see their Fig.~4). Thus, the \G350~morphology may be qualitatively explained assuming its evolution in a multi-component interstellar medium. However, detailed numerical simulations are required. The cloudlet evaporation model of \citet{white1991} is also frequently used to explain MM SNR properties. This model predicts the radial density gradient which is not observed for \G350~(Table \ref{t:con}). However, this cannot be considered as a solid argument since the model predictions for the non-spherically symmetric case can be different. \subsection{The region 6} The emission from the region 6 is harder than that from the rest of the SNR and the additional power law component is needed to describe its spectrum. Two weak point sources are presented inside the region 6. The first one (RA = $261\fdg790$, Dec = $-38\fdg478$) shows a softer emission (it is not seen in 3--7 keV image). A possible optical counterpart was found for this source in the USNO-B1.0 catalogue (ID 0515$-$0517969), OM, VPHAS+ and VVV images. The second source (RA = $261\fdg794$, Dec = $-38\fdg485$) has hard emission and does not have an optical counterpart. It is possible that the PL emission results from the incomplete subtraction of this source by {\sc cheese} task. We tried to use larger aperture of 30 arcsec to mask the source but this did not remove the PL component. On the other hand, extracting spectra of the region 6 including point sources\footnote{ The small distance between the sources does not allow to separate their emission unambiguously.} and fitting them with {\sc vapec}+PL model resulted in a similar $\Gamma=1.9\pm0.5$ and a larger PL normalization $N_{\rm PL}=(4.3\pm 2.5)\times 10^{-5}$ ph cm$^{-2}$ s$^{-1}$ keV$^{-1}$ than in previous case (Table~\ref{t:best-fit}). The unabsorbed 0.3--10 keV flux in the PL component changed from 1.4$\times$10$^{-13}$ to 2.6$\times$10$^{-13}$ erg~cm$^{-2}$~s$^{-1}$. For $D=3.7$~kpc, these numbers correspond to the X-ray luminosity in the range of $(2-5)\times 10^{32}$~erg~s$^{-1}$. The obtained PL parameters are typical for pulsar+pulsar wind nebula (PWN) systems with ages $\gtrsim 10$ kyr \citep{kargaltsev2008}. This together with the spatial extent of the region~6 of $1-2$ arcmin allows us to suggest it as a PWN candidate. However, we do not see any diffuse emission that resembles a PWN in the VLA 1.4 GHz image though it may be blended with the emission of the inner arc. As an alternative, to check if the hard emitting component in the region 6 may have a thermal origin, we fitted its spectrum with a {\sc vapec}$+${\sc vapec} model. The fit is acceptable with $\chi^2$/d.o.f. = 75/67 and suggests the presence of warm and hot plasma components. Within uncertainties, the warm {\sc vapec} component has the same parameters as in the {\sc vapec}$+$PL case, while the best fit temperature for the hot component is about 12 keV with a 90\% lower limit of 3.6 keV. This is atypical for SNRs where the second hot thermal component, if observed, has a temperature of $\la$ 3 keV \citep[e.g.][]{kawasaki2005} and makes the thermal interpretation of the hard component less plausible especially for an evolved SNR such as \G350. \subsection{J1726} \begin{figure} \begin{minipage}[h]{0.5\linewidth} \center{\includegraphics[scale=0.38,clip]{sed_j1726.eps}} \end{minipage} \caption{Dereddened flux density vs. frequency for the J1726 possible counterpart taken from Table~\ref{t:flux}. Data points for different observations are shown by different symbols as indicated in the inset.} \label{fig:opt_spec} \end{figure} The J1726 X-ray spectrum is well described by PL+BB(NSA) model with parameters typical for rotation powered pulsars. The corresponding column density is consistent within uncertainties with that of the SNR supporting the NS origin of the source. The detection of pulsations could confirm this, but pulsations were not found (Section~\ref{sec:j1726}). The derived upper limit for PF of 27\% is non-informative since PFs of many NSs are lower than this value. Moreover, the actual period for the putative NS can be smaller than the Nyquist limit for our observations of $\approx 150$ ms. Higher time resolution observations are required to solve this problem. An optical source with a non-stellar spectral energy distribution was found at the X-ray position of J1726. Assuming that the optical source is the J1726 counterpart, we calculated the X-ray to optical flux ratio $f_{\rm X}/f_{g}$. The magnitude $g$ was corrected for interstellar extinction adopting the extinction law of \citet*{cardelli1989} and $A_V=3$ (see Subsection 5.1). The unabsorbed X-ray flux was obtained in the 0.3--10 keV energy band (Table~\ref{t:fit-j1726}). We got $f_X/f_{g} \approx 1.5$. This is less than the values obtained for low-mass X-ray binaries ($\sim 10^2-10^3$) or isolated NSs \citep[$\gtrsim 10^3$;][]{universeinxrays}. While a chance spatial coincidence of a NS and an unrelated optical source can not be ruled out, the NS interpretation seems to be unlikely. On the other hand, the spectral energy distribution of the putative optical counterpart points to the CV interpretation. This is also supported by J1726 X-ray spectrum, which is well described by the two-temperature optically thin plasma model typical for CVs \citep{baskill2005,reis2013}. The dereddened optical fluxes obtained in different bands are presented in the last column of Table~\ref{t:flux} and shown in Fig.~\ref{fig:opt_spec}. The dereddening was performed using $A_V=0.9$ which corresponds to $N_{\rm H}$=2$\times$10$^{21}$ cm$^{-2}$ obtained from the 2-T {\sc mekal} fit. The optical source shows a flux excess in a narrow-band H$\alpha$ filter (Table~\ref{t:flux}) that is a common feature for CVs \citep[see, e.g.,][]{witham2006}. The X-ray to optical flux ratio $f_X/f_{g} \approx 7$ is also usual for CVs \citep{palombara2006}. Other interpretations of the J1726 nature are less plausible. For instance, J1726 cannot be an AGN since $N_{\rm H}$ obtained for the single PL fit, typical for AGNi, is much smaller than the total Galactic $N_{\rm H}$ in this direction of $9.5\times10^{21}$ cm$^{-2}$ derived from the H{\sc\,i} map by \citet{dickey1990}. | 16 | 9 | 1609.03464 |
1609 | 1609.06549_arXiv.txt | The atmospheric circulation in Venus is well known to exhibit strong super-rotation. However, the atmospheric mechanisms responsible for the formation of this super-rotation are still not fully understood. In this work, we developed a new Venus general circulation model to study the most likely mechanisms driving the atmosphere to the current observed circulation. Our model includes a new radiative transfer, convection and suitably adapted boundary layer schemes and a dynamical core that takes into account the dependence of the heat capacity at constant pressure with temperature. The new Venus model is able to simulate a super-rotation phenomenon in the cloud region quantitatively similar to the one observed. The mechanisms maintaining the strong winds in the cloud region were found in the model results to be a combination of zonal mean circulation, thermal tides and transient waves. In this process, the semi-diurnal tide excited in the upper clouds has a key contribution in transporting axial angular momentum mainly from the upper atmosphere towards the cloud region. The magnitude of the super-rotation in the cloud region is sensitive to various radiative parameters such as the amount of solar radiative energy absorbed by the surface, which controls the static stability near the surface. In this work, we also discuss the main difficulties in representing the flow below the cloud base in Venus atmospheric models. Our new radiative scheme is more suitable for 3D Venus climate models than those used in previous work due to its easy adaptability to different atmospheric conditions. This flexibility of the model was crucial to explore the uncertainties in the lower atmospheric conditions and may also be used in the future to explore, for example, dynamical-radiative-microphysical feedbacks. | \label{sec:intro} The Venus atmospheric circulation has several characteristics which remain poorly understood, such as the mechanism of formation and maintenance of the atmospheric super-rotation below 100 km. The super-rotation is characterized by a much faster rotation of the atmosphere compared to the rotation rate of the solid planet. This phenomenon quantifies the excess of axial angular momentum that an atmosphere possesses when compared with an atmosphere co-rotating with the underlying planet. From \cite{1986Read} this global phenomenon is quantified using the following equation, \begin{equation} \label{eqn:S} S = \frac{M_t}{M_0}-1, \end{equation} where $S$ is defined as the global super-rotation index, $M_t$ is the total axial angular momentum of the atmosphere and $M_0$ is the axial angular momentum of the atmosphere with zero zonal wind velocities relative to its underlying planet. The total angular momentum of the atmosphere per unit mass ($M_t$) is defined by: \begin{equation} \label{eqn:M} M_t = \int \int \int \frac{m a^2 \cos \phi}{g} d\phi d\lambda dp \end{equation} where $a$ is taken to be the radius of the planet, $\phi$ is the latitude, $\lambda$ is the longitude, $p$ is pressure, $g$ is the gravitational acceleration and $m$ is the angular momentum per unit mass ($m = a \cos \phi (\Omega a \cos \phi + u)$), $\Omega$ is the rotation rate of the planet and $u$ is the zonal component of the wind velocity). Using equations \ref{eqn:S} and \ref{eqn:M} and the observational vertical profiles of zonal winds and their uncertainties from \cite{1985Kerzhanovich}, we can estimate $S = 7.7_{-3.6}^{+4.2}$ (S$ \sim 1.5\times10^{-2}$ for the Earth, \citealt{1986Read}). The vertical profiles from \cite{1985Kerzhanovich} are associated with low latitudes. Three different profiles corresponding to the lowest, mean and highest wind values observed at each altitude, were used to build three global wind fields to compute three different values of $S$ (the mean value and its uncertainties). The three global wind fields were defined using the observational vertical wind profiles at the equator and were extrapolated to the pole region assuming that the atmospheric circulation follows a solid body rotation profile at each altitude. These three dimensional wind fields are axisymmetric. In general, the solid body rotation assumption slightly underestimates the values of the zonal winds at high latitudes, but is a very good approximation for the altitudes where the winds have the largest contribution to angular momentum density (at around 20 km, \citealt{1983Schubert}). In general, dynamical motions in the atmosphere of Venus are driven by a differential insolation in latitude, which might be expected to induce atmospheric circulation in the form of cells with rising atmospheric flow at low latitudes and descending at high latitudes. The presence of middle or high latitude local super-rotation, which refers to the typically barotropically unstable jet, is a consequence of the existence of those cells that transport angular momentum from low toward high latitudes. The presence of a large equator-to-pole Hadley circulation in each hemisphere (e.g., \citealt{1980Schubert}) is due to the slow planetary rotation, which weakens the Coriolis acceleration, increasing the efficiency of the latitudinal heat transport of the atmosphere. The total axial angular momentum of the atmosphere is mainly controlled by the mechanical surface-atmosphere interaction. During the ``spin-up'' of the atmosphere this mechanical interaction pumps axial angular momentum into the atmosphere. More difficult to explain is the presence of the observed equatorial super-rotation. The strong winds in the equatorial region are not produced or maintained by the influence of zonal mean mechanisms (\citealt{1969Hide}). Such super-rotation requires the presence of non-axisymmetric eddy motions, unless super-rotation was its initial condition. Using the equations of motion we can learn more about possible atmospheric mechanisms for generating and maintaining the strong winds in the equatorial region (\citealt{1997Gierasch}). The zonal momentum equation can be written as a conservation equation for axial angular momentum in the form: \begin{equation} \underbrace{\frac{\partial}{\partial t}(\rho m) }_{[A]}+ \underbrace{\frac{}{}\vec{\nabla}\cdot (\rho \textbf{v}m)}_{[B]} + \underbrace{\frac{\partial p}{\partial \lambda}}_{[C]} = \underbrace{\frac{}{}\vec{\nabla}\cdot(\tau \cdot \hat{\textbf{z}}\times\textbf{r})}_{[D]} \end{equation} where $\hat{\textbf{z}}$ is the unit vector in the direction of the planetary angular velocity ($\Omega$) and $\tau$ is the viscous stress tensor. $m$ in this equation is the axial angular momentum per unit mass as mentioned previously. Zonally averaging this equation and assuming the friction is negligible, the terms [C] and [D] drop out. This simplification makes it easier to interpret the conservation of angular momentum in a circulating atmospheric cell. Unless there is convergence of the angular momentum flux, $\textbf{F}_m = \rho \textbf{v} m$, towards a location of maximum angular momentum per unit of mass, it is not possible to produce or sustain, for example, the observed strong prograde winds at low latitudes in the Venus atmosphere. Note that the term ``prograde'' in this work refers to winds in the direction of the planet's rotation. It was demonstrated in \cite{1969Hide} that global or local super-rotation cannot be obtained in a purely inviscid, axisymmetric system that evolved from rest. This result is frequently called ``Hide's first theorem'' (\citealt{1986Read}), and implies that the excess of angular momentum, $S>0$, can only be obtained from non-axisymmetric motions. This non-axisymmetric phenomenon can be represented by the known Reynolds' stress terms (associated with zonal pressure torques) that are defined by: \begin{equation} \label{eqn:fm} [F_m] = \rho a \cos \phi ([u^{\star}v^{\star}],[u^{\star}w^{\star}]) \end{equation} where $\rho$ is the atmospheric density, $a$ is the planet radius, the square brackets denote a zonal average, $v$ is the meridional component of the wind velocity and $w$ is the wind speed in the vertical direction. The stars on each variable mean that they are disturbances in relation to their respective zonal average. The first and second terms are the meridional and vertical components of the eddy fluxes. From Eq. (\ref{eqn:fm}), the weight of each component of the Reynold's stress (horizontal and vertical) is in general related to different possible mechanisms that contribute to the formation and$/$or maintenance of the super-rotation. To be able to complete the puzzle as to the real nature of the general super-rotation, there is a need to identify the atmospheric processes involved in these two terms and quantify their contribution to the phenomenon. The Venus atmosphere has been explored by several space missions in the past: notably the Venera orbiters and entry probes, Pioneer Venus and Magellan, and more recently the European Venus Express mission. These missions made atmospheric data available which increased the interest in the development of global circulation models capable of interpreting these data and guiding their analysis. Numerical modelling of the global Venus atmospheric circulation started more than forty years ago, with the complexity and accuracy improving along the years (e.g. \citealt{1975Kalnay}, \citealt{2003Yamamotoa}, \citealt{2007Lee3} and \citealt{2010Lebonnois}). Recently, typical Venus numerical models that use very simplified representations of radiation and boundary layer processes (e.g., \citealt{2003Yamamotoa}; \citealt{2007Lee3}; \citealt{2013Lebonnois}), have suggested that the global atmospheric super rotation is at least partially maintained by the equatorward momentum transport via synoptic eddies from high latitude barotropically unstable jets. This mechanism is commonly known as the GRW mechanism (\citealt{1975Gierasch}; \citealt{1979RossowWilliams}). Further studies using simplified GCMs have also highlighted other possible mechanisms involving interactions between mid-latitude Rossby waves and equatorial Kelvin waves to form unstable modes that can lead to zonal acceleration in the tropics (\citealt{2010Mitchell}; \citealt{2014Potter}). Evolving towards more complex physically-based models we find the work by \cite{2007Ikeda} and \cite{2010Lebonnois}, who included for instance a self-consistent computation of temperature using a radiative transfer formulation. In these cases the diurnal cycle is not neglected, which revealed to be an important factor in the atmospheric dynamics produced. The diurnal cycle excites migrating thermal tides especially in the Venus cloud region due to the large extinction of solar energy there. The thermal tides play an important role maintaining the super rotation, since they transport prograde momentum vertically and predominantly at low latitudes, from above the cloud region towards the upper cloud deck (e.g., \citealt{1992Newman} and \citealt{2010Lebonnois}). These three momentum transport mechanisms: high latitude barotropic eddies, tropical Rossby-Kelvin instabilities or thermal tides, are thought to be the main possible mechanisms for the formation and maintenance of strong zonal winds at low latitudes: via the $[u^{\star}v^{\star}]$ and $[u^{\star}w^{\star}]$ terms respectively. In this study we have developed a new Venus General Circulation Model called the Oxford Planetary Unified (Model) System for Venus (\textbf{OPUS-Vr}) that includes a new radiation scheme (\citealt{2015Mendonca}). We use it to simulate the Venus atmosphere in a physically consistent manner. The main advantage of our radiation code against previous works is the explicit calculation of the solar fluxes and the easy adaptability to different optical structures. The OPUS-Vr is aimed at studying the atmospheric mechanisms that transport momentum, and explore the range of atmospheric conditions favourable to the formation of an atmospheric circulation similar to the one observed. In the next section we describe the numerical model used in this work. In section \ref{sec:Dbaseline} the results from the reference simulation are explored, and the main momentum transport mechanisms and waves are identified. In section \ref{sec:Dbaseline} we also compared the model results obtained with available observational data. In section \ref{sec:SurfAlb} a sensitivity test to the surface albedo is explored. Finally in sections \ref{sec:dics} and \ref{sec:conclu} a discussion on possible super-rotation mechanisms working within the lower Venus atmosphere is presented followed by the general conclusions. | \label{sec:conclu} Our new model is capable of producing an atmospheric circulation above the cloud base similar to the one observed. The mean circulation simulated is characterised in the cloud region by two planetary-scale Hadley cells in each hemisphere: one near the cloud base and another in the upper clouds. The existence of these two dynamical regions is mainly related to the significant absorption of solar radiation in the upper cloud region and the blocking at the cloud base of upwelling infrared radiation from the hot lower atmosphere. The atmospheric cells on Venus are in general larger than the ones found on the Earth's atmosphere, due to the weak coriolis acceleration on Venus (a slow rotating planet). These circulation patterns transport momentum poleward in the upper branches that drives the formation of mid-latitude jets but weakens the zonal winds at low latitudes. The eddy-zonal flow interactions have a crucial role in replenishing the equatorial region with angular momentum. The model results showed that the nature of the mechanisms involved in the formation of atmospheric super-rotation above and below the cloud base are different. In the upper cloud region, the radiative time-scale is smaller than a Venus day, and several harmonics of the thermal tides are produced. The components with the largest amplitudes comprise zonal wavenumbers one (diurnal tide) and two (semi-diurnal tide). In the upper cloud region, however, it is the wave number two component that has more impact on the atmospheric circulation. The Sun moves slowly in the retrograde direction in relation to the mean flow, which forces the tides to follow its position. This relative motion induces a positive acceleration of the flow in the region where the tides are excited (upper cloud region). In the region where the waves are absorbed (via radiative damping) by the atmosphere, predominantly above the cloud region, the forcing in the atmosphere acts in the reverse direction. The presence of the thermal tides is very clear in the observational data, but the same cannot be said of the other transient waves. The main difficulty in retrieving other eddy motions from observational data is the low resolution of the images available from any space mission to date at low latitudes. We also found that the flow in the upper cloud region is accelerated due to the wave structure of a free equatorial Rossby wave. In this case, the equatorward transport of momentum is achieved by a nonlinear interaction between the free Rossby wave and the equatorial structure of the diurnal tide. Also, this wave transport momentum verticaly due to the tilt of the phase front with altitude. Below the cloud base the radiative time-scale becomes much larger than a solar day, and the influence of the thermal tides in the atmosphere becomes negligible. In general, the mean atmospheric circulation in this region is apparent from the simulations as large, deep, equator-to-pole cells, which extend from the surface to the cloud region. Nevertheless, in common with the LMD Venus GCM (\citealt{2010Lebonnois}), OPUS-Vr was not capable of reproducing strong zonal winds comparable with the observations in this region. More observational and theoretical work is needed to improve our understanding of what are the atmospheric mechanisms driving the circulation in the lower atmosphere. The Venus atmospheric super-rotation is not a temporary state for its current atmospheric conditions, and supporting this idea are the results of long numerical simulations and the consistency between the observations over the last decades. However, numerical and observational studies show that the Venus atmospheric circulation is not steady. As an example, the variability of the zonal wind distribution at jet altitudes is significantly affected by long term oscillations (tens of Venus days) and by the approximately bidiurnal planetary mixed Rossby-gravity waves, which is seen in both observations and simulations. These low frequency waves (also found in the LMD Venus GCM, \citealt{2010Lebonnois}) may be the same ones as found in observations by \cite{2013Khatuntsev}. With our new and flexible radiative transfer scheme, our model presents an excellent capability to explore different atmospheric conditions which is essential to study the Venus atmospheric dynamics. Our model shows great promise for future research in this area, in particular due to its unique ability to explore radiation parameters in the lower atmosphere, where the observational and modelling uncertainties are very high. | 16 | 9 | 1609.06549 |
1609 | 1609.06914_arXiv.txt | \noindent Previous work on protoplanetary dust growth shows halt at centimeter sizes owing to the occurrence of bouncing at velocities of $\stackrel{>}{\sim} 0.1 \unit{m~s^{-1}}$ and fragmentation at velocities $\stackrel{>}{\sim} 1 \unit{m~s^{-1}}$. To overcome these barriers, spatial concentration of cm-sized dust pebbles and subsequent gravitational collapse have been proposed. However, numerical investigations have shown that dust aggregates may undergo fragmentation during the gravitational collapse phase. This fragmentation in turn changes the size distribution of the solids and thus must be taken into account in order to understand the properties of the planetesimals that form. To explore the fate of dust pebbles undergoing fragmenting collisions, we conducted laboratory experiments on dust-aggregate collisions with a focus on establishing a collision model for this stage of planetesimal formation. In our experiments, we analysed collisions of dust aggregates with masses between 1.4 g and 180 g, mass ratios between target and projectile from 125 to 1 at a fixed porosity of 65\%, within the velocity range of 1.5\textendash8.7 $\unit{m~s^{-1}}$, at low atmospheric pressure of $\sim 10^{-3}$ mbar and in free-fall conditions. We derived the mass of the largest fragment, the fragment size/mass distribution, and the efficiency of mass transfer as a function of collision velocity and projectile/target aggregate size. Moreover, we give recipes for an easy-to-use fragmentation and mass-transfer model for further use in modeling work. In a companion paper, we utilize the experimental findings and the derived dust-aggregate collision model to investigate the fate of dust pebbles during gravitational collapse. | Introduction} Over the past decade, a significant amount of work on protoplanetary dust growth has been contributed by modeller and experimenters, which has significantly advanced our understanding about the formation of planetesimals. In the field of planetesimal formation, broad consensus has been reached on the pre-gravitational dust-growth regime in which micrometer-sized dust grains grow to at least centimetre sizes by sticking collisions in protoplanetary discs. Based upon the first complete laboratory-based dust-aggregate collision model by \citet{GuettlerEtal:2010}, \citet{ZsomEtal:2010a} showed that dust aggregates experience a bouncing barrier when they reach millimetre sizes, which limits growth and leads to relatively compact pebble-sized dust aggregates with volume filling factors of $\phi \sim 0.4$ (i.e., 60\% porosity). The further growth from pebbles to planetesimals faces severe obstacles by the absence of direct hit-and-stick processes \citep{GuettlerEtal:2010}, the onset of fragmentation in collisions between dust aggregates of similar size around $\sim 1 \unit{m~s^{-1}}$ \citep{GuettlerEtal:2010} and the strong influence of radial drift, which leads to the rapid depletion of boulders around 1 m in size at 1 AU \citep{Weidenschilling:1977a}. This halt of growth at pebble sizes is in agreement with observations, which show the presence of mm-cm-sized dust particles in protoplanetary disks (see \citet{Testi:2014} for a review). However, \citet{OkuzumiEtal:2012} have shown that under very favourable conditions (sub-micrometer-sized water-ice particles), direct coagulation into planetesimals is feasible. As we are interested in a more generic formation scenario that is less restricted in terms of grain size, particle material and location in the protoplanetary disk, we hereafter do assume that the growth pathway demonstrated by \citet{OkuzumiEtal:2012} is not feasible for micron-sized or warm dust particles. Two competing models of planetesimal formation in the presence of the above-mentioned obstacles have been developed in the past years. Based upon an extensive body of laboratory work on mass transfer in high-velocity collisions between dust aggregates of dissimilar masses \citep{WurmEtal:2005a,TeiserWurm:2009a,GuettlerEtal:2010,Kotheetal:2010,Teiseretal:2011a,DeckersTeiser:2014}, \citet{WindmarkEtal:2012a}, \citet{WindmarkEtal:2012b} and \citet{Garaudetal:2013} describe the direct collisional formation of planetesimals, ignoring particle transport by radial drift. Although mass transfer in the process of fragmentation of the smaller projectile aggregate during an impact into the larger target aggregate has been clearly proven to exist, the formation of planetesimals of kilometre sizes or larger by this process faces severe problems, such as the rather large time scales required \citep{JohansenEtal:2014}, the role of counter-acting erosion \citep{SchraeplerBulm:2011}, and fragmentation in collisions between similar-sized planetesimals. A planetesimal-formation model relying on particle concentration and self-gravity has been proposed by \citet{JohansenEtal:2007}, who showed that the streaming instability, first described by \citet{YoudinGoodman:2005}, is capable of concentrating pebble-sized dust aggregates such that planetesimals can directly form by gravitational instability. Since then, this formation scenario has been refined and proven capable of forming planetesimals of up to several 100 km in size from dust aggregates with Stokes numbers in the range St$\sim 0.01$ to St$\sim 1$ within the radii 1 to 10 AU \citep{BaiStone:2010b,JohansenEtal:2012,CarreraEtal:2015}. Here, the Stokes number is defined as the ratio between the gas-grain coupling time and the inverse Keplerian frequency \citep{Cuzzi:1993}. At 1 AU, this range in Stokes numbers corresponds to cm- to m-sized dust aggregates in a minimum mass solar nebula model. As the formation of dust aggregates at the upper end of the size range faces the above-mentioned drift and fragmentation problems, this planetesimal-formation scenario is likely to operate with pebble-sized rather than boulder-sized dust particles. One of the main issues with previous studies on planetesimal formation via gravitational collapse is the use of inert dust, i.e. dust agglomerates were indestructible. However, numerical simulations predict that they collide with rather high velocities, typically a few $\unit{m~s^{-1}}$ according to \citet{JohansenEtal:2009}. At these velocities, aggregates are supposed to fragment as shown in the model of \citet{GuettlerEtal:2010}. Additionally, during the gravitational collapse of the pebble clouds, speeds high enough for fragmentation can be reached for planetesimals above a few 10 km in size \citep{Wahlberg:2014}. This fragmentation changes the size distribution of the pebbles and thus influences the porosity and packing of the planetesimal that forms. Nevertheless, \citet{Skorov:2012}, \citet{BlumEtal:2014} and \citet{BlumEtal:2015} have shown that the dust activity of comets as they approach the Sun (as well as the low mass density and thermal conductivity) can only be explained by the gravitational instability scenario of planetesimal formation, due to the resulting low tensile strengths of the accreted dust pebbles. In this paper, we will present new experimental work on the collision behavior of cm-sized dust aggregates in the velocity range up to $8.7 \unit{m~s^{-1}}$ for mass ratios between target and projectile agglomerates of 1 to 125. These results will be used in the companion paper (Wahlberg Jansson et al. 2016; hereafter Paper II) to simulate how fragmentation affects the gravitational collapse phase and the interior structure of planetesimals that form by gravitational instability. In Section \ref{sect:exp}, we introduce our new experimental setup. Section \ref{sect:samples} describes the sample preparation and sample properties. In Section \ref{sect:results}, the results and analyses of our experiments are presented. Based upon these results, in Section \ref{sect:fragmod} we propose a simple empirical model to describe the general outcome in aggregate-aggregate collisions, which will be applied in Paper II. Section \ref{sect:simul} describes briefly how we use the new data to better describe the collapse of a pebble cloud. In Section \ref{sect:conclusions}, we conclude our work and discuss its astrophysical implications. | Conclusion and Discussion} We developed a new experimental setup dedicated to the study of the low-velocity fragmentation behavior of porous dust aggregates. Aggregates consisted of micrometer-sized $\rm SiO_2$ grains and possessed volume filling factors of $\phi = 0.35$, i.e. porosities of 65\%. The sizes of the dust aggregates ranged between 1 cm and 5 cm, with collision velocities in the range from $1.5 \unit{m~s^{-1}}$ and $8.7 \unit{m~s^{-1}}$ (see Figure \ref{fig:VD}). In all cases studied, the smaller (or equal) sized projectile aggregate fragmented. The larger (or equal sized) target aggregate survived impact when the target-to-projectile size ratio was large and the impact velocity rather small (see Figure \ref{fig:mtprob}b). However, we found that the outcome in these cases is probabilistic between target survival and target fragmentation, with a probability for target survival given by Eq. \ref{eq:mtprob} (see Figure \ref{fig:mtprob}a). We described the fragmentation of the colliding dust aggregates by the mass of the largest fragment and a continuous area-frequency distribution function of the smaller fragments. When we express the mass of the largest fragments in units of the target-aggregate mass, we can describe its dependence on impact energy with a Hill function (see Eq. \ref{eq:hill}) with two free parameters, the energy $E_{0.5}$ for which the largest fragment is $\mu = 0.5$ and an exponent $n$ for which we find that $n=0.55$. Following our recipe summarized in Sect. \ref{sect:fragmod}, a full description of the fragmentation process in collisions between arbitrary dust aggregates is possible. Besides the application of our high-velocity dust-aggregation collision model in the description of the fate of dust aggregates in collapsing pebble clouds (see Sect. \ref{sect:simul} and Paper II), it will also be useful for mass-transfer based formation models of planetesimals \citep{WindmarkEtal:2012a,WindmarkEtal:2012b,Garaudetal:2013}. The successive growth of dust aggregates beyond the bouncing barrier by mass transfer in catastrophic collisions between dissimilar-sized dust aggregates is an essential part of these models. With the data and formal descriptions of the collision outcomes presented in this paper, the validity of models for the formation of planetesimals by direct sticking via mass transfer can be assessed with more realistic collision outcomes. | 16 | 9 | 1609.06914 |
1609 | 1609.06639_arXiv.txt | It has been shown that some aspects of the terrestrial planets can be explained, particularly the Earth/Mars mass ratio, when they form from a truncated disk with an outer edge near 1.0~au \citep{Hansen:2009p8802}. This has been previously modeled starting from an intermediate stage of growth utilizing pre-formed planetary embryos. We present simulations that were designed to test this idea by following the growth process from km-sized objects located between 0.7--1.0~au up to terrestrial planets. The simulations explore initial conditions where the solids in the disk are planetesimals with radii initially between 3 and 300~km, alternately including effects from a dissipating gaseous solar nebula and collisional fragmentation. We use a new Lagrangian code known as {\tt LIPAD} \citep{Levison:2012p12338}, which is a particle-based code that models the fragmentation, accretion and dynamical evolution of a large number of planetesimals, and can model the entire growth process from km-sizes up to planets. A suite of large ($\sim$ Mars mass) planetary embryos is complete in only $\sim$~1~Myr, containing most of the system mass. A quiescent period then persists for 10-20~Myr characterized by slow diffusion of the orbits and continued accretion of the remaining planetesimals. This is interupted by an instability that leads to embryos crossing orbits and embyro-embryo impacts that eventually produce the final set of planets. While this evolution is different than that found in other works exploring an annulus, the final planetary systems are similar, with roughly the correct number of planets and good Mars-analogs. | An important and challenging issue in understanding terrestrial planet formation are the large differences between Earth and Mars. The two planets are solar system neighbors but are separated by an order of magnitude in mass and accretion age \citep{Nimmo:2007p11241,Kleine:2009p9784,Dauphas:2011p19768}. Classical models of terrestrial planet formation with initial conditions that include smooth disks of solid material extending beyond $\sim$2~au and that started with a population consisting of both km-scale planetesimals and Moon-to-Mars mass embryos generally fail to capture either of these two constraints (see \citealt{Raymond:2009p11530} or a review by \citealt{Morbidelli:2012p11505}). Recent works have explored conditions with non-smooth surface density profiles of solids with some successes being found for a truncated disk with an outer edge at or near 1~au \citep{Hansen:2009p8802,Walsh:2011p12463,Izidoro:2014p15200,Jacobson:2014p18340,Levison:2015p20168}. The altered surface density was found to promote a scattering of Mars-analogs where they then avoid further embryo-embryo impacts and accretion events --- essentially starving Mars, keeping it small and ending its accretion much earlier than the Earth. However, most previous works begin modeling at an intermediate stage of growth - with ``planetary embryos'' amidst a sea of ``planetesimals'' - and then truncate the disk or remove mass in certain regions. However, this does not consider the context of how such initial conditions came to be. For example, if solid material never formed beyond 1.0~au, how differently would the initial generation of planetesimals behave as they grew into the planetary embryos? Would there have been diffusion of material off the sharp edges of the mass distribution? We test an end-member case, essentially an initial disk truncation, and test if a disk of planetesimals (diameters of 10s of km), situated between 0.7--1.0~au grow to become good matches for the terrestrial planets. \subsection{Previous Work} \cite{Hansen:2009p8802} explored a scenario where all solid mass currently in the inner solar system (2~$M_\oplus$) was initially between 0.7--1.0~au, and all of it in 400 similar sized objects (each were 2.98$\times$10$^{22}$~kg, or $\sim$1000~km radius) with no gas effects. These conditions succeeded in producing a good Earth/Mars mass ratio consistently as embryos were scattered off the edge of the annulus early and avoided further growth near 1.5~au. The accretion timescales for Mars-analogs were fast, largely on order of the estimated timescales from cosmochemical studies between 2-10~Myr \citep{Nimmo:2007p11241,Dauphas:2011p19768}. However, the Earth-analogs also formed rapidly, on roughly similar timescales, which is much faster than cosmochemical expectations of 30-100~Myr \citep{Kleine:2004p19341,Kleine:2009p9784}. Seeking a mechanism to truncate a disk of solid material, \citet{Walsh:2011p12463} invoked the inward-then-outward migration of the Jupiter (where the \citet{Walsh:2011p12463} migration of Jupiter is referred to as the ``Grand Tack''). This migration was constrained by the need to produce a disk with a truncation at 1.0~au, similar to that of \citet{Hansen:2009p8802}. The initial conditions included 1/4 to 1/2 Mars mass planetary embryos in a sea of thousands of planetesimals. While these initial conditions were similar to \citet{Obrien:2006p8571} the dynamical effects of Jupiter migrating was a powerful effect and pushed embryos onto crossing orbits. The final planetary systems in this work had a similar mass-semimajor axis distribution as found in \citet{Hansen:2009p8802} and later work by \citet{Obrien:2014p13867} found similar accretion timescales. \citet{Jacobson:2014p18340} investigated the outcomes for ``Grand Tack'' scenarios, where the bi-modal mass distribution was altered . Here, correlations were found such that increasing the mass ratio of embryos to planetesimals leads to longer accretion timescales, but also more dynamically excited final systems of planets. Largely, the radial mass distribution (RMC) of the final planets was unchanged for the wide range of parameters explored. Nebular effects have also been proposed as means to change the surface density profile of the gas disk. Ionization in the gas-disk could lead to regions with very different viscosity creating local mass distribution minimums at the boundaries \citep{Jin:2008p20003}. \citet{Izidoro:2014p15200} used this as motivation to study planet formation in numerous scenarios with depletions of solid material in annular regions beyond 1~au. For deep depletions of material similar edge effects were found as in \citet{Hansen:2009p8802} and \citet{Walsh:2011p12463}. Finally, \citet{Levison:2015p20168} explored the growth of planets directly from cm-sized ``pebbles''. While this work was not designed explicitly to generate an annulus, the accretion efficiency for pebbles (where ``pebbles'' refer to the direct accretion of cm or small particles) is strongly dependent on both the size of the seed body (embryo or planetesimal) and also the stopping time of the rapidly drifting pebbles. Thus \citet{Levison:2015p20168} finds for some initial conditions the rapid growth of embryos inside of $\sim$1.0~au with minimal growth at further distances in the inner solar system -- essentially generating an annulus of very large planetary embryos. While these works have studied formation from an Annulus in different ways they all started modeling at an intermediate stage of growth (note that \citet{Levison:2015p20168} relied on an entirely different mode of accretion). The most numerically tractable point for commencing a model is after the bi-modal mass distribution is established during ``Oligarchic growth'', where after a few to ten million years there may no longer be a gas disk, and there are only tens of embryos amidst a sea of planetesimals. To allow a very large number of planetesimals (thousands), the models typically do not consider gravitational interactions between planetesimals, but the embryos (tens of bodies typically) interact with each other and the planetesimals. As discussed above there are variations on these initial conditions, but nearly all are founded in this initial bi-modal distribution of mass. Similarly, most studies assume that when two bodies collide they merge perfectly, conserving momentum and mass. This is caused by the numerical problem of introducing new particles into the simulation by way of a collisional fragmentation event and the subsequent computational risk of large, and increasing, $N$. While various works have included aspects of fragmentation at different times \cite[see][]{Leinhardt:2009p10318,Kokubo:2010p9520,Chambers:2013p19990,Carter:2015p19516}, only a few models have approached planetesimal to planet simulations \citep[see][]{Kenyon:2006p11683,Morishima:2015p19487}. Finally, most of the works starting in Oligarchic growth phases assume the absence of any gas effects. This is partly due to the expected lifetime of the gaseous solar nebula (2-10~Myr) being similar to the expected times to reach the Oligarchic growth stages. While not computationally difficult to include, the typical absence of gas effects is also partly due to the uncertainties in precisely when and how the gaseous solar nebula dissipated -- whether it was a slow loss of mass, or inside-out or outside-in dispersal. However, we will show that in the simple case of an exponential decay of the nebula, even a very small fraction of the original solar nebula can strongly affect the outcome of the models by stabilizing the system for long periods of time. \subsection{This work} Most previous works' initial conditions include planetary embryos, and thus imply significant previous growth and evolution {\it before} the various truncation/depletion mechanisms happen. None start with an annulus of planetesimals, nor do any constrain how early or late in the stages of growth of the disk that such truncation mechanisms could successfully operate (technically, \citet{Levison:2015p20168} starts with planetesimals, but studies a very different growth process). Here, the question we are trying to answer requires a complete simulation from planetesimals to planets, and thus re-thinking both the initial conditions and also the gas effects requires including particle fragmentation. Specifically, regarding a disk truncation or annulus as a way to address the ``small Mars'' problem, starting with a bi-modal mass distribution implies previous growth --- the embryos have already grown. However, this would assume simple static growth within an annulus not allowing for diffusion of bodies or drift from drag forces. Thus, this test is simple, but relevant to the concept of forming the Earth/Mars mass ratio due to substantial changes to the mass distribution of the solid material in the disk, and may point to required timing or truncation mechanisms to satisfy constraints. We will include fragmentation throughout the simulations and will include a constantly decaying gas disk and all of its effects on the simulation. Note that this is the first in a series of papers exploring the growth of planetesimals to planets, and so related studies will follow. | The intent of this study was to explore whether there were limits on building good Mars-analogs and Earth/Mars mass ratios from a truncated disk or annulus. The end-member case of an annulus of km-sized planetesimals was modeled including gas and fragmentation effects, and found in most cases to produce adequate matches to the observed terrestrial planets, including Mars analogs. However, as explored above, while these may produce similar final systems of planets as in the original \citet{Hansen:2009p8802} work the evolution of the system was vastly different. The systems final properties are therefore potentially less diagnostic of the physics involved in their growth than is the expression of a planet's accretion profile in geochemical data. Rather, an approach that considers the constraints on growth timescales and also the compositional and internal evolution due to different growth profiles may be more important in assessing all of the important factors in Terrestrial Planet formation \citep{Fischer:2014p20911,Dwyer:2015p20491,Carter:2015p19516}. Generally, a Mars-analog in these simulations accumulates most of its mass rapidly through the accretion of planetesimals, avoids late embryo-embryo collisions and its orbit diffuses outward from $\sim$~1~au to its final semimajor axis at $\sim$1.5~au. This growth profile for Mars-analogs is not too different than that found in \citet{Hansen:2009p8802}, where Mars-analogs are scattered out and avoid further embryo-embryo accretion events after a few million years. The differences in these two evolutionary paths would lie primarily in the accretion profiles of Earth and Venus-analogs that alternately have long quiescent periods of accretion compared to somewhat regular embryo accretion events found in more classical models. In this way the systems in this work evolve similarly to that found in \citet{Levison:2015p20168} - whereby that work forms $\sim$15 planetary embryos between 0.7-1.5~au. This system of embryos is built rapidly and are stable for $\sim$10~Myr before embryo-embryo collisions begin. Finally, this work was primarily an exploration of the dynamics of these systems as they grow from planetesimal to planets with and without the affects of a gaseous nebula. These tests were ideal first test cases for {\tt LIPAD} as the small annulus required less total $N$ than will the upcoming full-disk terrestrial planet simulations using the same code. However, the work found that for this test case that the collisional aspects of the code are much less important than is the total impact caused by the presence and lasting nature of the gaseous solar nebula. Upcoming work will expand on this work using similar techniques to study growth in a full-disk (0.7--3.0~au) and scenarios including giant planet migration. | 16 | 9 | 1609.06639 |
1609 | 1609.08624_arXiv.txt | {The outcome of upcoming cosmological surveys will depend on the accurate estimates of photometric redshifts. In the framework of the implementation of the photometric redshift algorithm for the ESA {\it Euclid} Mission, we are exploring new avenues to improve current template-fitting methods. This paper focusses in particular on the prescription of the extinction of a source light by dust in the Milky Way. Since Galactic extinction strongly correlates with wavelength and photometry is commonly obtained through broad-band filters, the amount of absorption depends on the source intrinsic spectral energy distribution (SED), a point however neglected as the source SED is not known a-priori. A consequence of this dependence is that the observed $E_{\rm B-V}$ ($= A_{\rm B} - A_{\rm V}$) will in general be different from the $E_{\rm B-V}$ used to normalise the Galactic absorption law $k_{\rm \lambda}$ ($= A_{\rm \lambda} / E_{\rm B-V}$). Band-pass corrections are thus required to adequately renormalise the law for a given SED. In this work, we assess the band-pass corrections of a range of SEDs and find they vary by up to $20$\%. We have investigated how neglecting these corrections biases the calibration of dust into reddening map and how the scaling of the map depends of the sources used for its calibration. We derive dust-to-reddening scaling factors from the colour excesses of $z < 0.4$ SDSS red galaxies and show that band-pass corrections predict the observed differences. Extinction corrections are then estimated for a range of SEDs and a set of optical to near-infrared filters relevant to {\it Euclid} and upcoming cosmological ground-based surveys. For high extinction line-of-sights ($E_{\rm B-V} > 0.1$, $\sim8$\% of the {\it Euclid} Wide survey), the variations in corrections can be up to $0.1$~mag in the `bluer' optical filters ($ugr$) and up to $0.04$~mag in the near-infrared filters. We find that an inaccurate correction of Galactic extinction critically affects photometric redshift estimates. In particular, for high extinction lines of sights and $z < 0.5$, the bias (i.e.~the mean $\Delta z = z_{\rm phot} - z_{\rm real}$) exceeds $0.2\%(1+z)$, the precision required for weak-lensing analyses. Additional uncertainty on the parametrisation of the Milky Way extinction curve itself further reduces the photometric redshift precision. We propose a new prescription of Galactic absorption for template-fitting algorithms which takes into consideration the dependence of extinction with SED.} | {\it Photometric redshift precision for weak lensing studies} The new generation of wide sky surveys such as the ground-based Dark Energy Survey (DES; the Dark Energy Survey Collaboration 2016)\nocite{DES2016}, the Kilo Degree Survey (KiDS; de Jong et al.~2013)\nocite{deJong2013} and the Large Synoptic Survey Telescope (LSST; Ivezi\'c et al.~2008)\nocite{Ivezic2008} survey or the upcoming optical-to-near-infrared ESA {\it Euclid} mission will map the extragalactic sky in multiwavelength photometric bands for a large number of galaxies. One of the primary science goals of these surveys is to constrain the dark energy equation of state by means of powerful cosmological probes including weak gravitational lensing by large-scale structures (cosmic shear). The applications of weak lensing analyses for high precision cosmology demand both an accurate measurement of the source distortion and a statistical knowledge of their distance. Given the few $10^9$ galaxies expected to lie within the survey footprints, these studies will heavily rely on high quality photometric redshifts (photo-z hereafter). To measure the effect of dark energy with cosmic time, a number of works have adopted the weak lensing tomography approach \citep[e.g.~][]{Hu1999,Hoekstra2002} where the galaxies are binned in tomographic bins. The weak lensing signal is derived from the cross-correlation of the source distortions between redshift planes, thus providing information on the distribution of mass along the line of sight. The weak lensing tomography approach requires photometric redshifts of high precision to avoid tomographic bin overlaps and to allow an accurate determination of the mean redshift of sources in each bin. This leads to performance requirements on both the photometric redshift scatter and bias. The photometric redshift accuracy required to achieve the precisions on the cosmological parameters set by current cosmological surveys is however spectacularly challenging \citep{Bordoloi2010}. Various studies have for example determined that a requirement of the order of a few percent in the dark energy equation of state parameter $w$ would translate into a scatter $\sigma_{\rm z} < 0.05(1+z)$ and a bias $\langle z \rangle < 0.2\%(1+z)$ \citep[][and references therein]{Ma2006, Hearin2010}. The performance of photometric redshift techniques is continuously improving. Photometric redshifts were for example derived for sources in the CANDELS \citep{Grogin2011} GOODS-south field from the multi-wavelength $17$-band photometric catalogue of \citet{Guo2013}. \citet{Dahlen2013} combined the outputs of a number of widely used template-fitting codes (using for instance a simple median or a hierarchical Bayesian approach) and reached a scatter in their photo-z estimates of $\sigma_{\rm z} \sim 0.025(1+z)$ and a bias of $\langle z \rangle < 0.7\%(1+z)$. These results are encouraging, but more work is certainly needed to achieve the ambitious goals set by the new generation of cosmological surveys such as the {\it Euclid} mission. The all (or half) sky surveys mentioned earlier will also evidently lack the richness of multiwavelength data available in deep extragalactic surveys such as GOODS-south. In this paper, we focus on the minimisation of a source of photometric bias that originates from an often simplistic treatment of Galactic absorption. \vspace{4mm} {\noindent \it Galactic extinction maps} The interstellar dust of the Milky Way absorbs the UV-to-near-infrared light and contaminates observations of extragalactic objects. Galactic reddening maps are therefore used to correct the observed photometric measurements of distant sources. A number of works have derived reddening maps (commonly referred to as $E_{\rm B-V}$ maps) using the colour excesses of stars or standard cosmological sources with well-known intrinsic spectral energy distribution, in other words with low intrinsic colour dispersion such as quasars, luminous red galaxies etc. These studies assume that the dust properties of the Galactic interstellar medium are homogenous across the Milky Way and linearly rescale either HI column density maps (Burstein \& Heiles 1978)\nocite{Burstein1978} or dust column density maps to match the observed extinction of their calibration sources. One of the most widely used Galactic reddening maps was derived by Schlegel, Finkbeiner \& Davis (1998; SFD98 hereafter)\nocite{Schlegel1998} from a dust thermal emission map in the far-infrared, the $100\mu$m DIRBE/{\it IRAS}-combined, point-source removed, $6.1\arcmin$ resolution map. The dust column density $D$ was linearly rescaled to a reddening estimate $E_{\rm B-V}$ by a constant calibration coefficient $p$ following $E_{\rm B-V} = pD$. The scaling factor $p$ was estimated using the colour excess estimates of $\sim 100$, $z < 0.05$ brightest cluster galaxies (BCG) and $\sim 400$ field elliptical galaxies. SFD98 adopted the Milky Way absorption law functional form of \citet{O'Donnell1994} in the visible and Cardelli, Clayton \& Mathis (1989)\nocite{Cardelli1989} in the ultraviolet and infrared. More recent studies set out to recalibrate the DIRBE/{\it IRAS}-derived dust map of SFD98 using the colour excesses of stars or standard cosmological sources from the Sloan Digital Sky Survey \citep[SDSS;][]{York2000}. \citet{Schlafly2010} and Schlafly \& Finkbeiner (2011)\nocite{Schlafly2011} used the colour excesses of $\sim 260,000$ stars and claimed that SFD98 overpredicted the Galactic reddening by up to $14$\%. Peek \& Graves 2010\nocite{Peek2010} used $\sim150,000$ passive galaxies in SDSS selected using upper limits on their $H\alpha$ and $[OII]$ equivalent widths to derive a new extinction map. They stated that SFD98 underestimated the reddening estimates at high latitudes up to $10-15$\%. \citet{Schlafly2014} derived a map of dust reddening for a large fraction of the northern hemisphere at Dec. $> 30^{\circ}$ from the photometry of half a billion stars from the Panoramic Survey Telescope and Rapid Response System 1 (Panstarrs, PS1). All these studies favoured the Milky Way extinction law from Fitzpatrick (1999; F99 hereafter)\nocite{Fitzpatrick1999}. M{\"o}rtsell et al.~(2013)\nocite{Mortsell2013} also attempted to evaluate the scaling factor $p$ from the colour excesses of a variety of extragalactic sources such as quasars, BCG and red galaxies while simultaneously refining F99. More recently, a higher resolution all-sky dust map was derived from a combination of the $5\arcmin$-resolution {\it Planck} data and the IRAS $100\mu$m data ({\it Planck} Collaboration XI 2014; P14 hereafter)\nocite{Planck2014}. P14 evaluated a dust-to-reddening conversion factor from the colour excesses of $\sim 53,000$ SDSS quasars. They adopted an F99 reddening curve and favoured a linear rescaling of their point-source-removed thermal dust radiance map $R$ (in units of W m$^{-2}$ sr$^{-1}$) over the dust optical depth at $353$~GHz ($\tau_{\rm 353}$) map to estimate reddening. They showed in particular that $R$ seems less affected by the cosmic infrared background anisotropies and that it correlates with $N_{\rm H}$ over a wider range of column density than $\tau_{\rm 353}$. The methodologies used to derive reddening maps evidently differ from one study to another, for instance in the adopted prescription of the extinction law of the Milky Way or in the initial dust maps used for the calibration etc.; it would therefore be difficult to pin down all possible origins of the discrepancies between the existing reddening maps. In this paper however, we specifically investigate an inaccuracy of these works introduced by the systematic neglected fact that the source photometry is mainly obtained through broad-band filters and that, in consequence, the Galactic extinction does depend on the intrinsic spectral energy distribution (SED) of the observed source. The paper is organised as follows. Section 2 summarises the range of reddening expected along the line of sight of the {\it Euclid} wide field survey. Section 3 presents a number of caveats introduced by the failure to take into consideration the impact of the dependence of Galactic extinction with source SED. We first introduce the band-pass corrections required to renormalise the Milky Way absorption law for a given SED. As a proof of concept, we test the SED dependence of dust-to-reddening scaling factors using luminous red galaxies at $z < 0.4$ as standard crayons. We then quantify the impact of the source SED on photometric corrections. We then assess in section 4 how inaccurate Galactic extinction corrections lead to bias in source photometric redshift estimates. We summarise our findings in section 5. Appendix A introduces a prescription of Galactic extinction as a possible improved recipe for template-fitting codes, in particular for the {\it Euclid} template-fitting code Phosphoros. Appendix B investigates the dependence of the Galactic extinction with the uncertainties on extinction law prescription, in particular the uncertainties on the characteristic total-to-selective extinction $R_{\rm V}$. Appendix C confronts our analysis to the textbook calibration work of SFD98. | We evaluate the impact of a source SED on Galactic extinction. The use of calibration sources of a specific SED with broad-band filters systematically bias the calibration of the Galactic absorption law and band-pass corrections are required to adequately renormalise the absorption law for a given SED. We assess the range of band-pass corrections that could be expected from a library of extragalactic sources derived from the {\rm LePhare} COSMOS template library and a grid of intrinsic dust and redshift. We use the Fitzpatrick (1999) Milky Way absorption law that was calibrated using the colour excesses of B5 stars and find that band-pass corrections required to renormalise a B5 star-calibrated law can be in the range $0.8 < {\rm bpc}_{\rm sed} < 1.2$; in other words, the product of the absorption law by the reddening along the line-of-sight (i.e.~the `$E_{\rm B-V}$' read on a reddening map) needs to be scaled up/down by up to $20$\% to take into consideration the SED of a source. One of the direct consequences of neglecting these band-pass corrections is that the scaling of a reddening map obtained from a dust map using the colour excesses of sources of a given SED will be SED-dependent. We illustrate this point by deriving the dust-to-reddening scaling factors for three different types of sources i.e.~three distinct SED. For this work, we use $z < 0.4$ luminous red galaxies from SDSS divided in three redshift bins ($[0.15-0.25]$, $[0.25-0.3]$ and $[0.3-0.35]$). The value of Galactic extinction along the line-of-sight of the sources is derived by matching their SDSS photometry with a reference LRG template reddened by a grid of extinction. We then linearly correlate the values of the radiance $R$ map produced by the {\it Planck} team (P14) in the direction of the sources with the estimate of reddening. This analysis is repeated for each studied LRG redshift bin. The differences in scaling factors for the three samples of LRG can be predicted by the band-pass corrections derived directly from the SED. We also recover the dust-to-reddening scaling factor that was derived by the {\it Planck} team (P14) using quasars at $0.7 < z < 1.7$ to convert their radiance $R$ map into the released `$E_{\rm B-V}$' reddening map. We then estimate the range of Galactic extinction corrections to be expected for an SED library of extragalactic sources within a number of optical-to-near infrared filters pertinent to on-going or upcoming cosmological surveys. We find that Galactic extinction corrections can vary by up to $0.1$~mag in the optical bands for lines of sight of medium-to-high extinction $pD > 0.1$ which represents about $\sim8$\% of the {\it Euclid} Wide survey planned footprint. Corrections can vary up to $0.02$~mag for lines of sight $pD = 0.02$, the extinction average value commonly encountered in the direction of the most well studied extragalactic surveys. The impact of a source SED for Galactic extinction in the near infrared wavelengths is less significant but still found up to $0.04$~mag in the $Y$-band for $pD = 0.1$. The introduction of biases in the estimation of Galactic extinction photometric corrections naturally result in biases in the estimates of source photometric redshifts that could be dramatic for future cosmological surveys, especially in high extinction lines of sight. Appendix A presents an example of prescription of the Galactic extinction in template-fitting codes used for the determination of photometric redshifts. It describes in particular the Galactic extinction recipe that will be implemented in the template-fitting code of the {\it Euclid} Mission, Phosphoros. Upcoming works will focus on illustrating the improvement on photometric redshifts induced by the use of this exact prescription of Galactic extinction. They will particularly concentrate on testing photometric redshift estimates for sources along lines of sight of high extinction, classically avoided by extragalactic wide-field surveys. Ultimately, we will be releasing a set of online routines and applets that will provide for a given line of sight and SED the estimate of Galactic extinction for a set of common filters\footnote[11]{In the meantime, please feel free to contact the authors for early access to the routines.}. Appendix B investigates the impact of uncertainties of the extinction law and in particular of its characteristic total-to-selective ratio $R_{\rm V}$ on Galactic extinction estimates. We show that although the uncertainties on $R_{\rm V}$ do not cause systematic biases, they are reducing the precision of photometric measurements and by extension of photometric redshifts. Appendix C confronts our analysis to SFD98 and raises again the impact of the adopted extinction law prescription for Galactic extinction estimates. In the context of template-fitting codes, the a-priori knowledge of the template SED used to model the observed photometry allows to properly estimate Galactic absorption. Such implementation is however not easily applicable for non-template-based photometric redshift recipes such as machine-learning architectures. Machine-learning algorithms learn how to combine galaxy photometric measurements \citep[and other galaxy features; see e.g.~][]{Hoyle2015} to estimate source distances using training samples of sources with known redshift. Unless such machine is trained with spectroscopically confirmed sources that cover a wide and `complete' range of Galactic reddening, it will not be able to learn how to take into account the impact of Galactic extinction on photo-z estimates correctly. A possible alternative could be to combine the strength of both template-fitting codes and machine-learning algorithms and use the first guess of the source type or SED from template-fitting to correct (to a first order at least) the source photometry from Galactic extinction. Complementarily, Masters et al.~(2015)\nocite{Masters2015} recently studied the colour-redshift distribution of galaxies based on a self-organising map (SOM) analysis with each cell of the SOM corresponding approximately to a characteristic source SED. One of their main science drivers was to investigate the regions of their SOM lacking spectroscopic redshifts and estimate the missing spectroscopy potentially required to calibrate photometric redshifts. The analysis was conducted in the COSMOS field and therefore with low Galactic extinction along the line-of-sight. It would be interesting to investigate how the Galactic extinction and its dependence with source SED would affect the colour-redshift distribution. Finally, the success of upcoming weak-lensing analyses does not only depend on the estimation of photometric redshifts of high precision. It also requires an accurate determination of the shape measurement of sources, more specifically a precise measure of their distortion, the so-called shear signal. In the light of the dependence of Galactic extinction with galaxy SED/colour, we also raise concern about the fact that the amount of Galactic extinction will also be sensitive to the colour gradients within the source itself and that it might bias the determination of source shape measurements. | 16 | 9 | 1609.08624 |
1609 | 1609.09308_arXiv.txt | {While the velocity fluctuations of supergranulation dominate the spectrum of solar convection at the solar surface, very little is known about the fluctuations in other physical quantities like temperature or density at supergranulation scale. Using SDO/HMI observations, we characterize the intensity contrast of solar supergranulation at the solar surface. We identify the positions of ${\sim}10^4$ outflow and inflow regions at supergranulation scales, from which we construct average flow maps and co-aligned intensity and magnetic field maps. In the average outflow center, the maximum intensity contrast is $(7.8\pm0.6)\times10^{-4}$ (there is no corresponding feature in the line-of-sight magnetic field). This corresponds to a temperature perturbation of about $1.1\pm0.1$~K, in agreement with previous studies. We discover an east-west anisotropy, with a slightly deeper intensity minimum east of the outflow center. The evolution is asymmetric in time: the intensity excess is larger 8~hours before the reference time (the time of maximum outflow), while it has almost disappeared 8~hours after the reference time. In the average inflow region, the intensity contrast mostly follows the magnetic field distribution, except for an east-west anisotropic component that dominates 8~hours before the reference time. We suggest that the east-west anisotropy in the intensity is related to the wave-like properties of supergranulation.} | Solar granulation is a manifestation of thermal convection; hot gas rises to the surface, cools, and sinks in the intergranular lanes. Granular structure is easily observed in white-light intensity images, even though the measured contrast is reduced by, e.g., stray light \citep[e.g.,][]{sanchez_2000}. After deconvolving the intensity images, the contrast is roughly 15\% RMS in the red \citep{wedemeyer_2009}. The thermal signature of the larger-scale supergranulation \citep[see][for an extensive review]{rieutord_2010}, on the other hand, is hard to measure because it is small compared to the granulation contrast and competes with the brightness increase due to the network magnetic field, which surrounds the supergranules \citep[e.g.,][]{liu_1974,foukal_1984}. The latter effect results from reduced opacity in magnetic regions \citep[e.g.,][]{spruit_1976}. Using observations from the ground-based PSPT, \citet{goldbaum_2009} and \citet{rast_2003} found a brightness excess of ${\sim}0.1\%$ in supergranules, corresponding to a temperature perturbation of ${\sim}1~$K. To obtain this result, the authors carefully removed magnetic pixels using \ion{Ca}{II}~K images, conducted ensemble averaging over thousands of supergranules and applied azimuthal averaging, thus losing spatial and temporal information. \citet{meunier_2007a} measured a similar (0.8$-$2.8~K) temperature excess using space-based SOHO/MDI intensity images and magnetograms (high-resolution mode) in combination with a magnetic field exclusion that takes into account neighboring pixels, but with a similar lack of spatial information. Here we extent the previous studies on the supergranular brightness excess: How does the convective intensity peak evolve? Does the intensity contrast show an east-west anisotropy, as the magnetic field \citep{langfellner_2015a} or wave travel times \citep{degrave_2015}? To tackle these questions, we make use of high-quality data from the Helioseismic and Magnetic Imager (HMI) \citep{schou_2012} onboard the SDO spacecraft. | Using Planck's law, we have $\Delta I/I_0 \approx 4\Delta T/T_0$, where $T_0=5777~$K corresponds to HMI's iron absorption line (6173~\AA). With $\Delta I/I_0 = (7.8\pm0.6)\times10^{-4}$ at the center of the outflow, this gives $\Delta T \approx 1.1\pm0.1$~K. This result is consistent with the value ${\sim} 1$~K obtained by \citet{goldbaum_2009} (no error bar provided) and the range 0.8$-$2.8~K measured by \citet{meunier_2007a}. The east-west anisotropy of the intensity contrast that we detect at the equator consists of two components: (i) an anisotropy in the network that has the same spatial pattern as the already known magnetic field anisotropy \citep{langfellner_2015a} and (ii) an anisotropy that is distinct from the magnetic field anisotropy, presumably of convective origin. Since the opacity is reduced in magnetic regions and the brightness is increased, the anisotropy of type (i) can be regarded as an independent confirmation of the magnetic field anisotropy, using a different observable. The east-west anisotropies of the intensity and magnetic field signals at the equator are most likely connected to the travelling-wave properties of supergranulation and the superrotation of the pattern \citep{gizon_2003,schou_2003}. This connection remains to be specified by studying the evolution of the intensity and the magnetic field signals over longer times (several days). The decrease of the intensity peak from the time of maximum divergence can be interpreted as the termination of the driving of the supergranular outflow. The temperature excess may imply a pressure excess in the supergranule center that would accelerate the plasma horizontally. Due to inertia, the outflow does not stop immediately, but continues and weakens over time. After the convective brightness excess has vanished, the intensity contrast reflects the magnetic field distribution. The broadening of the magnetic field dip in the outflow region and the accumulation of magnetic field in the network beyond the time of maximum outflow, are qualitatively consistent with the assumption that supergranular flows advect the magnetic field \citep{orozco_2012}. Alternative models have been proposed that treat supergranulation not as a convective phenomenon but as a pattern resulting from the collective (non-linear) interaction of granules \citep{rieutord_2000,rast_2003a} or magnetic elements \citep{crouch_2007}. For example, the model by \citeauthor{crouch_2007} implies that the network magnetic field builds up before the supergranular inflows, contrary to what our measurements indicate. These alternative models do not make clear predictions for the intensity contrast, but our findings may impose additional constraints that could be incorporated in the future. | 16 | 9 | 1609.09308 |
1609 | 1609.02706_arXiv.txt | A giant planet embedded in a protoplanetary disk creates a gap. This process is important for both theory and observations. Using results of a survey for a wide parameter range with two-dimensional hydrodynamic simulations, we constructed an empirical formula for the gap structure (i.e., the radial surface density distribution), which can reproduce the gap width and depth obtained by two-dimensional simulations. This formula enables us to judge whether an observed gap is likely to be caused by an embedded planet or not. The propagation of waves launched by the planet is closely connected to the gap structure. It makes the gap wider and shallower as compared with the case where an instantaneous wave damping is assumed. The hydrodynamic simulations shows that the waves do not decay immediately at the launching point of waves, even when the planet is as massive as Jupiter. Based on the results of hydrodynamic simulations, we also obtained an empirical model of wave propagation and damping for the cases of deep gaps. The one-dimensional gap model with our wave propagation model is able to well reproduce the gap structures in hydrodynamic simulations. In the case of a Jupiter-mass planet, we also found that the waves with smaller wavenumber (e.g., $m=2$) are excited and transport the angular momentum to the location far away from the planet. The wave with $m=2$ is closely related with a secondary wave launched by the site opposite from the planet. | \label{sec:intro} During the past decade, more than a thousand extrasolar planets have been discovered. A survey of extrasolar planets has revealed the diversity of giant planets outside the solar system \citepeg{Burke2014}. Giant planets are born in protoplanetary disks, due to core accretion \citepeg{Mizuno1980,Kanagawa_Fujimoto2013} or to the gravitational instability of the gaseous disk \citepeg{Cameron1978,Zhu2012b}. Once formed, they undergo orbital migration, and gas accretion results in growth their mass. The diversity of extrasolar planets is closely connected to such processes \citepeg{Mordasini_Alibert_Benz_Klahr_henning2012,Ida_Lin_Nagasawa2013}. Planet forming regions in protoplanetary disks can now be directly imaged, such as by the Atacama Long Millimeter/Submillimeter Array (ALMA) and by eight-metre class optical and/or near-infrared telescopes. High-resolution images have revealed the presence of complex morphological structures in disks; these structures include spirals \citepeg{Muto2012,Grady2013, Christiaens_Casassus_Perez_VanderPlas_Menard2014,Benisty2015,Currie2015,Akiyama2016} and gaps \citepeg{Osorio2014,ALMA_HLTau2015,Momose2015,Akiyama2015,Nomura_etal2016, HLTau_HCO2016,Tsukagoshi2016}. Direct images are now finding possible signatures of planets being formed in disks \citepeg{Sallum2015}. To understand the origins of the disk structures and their possible connection to the formation of planets, it is important to construct appropriate quantitative models. A planet interacts gravitationally with the gas in the surrounding disk, and as a result, the planet excites density waves (spirals). If the planet is sufficiently massive, it also creates gap structures \citepeg{Lin_Papaloizou1979,Goldreich_Tremaine1980,Artymowicz_Lubow1994,Kley1999,Crida_Morbidelli_Masset2006}. Recently there have been a number of studies on the quantitative relationship between the mass of a planet and the gap structure (i.e., depth and width) \citep{Duffell_MacFadyen2013,Fung_Shi_Chiang2014,Kanagawa2015b,Duffell_Chiang2015,Fung_Chiang2016,Kanagawa2016a}, and the application to actual observations has been discussed \citep{Kanagawa2015b,Momose2015,Marel2016,Nomura_etal2016,Tsukagoshi2016}. However, there is still room for improvement of the current models. \cite{Kanagawa2016a} presented an empirical formula for the gap width: they defined it to be the location at which the gap surface density is depleted by $50\%$ from its original value. The overall structure of the gap created by the planet should be further investigated. The gap structure is closely associated with the damping of spiral density waves excited by a planet, because the planet exchanges the angular momentum to the disk via the waves \citepeg{Takeuchi_Miyama_Lin1996,Goodman_Rafikov2001,Rafikov2002a,Dong2011}. \cite{Duffell2015} constructed analytical models of the gap structure by using the wave damping model of \cite{Goodman_Rafikov2001}, which is particularly useful in the case of shallow gaps (created by relatively low-mass planets). However, the wave propagation in the case where the deep gap is formed by a high-mass planet is still poorly understood. The propagation of waves induced by the high-mass planet would be qualitatively different from that in the case of the low-mass planet, as implied by the parametrized study by \cite{Kanagawa2015a} (hereafter, K15). In this paper, we extend our previous model of the gap width and depth, and present a complete model of the gap shape; this is based on a number of two-dimensional long-term hydrodynamic simulations. We also provide a model of the wave propagation indicated by the hydrodynamic simulations, when the deep gap is formed. In Section~\ref{sec:basic_eq}, we briefly summarize the setup of the numerical simulations we performed. In Section~\ref{sec:results}, we obtain an empirical formula for the gap structure, showing the results of hydrodynamic simulations. From this empirical formula of the gap structure, we can build an empirical model of propagation of the density waves. In Section~\ref{sec:1dmodel}, we obtain the empirical model of the wave propagation. Adopting this model of the wave propagation, we provide a semianalytical radial one-dimensional model of the gap structure which is able to reproduce the gap structures given by the two-dimensional simulations. \RED{ In Section~\ref{sec:discussion}, we discuss an observational application of our formula. We also discuss how the wave excitation and propagation are changed along with an increase of a planet mass, showing the results of the two-dimensional hydrodynamic simulations. A discussion about hydrodynamical instabilities such as the Rayleigh instability and Rossby wave instability is included in this section. } Section~\ref{sec:summary} contains a summary and discussion. | \label{sec:basic_eq} \cite{Kanagawa2016a} performed a number of the two-dimensional simulations over a wide range of parameter space (planet mass, disk scale height and viscosity), and derived a relationship between the gap width and the planet mass. In this paper, we extend the model by further analyzing of the numerical simulations. In this section, we briefly review the setup of hydrodynamic simulations. We numerically calculated the gap formation processes in a protoplanetary disk in the presence of a planet. We assumed a geometrically thin and non-self-gravitating disk. We adopted a two-dimensional cylindrical coordinate system ($R,\phi$), and the origin was located at the position of the central star. We adopted a simple locally isothermal equation of state, and the distribution of the temperature is independent of time. The mass of the central star is also constant during the simulations. Using FARGO \citep{Masset2000}, which is widely used in the study of disk--planet interaction \citepeg{Crida_Morbidelli2007,Baruteau_Meru_Paardekooper2011,Zhu2011}, we solved the equations of continuity and motion of disk gas. In this paper, we adopt $\alpha$-prescription of \cite{Shakura_Sunyaev1973}, and then the kinetic viscosity is written as $\nu=\alpha h^2 \Omegak^{-1}$, where $h$ and $\Omegak$ are the disk scale height and the Keplerian angular velocity, respectively. The initial disk structures, boundary conditions and other computational setup (e.g., computational domain and resolutions) are as the same as those in \cite{Kanagawa2016a}. To complete the parameter survey, we additionally ran 6 simulations with $\hp/\rp = 2/15$, where $\hp$ is the disk scale height at $\rp$ which is the orbital radius of the planet; ($\mpl/\mstar$,$\alpha$) = ($10^{-3},4\times 10^{-3}$), ($2\times 10^{-3},4\times 10^{-3}$), ($5\times 10^{-4},10^{-3}$),($10^{-3},10^{-3}$), ($5\times 10^{-4},6.4\times 10^{-4}$), and ($10^{-3},6.4\times 10^{-4}$), and we also ran 2 simulations with small planet mass of $\mpl/\mstar=5\times 10^{-5}$, where $\mpl$ and $\mstar$ are the masses of the planet and the central star, respectively; $(\alpha$, $\hp/\rp$) = ($10^{-3}$, $1/20$) and ($10^{-3}$,$1/25$) (totally 34 runs). \RED{The survey covers the range of the planet mass as $5\times 10^{-5} < \mpl/\mstar < 2\times 10^{-3}$, the range of the disk aspect ratio as $1/30 < \hp/\rp < 2/15$, and the range of the disk viscosity as $10^{-4}<\alpha < 10^{-2}$. The parameter set of each run is described in \cite{Kanagawa2016a}, except above additional runs. } The gap is opened by the disk--planet interaction and is closed by the viscous diffusion. In steady state, the viscous angular momentum flux is balanced by the planetary torque. The timescale of the gap opening would be scaled by the viscous timescale \citepeg{Lynden-Bell_Pringle1974}. As an empirical formula, the gap width ($\gapwidth$), which is defined as a width of a region where the surface density is smaller than $0.5 \Sigma_0$ in steady state, where $\Sigma_0$ is the surface density outside the gap, is obtained as follows \citep{Kanagawa2016a}: \begin{eqnarray} \frac{\gapwidth}{\rp} &= 0.41 K'^{1/4}, \label{eq:gapwidth_0.5} \end{eqnarray} where \begin{eqnarray} K'&=\left(\frac{\mpl}{\mstar} \right)^2 \left( \frac{\hp}{\rp} \right)^{-3} \alpha^{-1}. \label{eq:kp} \end{eqnarray} The timescale of the gap opening ($t_{\rm vis}$) is \begin{eqnarray} t_{\rm vis}= \left(\frac{\gapwidth}{2\rp} \right)^2 \left( \frac{\hp}{\rp} \right)^{-2} \alpha^{-1} \Omega_p^{-1}. \label{eq:vistime_gap} \end{eqnarray} Using equation~(\ref{eq:gapwidth_0.5}), this timescale is obtained as \begin{eqnarray} t_{\rm vis} &= 0.24 \bracketfunc{\mpl/\mstar}{10^{-3}} \bracketfunc{\hp/\rp}{0.05}^{-7/2} \bracketfunc{\alpha}{10^{-3}}^{-3/2} \nonumber \\ & \qquad \qquad \times \bracketfunc{\mstar}{1M_{\odot}}^{-1/2}\bracketfunc{\rp}{10\rm{AU}}^{3/2} \mbox{Myr}. \label{eq:gap_vistime} \end{eqnarray} To obtain the steady state, we have to calculate the evolution until $t\sim t_{\rm vis}$ (see, Appendix~\ref{sec:timeevo_various}). Note that for nominal parameters, this timescale is usually shorter than or comparable with the migration timescale of the type II ($\sim R_p^2/\nu $) or disk lifetime ($\sim 1$Myr). Hence, a full-width gap can be observed. \label{sec:discussion} \subsection{Observational applications} \label{subsec:obs_applications} \subsubsection{Constraint from the gap radial structure} \label{sss:obs_width_relation} \begin{figure} \begin{center} \resizebox{0.49\textwidth}{!}{\includegraphics{gapwidths_ratios.eps}} \caption{ The ratio of the gap width defined by $\sigmaedge=0.1\Sigma_0$ to that defined by $\sigmaedge=0.5\Sigma_0$ (triangles) and $\sigmaedge=0.3\Sigma_0$ and $0.5\Sigma_0$ (crosses). The horizontal dotted lines indicate the values predicted by equation~(\ref{eq:ratio_widths}). \label{fig:gapwidth_ratio} } \end{center} \end{figure} Here we would like to discuss observational applications of our empirical formula of the gap structure of equation~(\ref{eq:gap}). Recently, many gap structures have been discovered by ALMA; these have been found not only in dust \citepeg{ALMA_HLTau2015} but also in gas \citepeg{HLTau_HCO2016}. Such gap structures in protoplanetary disks can be created by dust growth \citep{Zhang_Blake_Bergin2015}, sintering \citep{Okuzumi_Momose_Sirono_Kobayashi_Tanaka2016}, the effects of MRI \citep{Flock_Ruge_Dzyurkevich_Henning_Klahr_Wolf2015}, secular gravitational instability due to gas--dust friction \citep{Takahashi_Inutsuka2016a}, and disk--planet interaction. It is difficult to distinguish the origin from observations. However, using the relationship of the gap radial structure that is shown above, we may be able to assess whether the observed gap is due to the disk--planet interaction. Using equation~(\ref{eq:gapwidth_varsedge}), we can determine the ratio of the gap width for any two arbitrary surface densities $\Sigma_{\rm th,a}$ and $\Sigma_{\rm th,b}$ as \begin{eqnarray} \frac{\Delta_{\rm gap}(\Sigma_{\rm th,a})}{\Delta_{\rm gap}(\Sigma_{\rm th,b})} = \frac{\Sigma_{\rm th,a}/{\Sigma_0}+0.32}{\Sigma_{\rm th,b}/{\Sigma_0}+0.32}. \label{eq:ratio_widths} \end{eqnarray} As can be seen from the above equation, the ratio of the gap widths depends only on the surface densities defined by the widths. In Figure~\ref{fig:gapwidth_ratio}, we show two ratios of gap widths, defined by $\sigmaedge=0.1\Sigma_0$ and $0.5 \Sigma_0$, and by $\sigmaedge=0.3\Sigma_0$ and $0.5 \Sigma_0$. Using equation~(\ref{fig:gapwidth_ratio}), we obtain that the ratios are $0.51$ and $0.76$, respectively. As can be seen in the figure, the ratios obtained from the simulations are reasonably consistent with those obtained with equation~(\ref{eq:ratio_widths}). \subsubsection{Gap depth--width relation} \label{sss:obs_depth-width_relation} As shown in \cite{Kanagawa2016a}, there is a relationship between the depth and width of a gap. Combining equations~(\ref{eq:gapwidth_varsedge}) and (\ref{eq:smin}), we can derive this relationship in terms of the locations of $\sigmaedge$, as follows: \begin{eqnarray} &\frac{\Delta_{\rm gap} \left(\sigmaedge /\Sigma_0 \right)}{R_p} \left(\frac{\sigmaminp}{\Sigma_0 - \sigmaminp} \right)^{1/4} \left(\frac{\hp}{\rp}\right)^{-1/2} \nonumber\\ &\qquad \qquad \qquad \qquad \qquad \qquad \qquad = 1.16\frac{\sigmaedge}{\Sigma_0} + 0.35. \label{eq:rel_depth_width} \end{eqnarray} If the gap is created by a planet, its width, as measured by the above-surface density, should satisfy equation~(\ref{eq:rel_depth_width}). When the gap structure is completely resolved and the disk aspect ratio is precisely estimated, we can use the gap widths measured by the different surface density at the gap edge and equation~(\ref{eq:rel_depth_width}) to strictly judge whether the gap was created by the planet. \subsubsection{Applicability of the model} \label{sss:obs_caveats} We now discuss some considerations when equations~(\ref{eq:ratio_widths}) and (\ref{eq:rel_depth_width}) are applied to observations, even if the observed gap is sufficiently resolved. First, we must consider that the distribution of dust particles may be different from that of a gas when the relations are applied to an observation of dust thermal emissions. Because of dust filtration \citepeg{Zhu2012,Dipierro_Price_Laibe_Hirsh_Cerioli_Lodato2015,Picogna_Kley2015,Rosotti_Juhasz_Booth_Clarke2016}, a gap in dust can be deeper and wider than in gas if the size of the dust particles (or the Stokes number of particles) is relatively large. Hence, equations~(\ref{eq:ratio_widths}) and (\ref{eq:rel_depth_width}) should be used for observations of disk gas. Alternatively, the dust--gas coupling depends on the gas surface density, as well as on the size of the dust particles. If the gas density is sufficiently large, the gas and dust particles will be well mixed. In this case, equations~(\ref{eq:ratio_widths}) and (\ref{eq:rel_depth_width}) would provide a good estimate. Second, we assume that the gap structure is in steady state. As shown in Appendix~\ref{sec:timeevo_various}, the timescale required for the gap to be in steady state can be roughly estimated as $t_{\rm vis} \sim 0.1$ Myr. If an observed gap is younger than this timescale, the gap will be narrower than it would be in steady state. Such young gaps cannot be estimated by equations~(\ref{eq:ratio_widths}) and (\ref{eq:rel_depth_width}). We note that equations~(\ref{eq:ratio_widths}) and (\ref{eq:rel_depth_width}) should be used for relatively deep gaps. As shown in Figure~\ref{fig:gapwidths}, for a shallow gap ($K'<1$), the actual gap width measured for a larger surface density is slightly wider than that estimated by equation~(\ref{eq:rel_depth_width}). Because of this, in this case, equation~(\ref{eq:rel_depth_width}) would estimate the disk scale height to be about $1.5$ times the actual value. Moreover, as also discussed in \cite{Kanagawa2016a}, the observational uncertainties of the orbital radius of the planet and aspect ratio should be taken into account. Finally, we briefly make comments about our assumption of two-dimensional disks and spatially constant kinematic viscosity $\nu$. \cite{Fung_Chiang2016} performed three-dimensional simulations of disk--planet interaction in the case when a deep gap is induced by a planet. The gap profile (i.e., the depth and the width) is not very different from the results of two-dimensional calculations. Hence, our models may be valid even when three-dimensional effects are properly taken into account. The assumption of constant viscosity may not be always satisfied. \cite{Zhu_Stone_Rafikov2013} performed ideal MHD simulations of the interaction between a low-mass planet and a disk and showed that the effective viscosity $\alpha$ within the gap is approximately twice as large as that outside the gap region. Therefore, the gap may be shallower than our model. However, since the dependence of $\alpha$ on the gap depth is not very strong (see, equation~\ref{eq:smin}), this effect may not significantly influence our results. \subsection{Excitation and propagation of waves with deep gaps} \label{subsec:waveprop} \subsubsection{Planetary torque and angular momentum deposition in the disk} \label{subsec:wave_2d} In Section~\ref{sec:1dmodel}, we present the empirical model of wave propagation and obtain the semianalytical one-dimentional model of gap structure using this wave propagation model. \RED{ In the previous studies \citepeg{Goodman_Rafikov2001}, when the planet mass is sufficiently large as $\mpl> M_1 (\sim 1 \times 10^{-4}\mstar \mbox{ if }\hp/\rp = 1/20$), the angular momentum flux of the wave decreases quickly ($\fjwave \propto |R-\rp|^{-5/4}$) from the location near the launching point. However, our model of the wave propagation (equations~\ref{eq:deposited_torque},\ref{eq:xdep}, and \ref{eq:wdep}) implies that the wave propagation is different from that expected by the model of \cite{Goodman_Rafikov2001}, when the planet is large. } For a better understanding of wave propagation with a deep gap, we discuss how the wave propagation properties change as we increase the planet mass. \begin{figure*} \begin{center} \resizebox{0.98\textwidth}{!}{\includegraphics{integtorque_amfwave_a1e-3_h0.05_qvar.eps}} \caption{ ({\textit Top}) Cumulative planetary torque ($\tp$, solid line), angular momentum flux due to the waves ($\fjwave$, chain line), and cumulative deposited torque ($\tdeposit$, dashed line) for $\mpl/\mstar=10^{-4}$ (left), $5\times 10^{-4}$ (middle), and $10^{-3}$ (right). The disk aspect ratio and viscosity were set to $1/20$ and $10^{-3}$, respectively. ({\textit Bottom}) Azimuthal averaged surface density distribution in each case. \label{fig:torque_waves_in_simulation} } \end{center} \end{figure*} \RED{ In Figure~\ref{fig:torque_waves_in_simulation}, we show the cumulative planetary torque $\tp$, the angular momentum flux due to the waves $\fjwave$, and the cumulative deposited torque $\tdeposit$ given by two-dimensional hydrodynamic simulations, for a small planet ($\mpl/\mstar=10^{-4}$), a planet with moderate mass ($\mpl/\mstar = 5\times 10^{-4}$), and a giant planet ($\mpl/\mstar = 10^{-3}$). The disk aspect ratio and the viscosity are $1/20$ and $10^{-3}$, respectively. In the case of $\mpl/\mstar=10^{-4}$ (left panel of Figure~\ref{fig:torque_waves_in_simulation}), the wave excitation is in the linear regime. In this case, $\tp$ increases within the gap bottom region of $\rhosurf \simeq \sigmamin$, and it is saturated to be constant value outside the gap. The angular momentum flux of the waves $\fjwave$ also increases with the cumulative planetary torque when $R/\rp \lesssim 1.1$. For $R/\rp>1.1$, $\fjwave$ decreases while $\tp$ does not change, which indicates that the waves are damping due to the shock and deposit the angular momentum to the disk. The angular momentum flux of the waves $\fjwave$ is approximately proportional to $|R-\rp|^{-5/4}$ and quickly decreases within the density gap ($R/\rp \lesssim 1.3$). With decreasing $\fjwave$, the cumulative deposited torque $\tdeposit$ increases. Because $\fjwave=0$ outside of the gap, $\tdeposit$ converges to $\tp$ at the location apart from the planet. When $\mpl/\mstar=5\times10^{-4}$ (the middle panel in Figure~\ref{fig:torque_waves_in_simulation}), the evolution of the waves should be in the nonlinear regime. In this case, the cumulative planetary torque increases not only at the gap bottom but also at the gap edge of $\rhosurf \sim \Sigma_0$. The angular momentum flux of the waves is almost equal to $\tp$ when $R/\rp < 1.15$. For $R/\rp>1.15$, $\fjwave$ deviates from $\tp$ and decreases due to the shock. As compared with the case of $\mpl/\mstar=10^{-4}$, the decrease of $\fjwave$ is much slow, which indicates the angular momentum deposition of the waves is less effective as the planet mass increases. Outside of the gap, the behavior of $\tp$ is significantly different from that in the case of $\mpl/\mstar=10^{-4}$. Around $R/\rp = 1.5$, $\tp$ decreases and then it increases again around $R/\rp=1.8$, which indicates that the planet interacts with the gas at such the large distance from the planet. The angular momentum flux of the waves $\fjwave$ also changes along with $\tp$. The cumulative deposited torque $\tdeposit$ slowly increases in this region. In the right panel of Figure~\ref{fig:torque_waves_in_simulation}, the planet mass is $\mpl/\mstar=10^{-3}$, and the excitation of waves is highly nonlinear. The behaviors of $\tp$, $\fjwave$, and $\tdeposit$ are similar to these in the case of $\mpl/\mstar=5\times 10^{-4}$, but the decrease of $\fjwave$ is further slow, and then the increase of $\tdeposit$ is also very slow. } \subsubsection{Excitation and damping of waves in low-m modes} \label{subsec:wave_2d} To further examine the wave propagation when a deep gap is present, we investigate the wave resonances associated with small wavenumber using the results of two-dimensional simulations. We consider the Fourier components for $\rhosurf$, $\vrad$, $\vphi$, and $\Psi$ which are given by \begin{eqnarray} f_m(R,m)&=\frac{1}{2\pi} \int^{2\pi}_{0} f(R,\phi) \exp\left( -im\phi \right) d\phi, \label{eq:fourier_trans} \end{eqnarray} where $f$ is $\rhosurf$, $\vrad$, $\vphi$, or $\Psi$. For convenience, we will let the subscript $m$ indicate the $m$-th Fourier component. The torque density exerted on the $m$-th resonance is given by \citep{Goldreich_Tremaine1980} \begin{eqnarray} \left( \frac{d\tp}{dR} \right)_m &=4\pi R \Re(\Psi_m) \Im(\rhosurf_m), \label{eq:tqdens_m} \end{eqnarray} and the integrated torque exerted on the $m$-th resonance is given by \begin{eqnarray} \tp{}_{,m}&=\int^{R}_{\rp} \left( \frac{d\tp}{dR} \right)_{m} dR=\int^{R}_{\rp}4\pi R \Re(\Psi_m) \Im(\rhosurf_m) dR. \label{eq:int_torque_m} \end{eqnarray} The angular momentum flux doe to the $m$-th mode of the wave is \begin{eqnarray} &\fjwavem = \qquad \qquad \qquad \qquad \qquad \qquad \quad \qquad \qquad \qquad \nonumber\\ &\qquad 4\pi R^2 \left[ \Re( \rhosurf \vrad{}_{,m}) \Re(\delta \vphi{}_{,m}) + \Im( \rhosurf \vrad{}_{,m}) \Im(\delta \vphi{}_{,m}) \right]. \label{eq:amfm} \end{eqnarray} The cumulative torque deposited by the $m$-th mode $\tdeposit{}_{,m}$ can be written as \begin{eqnarray} \tdeposit{}_{,m} &= \tp{}_{,m} - \fjwavem. \end{eqnarray} \begin{figure*} \begin{center} \resizebox{0.98\textwidth}{!}{\includegraphics{smallmode_integtorque_amwave_a1e-3_h0.05_qvar.eps}} \caption{ Contributions of the cumulative planetary torque (top) and the cumulative deposited torque (bottom) from azimuthal modes in the same cases as Figure~\ref{fig:torque_waves_in_simulation}. The dashed, solid, dotted and dot-dashed double-dot-dashed lines denotes the contributions from $m=1$, $2$,$3$, $4$, and a sum of $m>4$, respectively. Only when $\mpl/\mstar=10^{-4}$, we only plot the sum of $m\leq 4$ (gray solid line) and $m>4$, because the contributions from the lower azimuthal modes are very small. \label{fig:smallmode} } \end{center} \end{figure*} In Figure~\ref{fig:smallmode}, we illustrate the cumulative torque exerted by the planet and that deposited on the disk for low-m modes. The parameters for the disk and the planet are the same as those in Figure~\ref{fig:torque_waves_in_simulation}. We plot the contributions from $m=1$ -- $4$ individually and the sum of contributions from modes larger than $4$. In the case of $\mpl/\mstar=10^{-4}$ (left panel of the figure), the planetary torque is mainly exerted by the resonances with $m>4$ since the wave excitation is in the linear regime \citep{Goldreich_Tremaine1980}. The cumulative deposited torques are also dominated by the contribution from $m>4$ waves. \RED{ In the cases where the plant mass is large as in the middle and right panels of Figure~\ref{fig:smallmode}, the contributions from the small $m$ resonances are significant. For instance, in the case of $\mpl/\mstar=10^{-3}$ (the right panel of the figure), the contributions of torque from the resonances of $1\leq m \leq 4$ are comparable to the sum of the contributions from $m>4$ resonances. The contributions from $m=3,4$ resonances are excited around the gap edge ($R/\rp \simeq 1.2$). The contributions from $m=1,2$ resonances are significantly excited in the outside of the gap. The variation of $\tp$ at large distances pointed out in previous subsection (and shown in Figure~\ref{fig:avgdens_comp_sims_models}) is originated from the contributions from $m=1,2$ resonances. Note that the contribution of $m=1$ resonance is related with the indirect term of the gravitational potential (which is connected with the gravitational interaction between the planet and the central star), which just oscillates around zero at larger distances from the planet, and the contribution of $m=1$ does not significantly influence the angular momentum flux of the waves. As in the planetary torque, the contributions of the cumulative deposited torque from the small $m$ modes become larger as the planet mass increases. When $m=2$, in particular, even at radii very far away from the planet ($R/\rp \sim 2$), the cumulative deposited torque slowly increases, in the case of $\mpl/\mstar=10^{-3}$. These behaviors of $\tp$ and $\tdeposit$ indicate that the waves with small $m$ modes (except $m=1$) significantly excites and carry the angular momentum to the large radii from the planet, with the increase of the planet mass. As the gap becomes deeper and wider, the contributions on the wave excitation from large $m$ resonances (near the planet) becomes smaller because the surface density at the resonance decreases. In consequence, the wave excitation on smaller $m$ resonances becomes significant, as discussed by \cite{Juhasz_Benisty_Pohl_Dullemond_Dominik_Paardekooper2015}. Moreover, \cite{Lee2016} has shown that the excitation of waves with small azimuthal wavenumber can be enhanced due to nonlinear effects. The significantly excitation of waves with small $m$ would be explained by the gap opening and the nonlinear effects. We should note that the wave with $m=2$ deposit the angular momentum outside the gap with $\rhosurf > \Sigma_0$ ($R/\rp > 1.8$). This indicates that wider gap is formed by the deposition of the angular momentum of $m=2$ wave, whereas opening such the wide gap requires very long time that is comparable with the disk life time (see equation~\ref{eq:vistime_gap}). } \begin{figure*} \begin{center} \resizebox{0.98\textwidth}{!}{\includegraphics{decomp_densmap_a1e-3_h0.05_q1e-3.eps}} \caption{ Two-dimensional surface density distribution in the case of $M_p/M_{\ast}=10^{-3}$, $h/R_p=1/20$, and $\alpha=10^{-3}$. Shown are the surface densities produced by all modes (left top), the $m>2$ modes (right top), the $m=1$ mode (left bottom), and the $m=2$ mode (right bottom). \label{fig:decomp_densmap} } \end{center} \end{figure*} The development of the $m=1, 2$ contributions significantly changes the morphology of the waves in addition to the transport of angular momentum. In Figure~\ref{fig:decomp_densmap}, we show the surface density produced by the $m=1,2$ components and the sum of the $m>2$ components, for the same case as shown in Figure~\ref{fig:smallmode}. The surface density produced by the $m$-th component is \begin{eqnarray} \Sigma (m,R,\phi) &= 2\pi \Sigma_m(R,\phi) \exp\left(im\phi \right), \label{eq:sigma_m} \end{eqnarray} where $\Sigma_m$ is the Fourier component of the surface density. The waves that are far from the planet are created by the contributions from the $m=1$ and $m=2$ resonances. In particular, the secondary wave that is launched from the site opposite the planet is mainly composed of the $m=1$ and $m=2$ components. Because the contribution from the $m=2$ component is strong, the secondary wave originates at a site opposite from the planet. \cite{Fung_Dong2015} have shown that the azimuthal separation of the primary and secondary waves depends on the planet mass. For a large planet, as shown in Figure~\ref{fig:decomp_densmap}, this separation is close to $180^{\circ}$, which is consistent with our results. For a smaller planet, the separation is smaller than $180^{\circ}$. In this case, the contributions from other lower modes (e.g., $m=3,4$), are not much smaller than these from $m=1,2$ (see Figure~\ref{fig:smallmode}). Because of these contributions, the launching point of the secondary wave would move toward the planet. It is worth noting that the secondary wave interacts with the planet. As seen in Figure~\ref{fig:decomp_densmap}, the secondary wave passes through the location at which $\phi=\phi_p$ around $R/\rp=1.4$. Near this location, the planetary torque due to the $m=2$ resonance increases and then decreases, as shown in the top panel of Figure~\ref{fig:smallmode}; this is a result of the interaction between the planet and the secondary wave. The cumulative deposited torque of $m=2$ also increases and decreases when the secondary wave passes through the location at which $\phi=\phi_p$; this is also a consequence of the interaction between the planet and the secondary wave, and especially of the $m=2$ component. \RED{ \subsection{Effect of instability for gap structures} \label{subsec:rbw} Previous studies \citepeg{Li2000,Tanigawa_Ikoma2007,Lin2012,Lin2014,Ono_Nomura_Takeuchi2014,Kanagawa2015a,Ono2016} pointed out a possibility that hydrodynamic instabilities (i.e., Rayleigh Instability, Rossby Wave Instability) occur at the edge of the gap induced by the planet. The condition for the onset of the RI is given by $d(R^2\Omega)/dR < 0$ \citep{Chandrasekhar1961}. In our parameter range, however, this condition is not satisfied because the gap structure is less steep. For a very deep gap with a massive planet as $\mpl/\mstar \simeq 4\times 10^{-3}$, the RI may be violated as shown by \cite{Fung_Chiang2016}. \begin{figure*} \begin{center} \resizebox{0.98\textwidth}{!}{\includegraphics{vorticities_h5e-2.eps}} \caption{ (Left)Distributions of the azimuthal averaged values of $\omega_z c_s^2$ for $\mpl/\mstar = 10^{-4}$ (solid), $5\times 10^{-4}$ (dotted), and $10^{-3}$ (cross), respectively, when $\hp/\rp=1/20$ and $\alpha=10^{-3}$. (Right)Two-dimensional distribution of $\omega_z c_s^2$ in the case of $\mpl/\mstar=10^{-3}$ in the left panel. \label{fig:vorticities_h0.05} } \end{center} \end{figure*} According to \cite{Lovelace_Colgate_Nelson1999}, the RWI can occur when the function of $\omega_z S^{2/\Gamma} = \omega_z c_s^2$ (when locally isothermal), where $\omega_z = (\nabla \times \bm{V})_z/\rhosurf$ is the potential vorticity and $c_s$ is sound speed, respectively, has an extreme value (a maximum or minimum), whereas \cite{Ono2016} have shown that Lovelace's condition is not a sufficient one. The left panel of Figure~\ref{fig:vorticities_h0.05} shows radial distributions of azimuthal averaged values of $\omega_z c_s^2$ for $\mpl/\mstar=10^{-4}$, $5\times 10^{-4}$, and $10^{-3}$ when $\hp/\rp=1/20$ and $\alpha=10^{-3}$. When $\mpl/\mstar=10^{-4}$ and $5\times 10^{-4}$, there is no extreme around the gap edge. When $\mpl/\mstar=10^{-3}$, we can find a local minimum and maximum around $R/\rp=1.2$. Even in this case, the two-dimensional distribution of $\omega_z c_s^2$ is almost axis-symmetric (see the right panel of the figure) and there is no vortex-like structure, it does not seem that the RWI affects the gap structure. The RWI can occur at the edge of the planet-induce gap in the case of the low viscosity (e.g.,\cite{Yu2010,Fu_Li_Lubow_Li2014,Zhu_Stone_Rafikov_Bai2014},see also Appendix~\ref{sec:RWI_inviscid}). The RWI may constraint the gap structure in this case, as stated by \cite{Hallam_Paardekooper2017}. However, with viscosity larger than $\alpha \sim 10^{-3}$ -- $10^{-4}$, the RWI is stabilized by the viscosity during the long-term evolution \citepeg{Lin2014,Fu_Li_Lubow_Li2014,Zhu_Stone2014}. In fact, there is no clear evidence which the RWI affects the gap structure in Figure~\ref{fig:vorticities_h0.05}. In this case, the effect of the RWI on the gap structure may not be essential. } Using the results of the survey of the hydrodynamic simulations (34 runs) over the wide parameter space, we derived a quantitative relationship between a planet and the gap radial structure. We also obtain an empirical model of the wave propagation, from the results of the survey. Using this empirical model of wave propagation to a semianalytical model of the gap structure provided by \cite{Kanagawa2015a}, we can obtain the gap structures which are in very good agreement with these obtained by the two-dimensional simulations. Our results can be summarized as follows: \begin{enumerate} \item \RED{We extended the scaling relation of the gap width given by \cite{Kanagawa2016a}. From this scaling relation, we derived an empirical formula for the surface density distributions of the gap (equation~\ref{eq:gap}), which accurately reproduces the results of hydrodynamic simulations; see Figure~\ref{fig:avgdens_simple}.} \item Our model puts a constraint on the origin of the observed gap structure, as discussed in Section~\ref{subsec:obs_applications}. Using equations~(\ref{eq:ratio_widths}) and (\ref{eq:rel_depth_width}), we can judge whether an observed gap is induced by the planet; see Figure~\ref{fig:gapwidth_ratio}. \item We confirmed the validity of the empirical model for wave propagation adopted in \cite{Kanagawa2015a} (equation~\ref{eq:deposited_torque}) with the parameters given by equations~(\ref{eq:xdep}) and (\ref{eq:wdep}). This model describes wave propagation that is consistent with the results of the hydrodynamic simulations; see Figure~\ref{fig:deposited_am_comp_sims_models}. \RED{Using this wave propagation model, we can reproduce the gap structure obtained by the two-dimensional hydrodynamic simulations by the one-dimensional model; see Figure~\ref{fig:avgdens_comp_sims_models}.} \item \RED{Our model of wave propagation indicates that the waves excited by the larger planet carry the angular momentum at larger distances from the planet.} As the planet mass is larger, waves with smaller azimuthal wavenumber (e.g.,$m=2$) are excited strongly. The angular momentum is transported to the distant locations from the planet, even when the planet mass is larger as Jupiter; see Figure~\ref{fig:smallmode}. \item The development of modes with small azimuthal wavenumber changes the morphology of spiral waves. When the planet mass is sufficiently large, the $m=2$ mode creates a secondary wave launched from the site opposite from the planet (see Figure~\ref{fig:decomp_densmap}), in addition to a primary wave originated from the location of the planet. The secondary wave would be closely related with the transport of angular momentum. \end{enumerate} \RED{ As the planet mass increases, the nonlinear effects become significant in excitation and propagation of waves, and thus the mechanism of the gap formation is different from that when the planet is small in linear theory. } However, these theoretical mechanisms are not yet completely understood. To understand the mechanism of the gap formation and the morphology of the density waves induced by a giant planet, it will be necessary to further investigate the wave excitation and propagation when there are deep gaps. \begin{ack} This work was supported by JSPS KAKENHI Grant Numbers 23103004, 26103701, 26800106, and 26800229, and the Polish National Science Centre MAESTRO grant DEC- 2012/06/A/ST9/00276. KDK was supported by the ALMA Japan Research Grant of the NAOJ Chile Observatory, NAOJ-ALMA-0135. Numerical computations were carried out on the Cray XC30 at the Center for Computational Astrophysics, National Astronomical Observatory of Japan and the Pan-Okhotsk Information System at the Institute of Low Temperature Science, Hokkaido University. \end{ack} \appendix | 16 | 9 | 1609.02706 |
1609 | 1609.00535_arXiv.txt | {Spectroscopic observations of a solar eclipse can provide unique information for solar and exoplanet research; the huge amplitude of the Rossiter-McLaughlin (RM) effect during solar eclipse and the high precision of solar radial velocities (RVs) allow detailed comparison between observations and RV models, and they provide information about the solar surface and about spectral line formation that are otherwise difficult to obtain. On March 20, 2015, we obtained 159 spectra of the Sun as a star with the solar telescope and the Fourier Transform Spectrograph at the Institut f\"ur Astrophysik G\"ottingen, 76 spectra were taken during partial solar eclipse. We obtained RVs using $I_2$ as wavelength reference and determined the RM curve with a peak-to-peak amplitude of almost 1.4\,km\,s$^{-1}$ at typical RV precision better than 1\,m\,s$^{-1}$. We modeled the disk-integrated solar RVs using well-determined parameterizations of solar surface velocities, limb darkening, and information about convective blueshift from 3D magnetohydrodynamic simulations. We confirm that convective blueshift is crucial to understand solar RVs during eclipse. Our best model reproduced the observations to within a relative precision of 10\,\% with residuals lower than 30\,m\,s$^{-1}$. We cross-checked parameterizations of velocity fields using a Dopplergram from the Solar Dynamics Observatory and conclude that disk-integration of the Dopplergram does not provide correct information about convective blueshift necessary for m\,s$^{-1}$ RV work. As main limitation for modeling RVs during eclipses, we identified limited knowledge about convective blueshift and line shape as functions of solar limb angle. We suspect that our model line profiles are too shallow at limb angles larger than $\mu = 0.6,$ resulting in incorrect weighting of the velocities across the solar disk. Alternative explanations cannot be excluded, such as suppression of convection in magnetic areas and undiscovered systematics during eclipse observations. To make progress, accurate observations of solar line profiles across the solar disk are suggested. We publish our RVs taken during solar eclipse as a benchmark curve for codes calculating the RM effect and for models of solar surface velocities and line profiles. } | The observation of radial velocities (RVs) of the Sun as a star is gaining interest in both solar and exoplanet communities. For solar applications, the availability of methods for computing one single parameter (the RV) from sometimes complex solar data and its comparison to other proxies opens new possibilities to search for physical relations. For the exoplanet community, the Sun is the one benchmark object on which methods can be tested with precision observations that offer more information than is available for any other star, such as limb darkening and surface velocity fields. Obtaining RVs in stars involves measuring the Doppler shift of spectral lines at different times of observation. There are several ways to precisely determine a shift of spectral lines. They all have in common that a change in the line profile shape caused by an emerging active region on the star, for instance, can mimic a Doppler shift of several m\,s$^{-1}$\citep[see, e.g.,][and references therein]{2016PASP..128f6001F}. Finding ways to account for this type of bias is currently an important step in the search for Earth-like extrasolar planets \citep[e.g.,][]{2013ApJ...770..133H, 2014Sci...345..440R, 2015ApJ...805L..22R, 2015arXiv150609072A}. In the context of stars, it has been realized that m\,s$^{-1}$ precision can only be reached when stellar surface velocities, and in particular the effect of magnetic activity on convective blueshift, are thoroughly understood \citep{2010A&A...512A..38L, 2010A&A...512A..39M, 2010A&A...519A..66M, 2014ApJ...796..132D, 2015ApJ...798...63M, 2016MNRAS.457.3637H}. The Sun provides an invaluable reference for understanding the influence of stellar surface velocity fields and active regions because it is possible to relate precision RV observations to spatially resolved solar surface information \citep{2010A&A...519A..66M, 2016MNRAS.457.3637H}. However, it is extremely difficult to obtain solar disk observations with m\,s$^{-1}$ precision or better, and current solar Dopplergrams are known to exhibit severe problems in this respect \citep{2012SoPh..275..285C, 2013ApJ...765...98W}. Furthermore, measuring precise Sun-as-a-star RVs is very challenging because the spatial extension of the Sun implies that feeding light from the Sun into a spectrograph involves the problem of collecting light from all areas of the solar disk equally. Earlier attempts have demonstrated this difficulty \citep{Jimenez1986AdSpR, Deming1987ApJ, Deming1994ApJ, McMillan1993ApJ} but there is new motivation for instruments facing this challenge, for example, observing the Sun with HARPS \citep{2015ApJ...814L..21D} and in the G\"ottingen Solar Radial Velocity Project used for this work \citep{2016arXiv160300470L}. A particularly fruitful exercise to understand the effects across an extended stellar disk in RV measurements is the observation of an eclipse of the Sun by one of its planets or the Moon. During eclipse, solar and stellar RVs exhibit the so-called Rossiter-McLaughlin (RM) effect \citep{1924ApJ....60...15R, 1924ApJ....60...22M} with excursion of the RV curve to the blue and the red in proportion to the rotational velocity of the star and to the amount of the surface eclipsed \citep[e.g.,][]{2005ApJ...622.1118O, 2007ApJ...655..550G, 2008ApJ...677.1324T, 2009ApJ...696.1230F}. Observations of planetary transits of the Sun can also be performed with night-time telescopes observing the light reflected from other bodies of the solar system in order to avoid the effect of the spatially extended solar disk \citep[e.g.,][]{2013MNRAS.429L..79M, 2015MNRAS.453.1684M}. The RM effect in stars is used to determine the geometry of planetary systems. Convective blueshift influences the shape of the RM curve and potentially biases measurements of planetary system geometries \citep{2011ApJ...733...30S, 2013A&A...550A..53B, 2016ApJ...819...67C, 2016A&A...588A.127C}. A planetary transit can also be used to observe local stellar line profiles \citep{2015csss...18..853D}, and in the same sense, observations of the eclipsed Sun provide a unique opportunity to gather information about local solar line profiles and about the influence of convective blueshift on RM measurements. \citet{2015PASJ...67...10T} carried out observations of the solar eclipse visible in Japan on May 21, 2012, with the aim to extract information on solar rotation. At maximum, the Moon eclipsed 93\,\% of the visible Sun during their observations. They showed an RM curve with a peak-to-peak amplitude of almost 2\,km\,s$^{-1}$ and calculated a model of the RV curve that matches within a standard deviation of 41\,m\,s$^{-1}$. Their model includes solar differential rotation and limb darkening, and they attempted to gain information on solar surface rotation from their RV curve. \citet{2015PASJ...67...10T} also provided information about the spectral line shape during eclipse and discussed deviations between ATLAS9 solar model calculations \citep{1993KurCD..13.....K} and their observations. In this work, we investigate our solar eclipse observations from Mar 15, 2015, with a maximum eclipse of 75.9\,\%. We argue that solar surface rotation and large-scale granulation patterns are relatively well described from different tracers \citep{1992ASPC...27..205S, 2004A&A...428.1007D}. On the other hand, the effect of convective blueshift is less well quantitatively understood, in particular its variation across the solar disk \citep[e.g.,][]{1978SoPh...58..243B}. In addition, convective flows can be suppressed in the presence of magnetic fields \citep{1982Natur.297..208L}, with the consequence that magnetic areas affect RV observations of active stars and the Sun. Therefore, we assume in our work that the geometry of the eclipse as well as surface rotation and granulation velocity fields are correctly known from independent information. As the most critical parameters for modeling solar RVs we identify line intensity changes across the solar disk and the dependence of convective blueshift on solar limb angle. | On March 20, 2015, we obtained 159 high-resolution FTS spectra of the Sun as a star, 76 of them during partial solar eclipse. From each spectrum, we derived the apparent absolute RV. Before and after the eclipse, our RVs are tracking barycentric motion. During eclipse, they provide an exquisite benchmark RM curve with a peak-to-peak amplitude of almost 1.4\,km\,s$^{-1}$. The geometry of the event is fully determined by our knowledge about solar, lunar, and terrestrial motion. Doppler velocities on the solar surface and limb darkening are also well known, but information about the limb angle dependence of solar convective blueshift and spectral line shape is incomplete. With this paper, we publish the RV values of our observations. The time series can be used to test line profile calculations and codes calculating RM curves for exoplanet observations. The amplitude of the RM curve is orders of magnitudes higher than in typical exoplanet observations, while the uncertainties in the RVs are very small (S/N$>1000$). In contrast to other stars, much information is available about solar surface velocities. We tested whether we were able to reproduce the RM curve during solar eclipse using a set of different approaches. First, we parameterized limb darkening and solar surface rotation using standard descriptions but neglected convective blueshift. We calculated surface integrated RVs by computing the intensity-weighted average of the radial velocities over the visible solar disk. This first attempt resulted in an RV curve that qualitatively matched the observations, but revealed a clear offset with the RM curve being off to the blue. In our second model, we included a qualitative prescription of convective blueshift leaving all other steps as in our first model. This resulted in a much better reproduction of the observed RV curve, but still revealed residuals as high as 40\,m\,s$^{-1}$ especially during the first half of the eclipse. In our model~3 we calculated spectral line profiles for different locations on the solar surface and computed a disk-integrated line from which we determined the apparent solar RVs like in our observations. For the line profiles, we used results from a 3D-MHD calculation that inherently includes information on convective blueshift and the variation of the profile shape. While this model provided the best match to our observations, residuals were still as high as 30\,m\,s$^{-1}$, the relative errors were smaller than 10\,\%. As a consistency check, we used solar surface observations from HMI in model 4 to describe limb darkening and solar velocity fields. The results of this exercise are very similar to those from model~1. This showed the limitation of the HMI Dopplergrams for full solar disk calculations. It is relevant for investigations of convective blueshift that use Dopplergram observations as reference \citep[e.g.,][using SOHO/MDI and SDO/HMI, respectively]{2010A&A...519A..66M, 2016MNRAS.457.3637H}. With the goal to identify the reason for the mismatch between our models and the RV observations, we inspected the line profiles modeled and observed during different phases of the eclipse. We found a mismatch of line depths close to phases 0.25 and 0.75, but a rather good match at phase 0.5 and outside eclipse. We concluded that the line depths at intermediate limb angles, $\mu$ between 0.6 and 1, may be systematically overestimated, which could lead to an incorrect weighting of velocities across the solar disk. Identifying reasons for this problem is beyond the scope of this work, but we see the need to obtain high-quality spectral observations of the solar surface at well-defined limb angles with high wavelength accuracy. As the main cause for the mismatch between our model calculations and observed RVs, we identified our limited knowledge about line profiles and convective blueshift as a function of solar limb angle. Alternative explanations include the variation of blueshift with magnetic field strength, incorrect assumptions about solar rotation and granulation, and so far undiscovered mechanisms affecting the spectral observation of the spatially extended solar disk during eclipse. The solution of this puzzle will have ramifications for solar physics because a better understanding of the local line profiles is required, and for exoplanet research because it will lead to a more accurate description of the RM curve in particular, and of stellar RVs in general. | 16 | 9 | 1609.00535 |
1609 | 1609.09414_arXiv.txt | In the context of structure formation with ultralight axion dark matter, we offer an alternative explanation for the mass relation of solitonic cores and their host halos observed in numerical simulations. Our argument is based entirely on the mass gain that occurs during major mergers of binary cores and largely independent of the initial core-halo mass relation assigned to hosts that have just collapsed. We find a relation between the halo mass $M_h$ and corresponding core mass $M_c$, $M_c\propto M_h^{2\beta-1}$, where $(1-\beta)$ is the core mass loss fraction. Following the evolution of core masses in stochastic merger trees, we find empirical evidence for our model. Our results are useful for statistically modeling the effects of dark matter cores on the properties of galaxies and their substructures in axion dark matter cosmologies. | \label{sec:intro} Ultralight scalar fields can be a viable candidate for dark matter if they are in a very cold state (e.g.,\cite{Turner:1983he,Press:1989id,Sin:1992bg,Sahni:1999qe,Hu:2000ke,Goodman:2000tg,Peebles:2000yy,Amendola:2005ad}). If consisting of particles of mass $\sim10^{-22}{\rm eV}$ \cite{Marsh2014,Schive2014a,Schive2014b,Bozek:2014uqa,Marsh2015b,Schive2016,Marsh2015,Sarkar2016,Calabrese2016,Gonzales-Morales:2016mkl} these candidates can potentially solve the well-known problems faced by pure cold dark matter (CDM) models on small scales (see \cite{Hui:2016ltb} for a recent review). Possible constituents are ultralight axions (ULAs) that are produced nonthermally via the misalignment mechanism~\cite{Witten:1984dg,Svrcek:2006yi,Arvanitaki:2009fg}. If self-interactions can be neglected, this type of dark matter candidate is often referred to as fuzzy dark matter (FDM)~\cite{Sahni:1999qe,Hu:2000ke}. Unlike CDM which produces cuspy halo profiles, FDM produces flat halo cores~\cite{Schive2014a,Schwabe:2016rze,Veltmaat:2016rxo} on scales smaller than the de Broglie wavelength of particles with the halo's virial velocity, the so-called quantum Jeans length~\cite{Hu:2000ke,Woo:2008nn}. Below this scale, quantum effects suppress gravitational collapse. By performing a Jeans analysis, it is found in~\cite{Schive2014a} that the cored halo profile corresponding to FDM with mass $m_a=0.81\times10^{-22}{\rm eV}$ can well reproduce the radial distribution of stars and their velocity dispersion in the Fornax dwarf spheroidal (dSph) galaxy. Further analysis on multiple stellar subpopulations in the Fornax and Sculptor dSph galaxies is done in~\cite{Marsh2015b} and an upper bond, $m_a<1.1\times10^{-22}{\rm eV}$, on the FDM mass is found by assuming that FDM alone can resolve the cusp-core problem. A similar constraint is found in~\cite{Chen:2016unw} from Jeans analysis of eight classical dSph galaxies. In~\cite{Gonzales-Morales:2016mkl}, it is demonstrated that Jeans analysis may be biased due to uncertainties in the assumed halo profile. Instead, a more stringent unbiased constraint, $m_a<0.4\times10^{-22}{\rm eV}$, is obtained in~\cite{Gonzales-Morales:2016mkl} by analyzing the averaged velocity dispersion of dSph galaxies. Coherent oscillations of FDM also lead to a sharp suppression of the power spectrum~\cite{Hu:2000ke} and halo formation~\cite{Marsh2014,Schive2016,Du:2016zcv,Corasaniti:2016epp} below the Jeans scale. In turn, this cutoff scale for FDM halos puts a lower bound on the FDM mass since deviations from CDM cannot violate the constraints given by current observations. Using the cosmic microwave background and galaxy clustering data, \cite{Hlozek:2014lca} find a lower bound on the FDM mass, $m_a\gtrsim10^{-24}{\rm eV}$. Constraints from UV luminosity functions and reionization are much tighter, e.g.~\cite{Corasaniti:2016epp} find $m_a\gtrsim1.6\times10^{-22}{\rm eV}$ (see also~\cite{Bozek:2014uqa} and \cite{Schive2016}). This lower bound is in tension with the upper bound obtained from dwarf galaxies. Furthermore, the Ly$\alpha$ forest also puts a tight constraint on the FDM mass similar to the case of warm dark matter (WDM)~\cite{Viel:2013apy,Marsh2014,Schneider:2016uqi}. Thus, FDM may also suffer from the \emph{Catch 22} problem~\cite{Maccio:2012qf} like WDM, i.e. either producing too small halo cores or too few low-mass halos. However, as discussed in~\cite{Gonzales-Morales:2016mkl}, to get more consistent constraints we need to consider details of the interplay between FDM and baryonic physics. The baryonic feedback may help reconcile the tension between different observations~\cite{Wetzel:2016wro}. Simulations of cosmological structure formation~\cite{Schive2014a} and merging solitonic solutions~\cite{Schive2014b,Schwabe:2016rze} based on the Schr\"{o}dinger-Poisson (SP) equations indicate that FDM halos contain distinct cores surrounded by Navarro-Frenk-White-like profiles~\cite{Navarro:1995iw}. \cite{Schive2014a} find that the mass of these cores, $M_c$, is related to the halo mass, $M_h$, by a power law relation, $M_c\propto M_h^{1/3}$. They propose an explanation based on the relation $M_c=\alpha\left(|E|/M\right)^{1/2}$, where $E$ is the total energy, $M$ is the total mass, and $\alpha$ is a constant of order unity, which they motivate heuristically with nonlocal consequences of the Heisenberg uncertainty relation. Identifying $E$ and $M$ with the energy of the halo $E_h$ and its virial mass $M_h$, they arrive at the numerically measured core-halo mass relation~\cite{Schive2014b}. Note that while $M_c \sim |E|^{1/2} M^{-1/2}$ is consistent with the intrinsic scaling properties of the SP equations (see, e.g., \cite{Guzman2004}), it is not unique (i.e., it can be multiplied by any scale invariant combination of $|E|$ and $M$). Removing any residual effects of the scaling symmetry by constructing and analyzing scale invariant quantities, \cite{Schwabe:2016rze} were unable to reproduce this relation in simulations of solitonic core mergers. Furthermore, the model of \cite{Schive2014b} does not account for the combined evolution of $M_c$ and $M_h$ by halo mergers after the initial collapse of density perturbations which is known to be an important ingredient in hierarchical structure formation. Comparing the initial and final masses of merging cores, \cite{Schwabe:2016rze} find a universal behavior of the core mass loss in mergers that depends nearly entirely on the mass ratio. Implementing this relation in a semianalytic model (SAM) for galaxy formation, \cite{Du:2016zcv} studies the effects of the core on the substructure of Milky way-sized FDM halos. Here, we present a model for the core mass as a function of halo mass which is entirely based on the fractional core mass loss during major mergers. No further assumptions about the quantum nature of FDM halos and cores are necessary. In particular, our model is independent of the dynamics of halo formation by gravitational collapse and hence insensitive to the initial core-halo mass relation of newly formed halos. We find a simple relation between the core and halo mass whose slope is a function of the core mass loss fraction. We provide numerical evidence for this dependence using stochastic merger trees. The existence of compact cores in the halo substructure has many potentially observable direct signatures in, for instance, rotation curves of dwarf galaxies \cite{Rhee:2003vw,Oman:2015xda}, gravitational lensing \cite{Mao:1997ek,Metcalf:2001ap,Hezaveh2014,Hezaveh2016}, globular cluster streams in the Milky Way \cite{Odenkirchen:2000zx,Grillmair:2006bd}, or the thickening of the thin galactic disk \cite{Toth1992,Quinn1993,Navarro:1994zk,Walker:1995ef,Sellwood:2013npa,Benson:2003uc,Kazantzidis:2007hy}. Indirectly, the effects of compact cores on star formation at high redshifts may be probed by the reionization history and the high-$z$ galaxy luminosity function. In addition to providing a simple explanation for the core-halo mass relation, it is straightforward to produce realizations of our stochastic model from modified EPS merger trees \cite{Du:2016zcv}. Since--as we will show--the core mass is determined by the individual accretion history, it can be modeled more realistically using individual mass accretion histories that recover not only the mean core-halo mass relation but also its scatter. | \label{sec:conclusions} By considering the merger history of dark matter halos in scenarios with ultralight bosonic dark matter, we offer an alternative explanation for the core-halo mass relation observed in cosmological simulations. We provide evidence for our model using stochastic merger trees and show that the core-halo mass relation depends only on the mass loss fraction of cores during binary mergers, $M_c\propto M_h^{2\beta-1}$. We find that for $\beta=0.7$~\cite{Schwabe:2016rze}, this relation fits numerical data from cosmological simulations very well~\cite{Schive2014b}. A similar approach may be employed to predict the statistical distribution of gravitationally bound substructures (axion stars or miniclusters) in scenarios with more massive axionlike particles or QCD axions. Instead of a single solitonic core, each dark matter halo hosts a large number of these objects and mergers take place both inside of individual halos and during halo mergers. Although a unique core-halo mass relation does not exist in this case, the universal mass gain for each substructure merger may still allow the construction of a stochastic model similar to ours. We will explore this possibility in future work. | 16 | 9 | 1609.09414 |
1609 | 1609.02530_arXiv.txt | {The recent development of brand new observational techniques and theoretical models have greatly advanced the exoplanet research field. Despite significant achievements, which have allowed the detection of thousands extrasolar systems, a comprehensive understanding of planetary formation and evolution mechanisms is still desired. One relevant limitation is given by the accuracy in the measurements of planet-host star ages. The star GJ 504 has been found to host a substellar companion whose nature is strongly debated. There has been a recent difference of opinion in the literature owing to the uncertainty on the age of the system: a young age of $\sim$ 160 Myr would imply a giant planet as a companion, but a recent revision pointing to a solar age ($\sim$ 4 Gyr) instead suggests a brown dwarf.} {With the aim of shedding light on this debated topic, we have carried out a high-resolution spectroscopic study of GJ 504 to derive stellar parameters, metallicity, and abundances of both light and heavy elements, providing a full chemical characterisation. The main objective is to infer clues on the evolutionary stage (hence the age) of this system.} {We performed a strictly differential (line-by-line) analysis of GJ 504 with respect to two reference stars, that is the planet-host dwarf $\iota$ Hor and the subgiant HIP 84827. The former is crucial in this context because its stellar parameters (hence the evolutionary stage) is well constrained from asteroseismic observations. Regardless of the zero point offsets, our differential approach allows us to put tight constraints on the age of GJ 504 with respect to $\iota$ Hor, thereby minimising the internal uncertainties.} {We found that the surface gravity of GJ 504 is 0.2 $\pm$ 0.07 dex lower than that of the main-sequence star $\iota$ Hor, suggesting a past turn-off evolution for our target. The isochrone comparison provides us with an age range between 1.8 and 3.5 Gyr, with a most probable age of $\approx$ 2.5 Gyr. Thus, our findings support an old age for the system; further evidence comes from the barium abundance, which is compatible with a solar pattern and not enhanced as observed in young stars.} {We envisaged a possible engulfment scenario to reconcile all the age indicators (spectroscopy, isochrones, rotation, and activity); this engulfment could have occurred very recently and could be responsible for the enhanced levels of rotation and chromospheric activity, as previously suggested. We tested this hypothesis, exploiting a tidal evolution code and finding that the engulfment of a hot Jupiter, with mass not larger than $\approx$ 3 M$_{j}$ and initially located at $\approx$ 0.03 AU, seems to be a very likely scenario.} \keywords {stars: abundances --stars: fundamental parameters --stars: individual (GJ 504) -- stars: solar-type} | Hunting for planetary systems outside our solar system, with the ultimate goal of revealing habitable worlds, is undoubtedly one of the most intriguing and fascinating topics in modern astrophysics. Outstanding efforts have been put forwards to detect (and characterise) extrasolar systems, such that 5437 planet candidates have been identified so far\footnote{Source \url{http://exoplanets.eu}, as for July 2016}. The presence of those systems has been mostly inferred via indirect methods (e.g. radial velocity variations or transits). However, direct imaging techniques are starting to provide us with powerful and complementary tools to detect planetary companions, thanks to the advent of new-generation instruments purposely designed for this goal, such as GPI, ScEXAO, and SPHERE (see \citealt{macintosh14}; \citealt{martinache09}; \citealt{beuzit08}, respectively). Unfortunately, beyond the technical limitations, which are nowadays primarily due to adaptive optics performance, several other fundamental issues affect our comprehension of the substellar regime, preventing us from gathering extensive knowledge. Along with several controversies related to, for instance, planetary atmosphere models, evolutionary stages of substellar objects, different formation, and early evolution scenarios (core accretion versus disc instability, and/or hot-start versus cold-start models; \citealt{pollack96}; \citealt{kratter10}; \citealt{marley07}; \citealt{fortney08}), the major concern in characterising exoplanetary systems is represented by the uncertainty on the stellar age. The age estimate of the planet-host star dramatically impacts the inferred mass for substellar companions and hence the calibration of the age-luminosity relationship for substellar objects. Thus, it is crucial to our understanding of how planets have formed. Expanding upon our previous work on GJ 758 (see \citealt{vigan16}), in this paper we present a similar investigation focussing on the curious case of GJ 504. The age of this star is hotly debated. The short rotational period of P$_{rot}$=3.3 days (\citealt{donahue96}) and the pronounced level of chromospheric and coronal activities seem to suggest a young age for the system, that is 160$^{+70}_{-60}$ Myr from gyrochronology and, from the chromospheric level, slightly older at 330$\pm$180 Myr, as traced by the Ca~{\sc ii} H and K emission. Assuming the young age of 160 Myr and spectroscopic parameters as published by \cite{valenti05}, \cite{kuzuhara13} concluded that GJ 504 is a young, main-sequence star with $T_{\rm eff}$$\approx $6200 K and log$g$ = 4.6 dex (with $g$ in cm s$^{-2}$). These authors were able to claim that the substellar companion orbiting at 43.5 AU, discovered through direct imaging observations within the framework of the SEEDS survey, has an inferred mass of 4$^{+4.5}_{-1.0}$ M$_{Jup}$. This is the lowest mass companion ever directly imaged to date. However, \cite{fuhrmann15} questioned this finding, arguing that the star is actually much older and is comparable with a turn-off stage of evolution and a solar-like age. In addition, these authors envisaged a possible merging scenario, with a now engulfed substellar object, to account for the enhanced level of activity and rotation. This study suggested an upward revision for the companion mass, implying that we are dealing with a brown dwarf (M$\approx$30-40 M$_{Jup}$) rather than a giant planet. An additional motivation for the interest in the GJ 504 system is derived by \cite{skemer16}, who found an indication that the substellar companion is more enriched ([M/H]$\approx$+0.5) in metal than its parent star ([M/H] $\approx$+0.1-0.3). We carried out a strictly differential analysis of GJ 504 with respect to three other reference stars ($\iota$ Horologii, HIP 84827, and the Sun) to shed light on this complicated picture, by obtaining in this way very accurate spectroscopic parameters and abundances. We derived temperature, gravity, metallicity, and several key species abundances, from the light Li and C up to elements synthesised via neutron-capture reactions (Ba). The manuscript is organised as follows: In Section \ref{sec:abu} we present the spectroscopic observations and abundance analysis procedure and results. In Section~\ref{sec:age} we extensively discuss all the other age indicators, whereas we sketch a plausible engulfment scenario to account for all the observed properties in Section~\ref{sec:merging}. We stress that in the present work we are focussing on the fundamental properties of the parent star. The complete characterisation of the substellar companion GJ 504b will be published in a companion paper (Bonnefoy et al., in preparation). | In this paper we have presented a comprehensive scrutiny of the fundamental properties for GJ 504. The star hosts a substellar companion located at $\approx$ 44 AU, which has been claimed to be one of the lowest mass objects ever observed via direct imaging techniques (\citealt{kuzuhara13}). However, this conclusion relied on a young age for GJ 504, namely less than $\sim$ 200 Myr, as indicated by the rotational period of P=3.33 days. Conversely, \cite{fuhrmann15}, thanks to high-resolution spectroscopic observations, found evidence for a much older age ($\approx$ 3 - 6 Gyr), suggesting that the substellar companion is actually a brown dwarf rather than a giant planet. In order to reconcile this old age, coming from isochrones and spectroscopy, with the indications of youth given by rotational and activity properties, these authors conceived the possibility of a merging event. The star GJ 504 might have engulfed a planetary-mass object that caused the spin up of rotational velocity along with enhancements in the chromospheric activity level. To shed light on this debated and complex picture, we exploited a strictly differential (line-by-line) analysis of GJ 504 with respect to two reference stars, namely $\iota$ Hor and HIP 84827. The role of $\iota$ Hor is particularly critical in this respect because its properties are very well constrained from an independent tool, i.e. asteroseismic observations (see \citealt{vauclair08}). Our results indicate that the surface gravity of GJ 504 is 0.2 $\pm $ 0.07 dex lower than that of the main-sequence star $\iota$ Hor, implying a slightly more advanced evolutionary stage for the object. The isochrone comparison provides us with an age range between 1.8 and 3.5 Gyr, which is qualitatively in agreement with Fuhrmann \& Chini results, although slightly younger. Interestingly, the solar Ba abundance also points to a very old age for the system, which is at variance with young stars in clusters and in the field (\citealt{dorazi09}; \citealt{desidera11}) that are known to be characterised by an extremely high Ba content ([Ba/Fe] up to 0.6 dex). To investigate the merging scenario suggested by Fuhrmann \& Chini, we ran a tidal evolution code, assuming an age of roughly $\approx$ 2 Gyr and imposing the condition that the engulfment event could not have occurred more than 200 Myr ago; otherwise the star would have lost the angular momentum and slowed down. Our tests indicate that a very plausible system architecture would result in an initial configuration of a planetary companion (with mass not larger than $\sim$ 3 M$_{J}$) located at 0.03 AU. This is probably because Kozai cycles (due to the presence of the external sub-stellar companion) have caused an inward migration. From such a small distance, the low-mass body has been affected by the stellar tides and slowly started to spiral down on the central star. If this were the case, we would expect to reveal planetary remnants such as rocky cores of the now defunct hot Jupiter in the proximity of star GJ 504. Current measurements prevent us from investigating this issue, but we plan to have purposely designed observations to detect such a small signal in radial velocity variations. | 16 | 9 | 1609.02530 |
1609 | 1609.06388.txt | Extreme adaptive optics systems are now in operation across the globe. These systems, capable of high order wavefront correction, deliver Strehl ratios of $\sim90\%$ in the near-infrared. Originally intended for the direct imaging of exoplanets, these systems are often equipped with advanced coronagraphs that suppress the on-axis-star, interferometers to calibrate wavefront errors, and low order wavefront sensors to stabilize any tip/tilt residuals to a degree never seen before. Such systems are well positioned to facilitate the detailed spectroscopic characterization of faint substellar companions at small angular separations from the host star. Additionally, the increased light concentration of the point-spread function and the unprecedented stability create opportunities in other fields of astronomy as well, including spectroscopy. With such Strehl ratios, efficient injection into single-mode fibers or photonic lanterns becomes possible. With diffraction-limited components feeding the instrument, calibrating a spectrograph's line profile becomes considerably easier, as modal noise or imperfect scrambling of the fiber output are no longer an issue. It also opens up the possibility of exploiting photonic technologies for their advanced functionalities, inherent replicability, and small, lightweight footprint to design and build future instrumentation. In this work, we outline how extreme adaptive optics systems will enable advanced photonic and diffraction-limited technologies to be exploited in spectrograph design and the impact it will have on spectroscopy. We illustrate that the precision of an instrument based on these technologies, with light injected from an efficient single-mode fiber feed would be entirely limited by the spectral content and stellar noise alone on cool stars and would be capable of achieving a radial velocity precision of several m/s; the level required for detecting an exo-Earth in the habitable zone of a nearby M-dwarf. | \label{sec:intro} Spectroscopy is a powerful technique widely used in astronomy. It was initially developed as a tool to determine the chemical composition and abundances of a target, and the first stellar spectrum was that of the sun taken by Fraunhofer in $1814$. Spectroscopy is now regularly used to study the kinematics of stellar systems, including their proper motion and orbital properties (in case of binarity or an exoplanetary system) and distance from Earth (i.e. redshift, in the case of distant galaxies). Low resolution spectroscopy has been used to map the structure of galaxies throughout the Universe~\citep{colless2001} and to confirm the presence of dark energy/matter and the expansion of the Universe~\citep{riess1998}. More locally, high-resolution spectroscopy has been used to understand stars and stellar structure by studying well-constrained systems like eclipsing binaries~\citep{las2002}. It has also been prolific in its detection yield of exoplanets~\citep{mayor95} and has been recently used for exoplanet characterization as well~\citep{snell2010,snell2014}. Spectroscopy provides a vast amount of information and is one of the key techniques employed throughout all areas of astronomy. Classically, spectrographs were mounted at either the Cassegrain or Coude focus, and fed via a slit~\citep{huggins1895}. While having the instrument directly mounted to the telescope provides the highest throughput and is therefore still common for instruments intended for spectroscopy of dim objects (MOSFIRE~\citep{mclean2012}, for example), it has several shortcomings for high-resolution spectrographs, where stability is often key. Instruments moving with the telescope (e.g. Cassegrain focus) experience a varying gravity vector and hence are prone to flexure. The Coude mounted version has the benefit of constant gravitational load and better insulation from environmental changes like temperature, as the Coude room can be isolated from the dome; however, beam transport requires multiple reflections and is somewhat lossy. Fiber fed instruments take this concept further by decoupling the spectrographs location from the telescope's foci. The advantages of feeding the instrument with a fiber instead of a slit were recognized early on, and in the late $1970$'s, the first fiber fed spectrographs were demonstrated~\citep{hill1980,hill1988}. Fibers provide an additional degree of decoupling in that they can be fed into a vacuum tank, where it is possible to realize a rigid fiber slit mounted directly to the spectrograph bench~\citep{quir2014}. Also, they provide what is termed scrambling (see Section~\ref{sec:enhspec}), which stabilizes the input illumination to the spectrograph optics~\citep{Stu2014}. Multimode fibers (MMF) with large cores and opening angles of the accepted light cone (quantified as the numerical aperture: the sine of the half-cone angle) were chosen for their superior collecting power from the seeing-limited telescopes of the day and have since been almost exclusively used to feed spectrographs. However, it was quickly realized that these fibers have several shortcomings. One issue was the fact that the focal ratio coming out of the fiber was degraded with respect to the injected focal ratio; this effect is known as focal ratio degradation (FRD)~\citep{ramsey1988}. With a larger cone angle at the output of the fiber, the spectrograph optics must be designed larger than desired, increasing the size and cost and reducing the stability of the instrument. The second issue was that the fibers did not fully scramble the image of the PSF incident at the input and indeed preserved significant spatial structure at the output. This resulted in a spatially and temporally varying slit illumination\footnote{We define the following terms here: the term PSF is used to describe the actual point-spread-function of the telescope including the atmosphere, which gives the light intensity distribution in the focal plane and is incident on the fiber input; the slit illumination (SI) or near field, refers to the input illumination of the spectrograph, either through a slit or at the exit of the coupling fiber (where the term near field comes from); the far-field illumination (FF) is the intensity distribution in the pupil of the spectrograph (far-field of a coupling fiber); and the spectral line spread function (SLSF~\citep{spro2012a, spro2012b}) describes the profile of a monochromatic line as seen by the spectrograph in its focal plane.} structure being injected into the spectrograph, complicating calibration. To address this shortcoming, scrambling techniques to eliminate fluctuations in the spatial structure of the near and far fields have been developed which include using non-circular fibers~\citep{chaz2012,roy2014} as well as double-scrambling with ball lenses~\citep{hunter1992}. Also, modal noise (see below) limits the ultimate signal-to-noise ratio (SNR) achievable with a MMF-fed instrument to levels below what is required for high precision Doppler measurements or stellar spectroscopy~\citep{baud2001}. To reduce modal noise, fiber shakers have been developed which provide mechanical agitation~\citep{plav2013}. Although these scrambling techniques work well, they are not perfect and will limit the accuracy of SLSF calibration for ultra-high precision measurements. Over the past $20$ years, adaptive optics systems have improved significantly and now diffraction-limited PSFs are commonplace at large observatories and are integral parts of future extremely large telescopes (ELTs). With a diffraction-limited telescope PSF, it becomes possible to utilize single-mode fibers (SMF) instead, which eliminate FRD and modal noise. We have recently demonstrated highly efficient coupling to SMF with the SCExAO instrument at Subaru Telescope~\citep{jovanovic2014,jovanovic2016a}. This opens the door to applying photonic technologies to efficient instrument design. With their compact footprint and broad functionality, photonic technologies will no doubt play a key role in future instrumentation. The aim of this work is to provide an overview of how the stable, high Strehl PSFs provided by advanced adaptive optics systems can enhance astronomical spectroscopy~\citep{Schwab2012}. This is an exciting topic that was recently discussed in a short article by~\cite{crepp14}. In Section~\ref{sec:ExAO}, we review the state-of-the-art in adaptive optics systems, and in Section~\ref{sec:enhspec}, we describe how these properties could be harnessed to develop innovative spectrograph concepts and enhance spectroscopy. Section~\ref{sec:future} details other innovative applications of adaptive optics for spectroscopy and Section~\ref{sec:app} outlines one possible instrument concept enabled by this technology. Section~\ref{sec:summary} rounds out the work with some concluding remarks. | \label{sec:summary} The high Strehl ratios ($\sim90\%$) and unprecedented PSF stability provided by ExAO systems can be used to advance stellar spectroscopy. With these properties, efficient direct injection of light into single-mode fibers or few port photonic lanterns becomes possible, eliminating imperfect input scrambling and modal noise and hence improving PSF calibration and spectrograph precision. Once the light is efficiently coupled to a diffraction-limited device it is possible to exploit a host of photonic functionalities with great efficiency, including spectral filters~\citep{trinh2013} and compact calibration combs~\citep{schwab15}. This enables miniaturized instruments to be developed that could be placed directly in the focal plane of an ELT. These instruments can be extremely compact and robust and they can be mass produced. Other features of ExAO facilities can also enable new capabilities like spatially resolved spectroscopy of close-separation binaries or even post-coronagraphic spectroscopy. By reducing the photon noise, it will be possible to detect and study molecular features in the atmospheres of giant planets in greater detail. ExAO systems are very versatile and will undoubtedly find applications in many fields other than their intended one of high contrast imaging. The key is to identify their potential and harness it. | 16 | 9 | 1609.06388 |
1609 | 1609.02689_arXiv.txt | Quasi-periodic pulsations (or QPPs) are periodic intensity variations in the flare emission, across all wavelength bands. In this paper, we review the observational and modelling achievements since the previous review on this topic by \citet{nakariakov2009qpp}. In recent years, it has become clear that QPPs are an inherent feature of solar flares, because almost all flares exhibit QPPs. Moreover, it is now firmly established that QPPs often show multiple periods. We also review possible mechanisms for generating QPPs. Up to now, it has not been possible to conclusively identify the triggering mechanism or cause of QPPs. The lack of this identification currently hampers possible seismological inferences of flare plasma parameters. QPPs in stellar flares have been detected for a long time, and the high quality data of the Kepler mission allows to study the QPP more systematically. However, it has not been conclusively shown whether the time scales of stellar QPPs are different or the same as those in solar flares. | \label{S-Introduction} In the past 15 years, it has become obvious that the solar corona hosts a variety of magnetohydrodynamic (MHD) waves. First, slow waves were detected in coronal plumes and loops \citep{deforest1998, berghmans1999}, quickly followed by kink waves excited by flares \citep{nakariakov1999}. For an overview of the earlier observations, we recommend a thorough reading of \citet{nakariakov2005} or \citet{demoortel2012b}. More recently, it was found that kink waves are observed in nearly all coronal loops \citep{tomczyk2007,vd2008,mcintosh2011,nistico2013}, and that slow waves are also detected in all coronal loops at all times \citep{krishnaprasad2012,banerjee2015}. \par The violent nature of the flare is bound to excite waves in its vicinity. Often, the direct observations of flares and their surroundings show clear oscillatory phenomena. Indeed, it is obvious that transverse waves are excited in nearby loops by the flare disturbance. In particular, the flare light curve shows periodic intensity increases and decreases. These are called {\em quasi-periodic pulsations (or QPPs)}. The classical example is the so-called ``7 sisters'' flare \citep{kane1983}, as depicted in Figure~\ref{fig:7sisters}, where a clear 8 s period is observed. \begin{figure} \centerline{\includegraphics[width=.5\linewidth]{kane1983}} \caption{Intensities as a function of time for the ``7 sisters'' flare. Picture taken from \citet{kane1983}.} \label{fig:7sisters} \end{figure} QPPs in solar flares have typical periods between a few seconds to a few minutes. For stellar flares, the periods may reach tens of minutes. These time scales puts the QPP oscillations in the MHD regime. \par QPPs were reviewed in depth by \citet{nakariakov2009qpp}. The current article aims to give an overview of recent theoretical and modelling results. Even though many new results have been obtained, still there is no consensus reached on what physical mechanism is responsible for the generation of QPPs. \par The importance of studying QPPs cannot be overstated. If the QPPs are generated by quasi-steady processes (also called magnetic dripping), their period probably contains information on the ongoing physical processes during the flare. If they are generated by magnetic dripping or MHD waves, they offer us a unique tool to seismologically probe the flare site. By comparing the QPP periods to theoretical models, it is possible to gauge plasma parameters in the flare surroundings, which are difficult to measure otherwise (\textit{e.g.} magnetic field). Moreover, it has been argued by \citet{fletcher2008,russell2013} that MHD waves are responsible for the energy transport from the reconnection site to the main emission sites in the flare foot-points. The QPPs could be a direct sign of this transport mechanism.\par Up to now, however, it is unclear what causes the QPPs (as mentioned above). This hampers progress in using QPPs for seismological purposes (using either the quasi-steady/dripping or wave description). | % \label{sec:conclusion} In this article, we have reviewed the observational and theoretical results on quasi-periodic pulsations (a.k.a. QPPs). These QPPs are periodic intensity modulations in the flare emission (across all wavelength bands). In this paper, we have focussed on reporting on the works since the publication of the review by \citet{nakariakov2009qpp}. \par Despite the many observational and theoretical advances in the last years, it has not been possible to determine what physical mechanism is responsible for causing the QPPs. There are three main lines of thought: (i) the periods could be created by external modulation of the reconnection (Section~\ref{sec:external}), (ii) the periods could be inherent to the reconnection process (Section~\ref{sec:flare}), or (iii) the periods are modulations of the flare particles by waves in the flaring loop (Section~\ref{sec:postflare}). We believe that the determination of the correct mechanism (at least for the long-period QPPs) will take place in the next few years, by comparing spatial information from the EUV to the temporal emission, where recent results have been achieved. It could be that the ALMA instrument will play a role in this \citep{wedemeyer2015}. \par We stress that QPPs carry a lot of potential for further research. Once the correct mechanism has been identified (and it is related to MHD waves), it is possible to use them for seismology: by comparing the QPP periods to models, we can determine physical quantities near the flare site. \par An interpretation in terms of MHD waves in the flaring loops (Section~\ref{sec:postflare}) is currently the most popular, because it explains most of the observational features (such as multi-periodicity). Making this assumption, it is possible to make progress in measuring flare parameters seismologically. Most of the solar and stellar QPP seismology models are currently based on crude models. However, it is a promising start, and more advanced models can lead to better understanding of the flare surrounding. In any case, it is important that flare models naturally incorporate QPPs, because they seem to occur in most flares.\par In our view, progress in the understanding of QPPs can be mainly made in two ways: \begin{enumerate} \item Observationally identifying the source location for the QPPs. \item Constructing realistic models for flares and their QPPs, which include forward modelling. \end{enumerate} \begin{acks} Inspiration for this review was obtained during ISSI and ISSI-BJ workshops. TVD thanks the Odysseus type II funding (FWO-Vlaanderen), IAP P7/08 CHARM (Belspo), GOA-2015-014 (KU~Leuven). EK is a beneficiary of a mobility grant from Belspo. Disclosure of Potential Conflicts of Interest: The authors declare that they have no conflicts of interest. \end{acks} | 16 | 9 | 1609.02689 |
1609 | 1609.02476_arXiv.txt | The performance of a wide-field adaptive optics system depends on input design parameters. Here we investigate the performance of a multi-object adaptive optics system design for the European Extremely Large Telescope, using an end-to-end Monte-Carlo adaptive optics simulation tool, DASP, with relevance for proposed instruments such as MOSAIC. We consider parameters such as the number of laser guide stars, sodium layer depth, wavefront sensor pixel scale, actuator pitch and natural guide star availability. We provide potential areas where costs savings can be made, and investigate trade-offs between performance and cost, and provide solutions that would enable such an instrument to be built with currently available technology. Our key recommendations include a trade-off for laser guide star wavefront sensor pixel scale of about 0.7~arcseconds per pixel, and a field of view of at least 7~arcseconds, that EMCCD technology should be used for natural guide star wavefront sensors even if reduced frame rate is necessary, and that sky coverage can be improved by a slight reduction in natural guide star sub-aperture count without significantly affecting tomographic performance. We find that adaptive optics correction can be maintained across a wide field of view, up to 7~arcminutes in diameter. We also recommend the use of at least 4 laser guide stars, and include ground-layer and multi-object adaptive optics performance estimates. | \label{sect:intro} The next generation of ground based astronomical telescopes will be the \elts \citep{eelt,tmt,gmt}, which are due to see first light within the next decade. All of these telescopes rely on \ao systems \citep{adaptiveoptics} to provide compensation for the degrading effects of atmospheric turbulence, thus allowing the scientific goals of these facilities to be met. Extensive simulation of \ao systems is required during the instrument design phases, so that predicted performance estimates can be made and design trade-offs explored and \ao designs optimised. High fidelity modelling of the performance of \ao and telescope systems can be implemented using Monte-Carlo simulation, which involves playing a time sequence of input atmospheric perturbations through the \ao system and telescope models. For \elt-scale instruments, these simulations are computationally expensive. Here, we report on an exploration of the parameter space related to a \moao system design study for the 39~m \eelt. We use \thedasp \citep{basden5,basden11} to perform this modelling. Our models are based on a system with both \lgss and \ngss (the number of which we explore), and we use a high resolution atmospheric model \citep[as used in previous studies, e.g.\ ][]{basden21} which is stratified into 35 discrete layers of turbulence. Our results have relevance for proposed and future \moao instruments such as MOSAIC, and also more generally for other wide-field \ao systems. Within this study, we investigate factors such as the number of \lgss, the elongation of \lgss as seen by the \wfss (due to the extent of the mesosphere sodium layer depth), the number, position and magnitude of \ngss within the field of view, \dm requirements, detector requirements, \wfs sensitivity and field of view and the effect of turbulence strength. We explore \ao performance across the field of view, and our default results are presented on-axis, which we show to be pessimistic compared to most of the rest of the field of view. We also consider the performance improvements achievable by operating the \lgs and \ngs \wfss at different frame rates, and consider the use of currently available commercial cameras as wavefront sensors. The results that we present can be used to aid design decisions for \elt instrumentation, and also to provide a benchmark for simulation comparison. These results are complementary to those from other modelling tools for \elt \moao instrumentation (which for a single on-axis channel can be viewed as \ltao), for example \citet{miskaltao,2014SPIE.9148E..6FA}. In \S2 we present the key parameters of our simulations, and details of the parameter space that is explored, along with key algorithms. In \S3 we present our resulting estimates of \ao system performance, and we conclude in \S4. | We have performed detailed Monte-Carlo modelling of several of the design parameters for a potential \eelt \moao instrument using a full Monte-Carlo end-to-end \ao simulation tool, \dasp. The recommendations that we draw from this study include \lgs pixel scale (typically optimal at 0.7~arcseconds per pixel), minimum sub-aperture size (at least $10\times10$ pixels), a study of number of guide stars, and reduction in \ngs \wfs frame rates so that a minimum detected flux is received (at least 10 photons per sub-aperture when using an \emccd). We identify current commercial cameras that would be suitable for wavefront sensors (reducing the risk associated with such an instrument). We include a study of several demanding \ngs asterisms, taken from availability of stars within a cosmological field, and also consider performance as a function of \lgs return flux. We also consider different \dm sizes and geometries and the effect of differential rotation between \dms and on-sky guide star positions. Although Strehl ratios are typically fairly low (6--10\% at H-band), the ensquared energy requirements for a typical spectrograph (e.g.\ MOSAIC) are likely to be met, being in the range of 50\% for energy within a 150~mas box. | 16 | 9 | 1609.02476 |
1609 | 1609.05898_arXiv.txt | The host star metallicity provide a measure of the conditions in protoplanetary disks at the time of planet formation. Using a sample of over 20,000 \textit{Kepler} stars with spectroscopic metallicities from the \texttt{LAMOST} survey, we explore how the exoplanet population depends on host star metallicity as a function of orbital period and planet size. We find that exoplanets with orbital periods less than 10 days are preferentially found around metal-rich stars ([Fe/H]$\simeq{} 0.15\pm0.05$ dex). The occurrence rates of these hot exoplanets increases to $\sim{} 30\%$ for super-solar metallicity stars from $\sim{} 10\%$ for stars with a sub-solar metallicity. Cooler exoplanets, that reside at longer orbital periods and constitute the bulk of the exoplanet population with an occurrence rate of $\gtrsim{} 90\%$, have host-star metallicities consistent with solar. At short orbital periods, $P<10$ days, the difference in host star metallicity is largest for hot rocky planets ($<1.7 ~R_\oplus$), where the metallicity difference is [Fe/H]$\simeq 0.25\pm0.07$ dex. The excess of hot rocky planets around metal-rich stars implies they either share a formation mechanism with hot Jupiters, or trace a planet trap at the protoplanetary disk inner edge which is metallicity-dependent. We do not find statistically significant evidence for a previously identified trend that small planets toward the habitable zone are preferentially found around low-metallicity stars. Refuting or confirming this trend requires a larger sample of spectroscopic metallicities. | Stellar metallicity is a good proxy of the initial metallicity of the protoplanetary disks, which in turn has an important impact on planet formation. Together with the disk mass, the disk metallicity determines the amount of solids available in protoplanetary disks for planet formation. Higher mass stars host more massive disks \citep[e.g.][]{2013ApJ...771..129A,2016arXiv160803621P} and a larger metallicity corresponds to a larger amount of condensible solids in the disk. Therefore, higher stellar masses and metallicities result in more building blocks available in the disk for planet formation. For gas giant planets, a correlation between planet occurrence and stellar metallicity \citep{2000A&A...363..228S, 2010PASP..122..905J,2012Natur.486..375B, 2013A&A...551A.112M} and stellar mass \citep{2007ApJ...670..833J,2010PASP..122..905J,2015A&A...574A.116R} has been well established. Theoretically, this can be understood as massive cores need to reach a critical mass of $\sim 10 ~M_\oplus$ to undergo runaway gas accretion before the gas dissipates, which is more likely to occur in disks with more solids \citep[e.g.][]{2004ApJ...616..567I,Alibert:2011kg,2012ApJ...751...81J,2012A&A...541A..97M}. \begin{figure} \includegraphics[width=\linewidth]{fig1.pdf} \caption{Histogram of spectroscopic metallicities of main-sequence stars in the \textit{Kepler} field from \texttt{LAMOST} ( blue). Photometric metallicities from \cite{2014ApJS..211....2H} for a sample of \textit{Kepler} targets brighter than Kpmag$=14$, representative of the brightness limit of the \texttt{LAMOST} stellar sample, are shown for comparison (hatched). \label{f:hist} } \end{figure} For smaller planets, those that have been found in abundance with the \textit{Kepler} spacecraft, correlations between planet occurrence and host star mass and metallicity are different from those for giant planets, and are less straightforward to interpret. The occurrence rate of these planets is \textit{anti}-correlated with stellar mass \citep{2012ApJS..201...15H,2015ApJ...798..112M}. This indicates a \textit{larger} amount of solids forming planets around low-mass stars \citep{2015ApJ...814..130M}, at least at short ($\lesssim ~1$ yr) orbital periods, in contrast with observed protoplanetary disk dust masses \citep{2013ApJ...773..168M,2013ApJ...771..129A,2016ApJ...827..142B,2016arXiv160405719A,2016arXiv160803621P}. The correlation between stellar metallicity and planet occurrence rate disappears towards lower mass planets, indicating that these planets can form around stars with a wide range of metallicities \citep{2008A&A...487..373S, 2012Natur.486..375B}. The large number of transiting planets with spectroscopically determined metallicities indicate only small rocky planets ($\lesssim ~1.7 R_\odot$) show no correlation with stellar metallicity, while larger mini-Neptunes ($1.7-3.9 R_\odot$) show a correlation with metallicity but one that is weaker than for giant planets \citep{2014Natur.509..593B,2015ApJ...808..187B}, however see \cite{2015ApJ...799L..26S}. A metallicity correlation for mini-Neptunes is also observed in a sample with measured planet masses \citep{2016MNRAS.461.1841C}. Another potential diagnostic of the planet formation process is the dependence of planet orbital period on host star metallicity. Different studies have pointed out underpopulated regions in the host star metallicity-orbital period diagram for small planets, at various orbital periods ranging from 5 to 70 days \citep{2013ApJ...763...12B,2013A&A...560A..51A,2015MNRAS.453.1471D,2016OLEB..tmp...30A}. In this paper, we revisit these results in the context of the exoplanet population. We use a dataset of over 20,000 medium-resolution spectroscopic metallicities for \textit{Kepler} target \textit{stars} from the \texttt{LAMOST}-\textit{Kepler} project \citep{2015ApJS..220...19D,2016arXiv160609149F}. This large dataset provides a homogeneous planet survey with a well-characterized detection bias, enabling us to estimate survey completeness. \textit{For the first time, we are able to calculate planet occurrence rates based on spectroscopically determined stellar metallicities.} We describe the target sample and methodology in section \ref{s:analysis} and present the main results of the metallicity dependence of the planet population on orbital period in section \ref{s:results}. We evaluate a potential trend towards the habitable zone in \S \ref{s:HZ}, and discuss potential origins for the excess of hot rocky planets around high-metallicity stars in section \ref{s:D}. We summarize our results and present and outlook for future research in section \ref{s:C}. \begin{table} \title{Planet Occurrence Rates and Host Star Metallicities} \centering \begin{tabular}{l l l l l | l}\hline\hline KOI & $R_P$ & $P$ & $f_{\rm occ}$ & \FeH & $f_{\rm occ}$ \\ & [cm] & [day] & & [dex] & \\ \hline K00001.01 & 7.9e+09 & 2.5 & 2.9e-04 & 0.28 & 5.6e-04 \\ K00005.01 & 3.8e+09 & 4.8 & 4.8e-04 & 0.36 & 9.1e-04 \\ ... & ... & ... & ... & ... & ... \\ K06242.03 & 8.9e+08 & 78.9 & 6.5e-03 & -0.41 & 1.5e-02 \\ K06246.01 & 1.0e+09 & 9.1 & 8.3e-04 & 0.32 & 1.6e-03 \\ \hline\hline\end{tabular} \par % \caption{Planet Occurrence Rates and Host Star Metallicities. The final column denotes the occurrence rate of the planet in the super-solar ($\FeH \ge 0$) or sub-solar ($\FeH < 0$) metallicity sample. Table \ref{t:occ} is published in its entirety in the electronic edition of the Astrophysical Journal. A portion is shown here for guidance regarding its form and content.} \label{t:occ} \end{table} | \label{s:C} We have characterized the orbital-period dependence of the \textit{Kepler} exoplanet population using $>20,000$ medium-resolution spectroscopic metallicities from the \texttt{LAMOST} survey for main-sequence G and F stars. For the first time we are able to calculate planet occurrence rates for the \textit{Kepler} sample based on spectroscopic metallicities. We find that: \begin{itemize} \item The metallicities of the stellar sample is consistent with solar, while giant planets have an increased host star metallicity of $0.14\pm0.04$ dex. \item The exoplanet population, which is dominated by planets smaller than $\sim 4$ Earth radii, shows an increased host star metallicity of [Fe/H]$\simeq{} 0.15\pm0.05$ dex) interior to a 10-day orbital period. At longer orbital periods metallicities are consistent with solar. \item The super-solar metallicity at short orbital periods is most significant for rocky planets ($R_P<1.7 ~R_\oplus$), where metallicity differs by [Fe/H]$\simeq 0.25\pm0.07$ dex. The difference in metallicity for hot mini-Neptunes ($R_P=1.7-3.9 ~R_\oplus$) is smaller and less significant at [Fe/H]$\simeq 0.08\pm0.05$ dex. Hot gas-giants ($R_P>3.9 ~R_\oplus$) do not show a significant metallicity variation with orbital period ([Fe/H]$\simeq 0.1\pm0.12$ dex). \item The occurrence rate of planets interior to a 10-day orbital period is almost three times higher for super-solar metallicity stars ([Fe/H]$\ge 0$, $29.6 \pm 2.0 \%$) than for stars with a sub-solar metallicity ([Fe/H]$<0$ , $11.9 \pm 1.4 \%$). Exterior to 10-day orbital periods, there is no significant difference between planet occurrence rates around stars of super-solar metallicity ($89.0 \pm 6.5 \%$) and sub-solar metallicity ($89.9 \pm 7.6 \%$). \item The increased host star metallicity of hot rocky planets suggests they may share a formation mechanism with hot-Jupiters that is distinct from the population of rocky planets and mini-Neptunes at orbital periods that shows no metallicity dependence. Alternatively, planet formation regions may extend closer-in around stars with higher metallicity, which is supported by hot rocky planets also appearing in multi-planets systems, in contrast to hot Jupiters that are typically single. \item We do not find statistically significant evidence for a trend previously identified by \cite{2016OLEB..tmp...30A} that small planets toward the habitable zone are preferentially found around low-metallicity stars. Although the occurrence rates of planets smaller than 2 Earth radii with orbital periods between 70 and 200 days are higher for stars with a sub-solar metallicity ($26.7 \pm 7.1 \%$) compared to stars of super-solar metallicity ($13.8 \pm 4.6 \%$) , this difference is only $1.5\sigma$. \end{itemize} Although the \textit{Kepler} spacecraft has finished its main mission, ongoing characterization of the stellar content of the \textit{Kepler} field with surveys like \texttt{LAMOST} and in the future \texttt{GAIA} can shed new light on the planet formation process. In particular trends for exoplanets in the habitable zone, where the detection efficiency is low, can be refuted or placed on a strong statistical footing with a larger dataset of spectroscopic metallicities. \vspace{1em}{\it Acknowledgments:}\\ This paper includes data collected by the \textit{Kepler} mission. Funding for the \textit{Kepler} mission is provided by the NASA Science Mission directorate. We thank the referee and the statistical editor for a constructive review that has improved the quality of the paper. The authors thank Mario Flock and Aline Vidotto for helpful discussions on the disk inner edge. We would also like to thank Dean Billheimer and his students for statistical advice through the Statistical Consulting course. JM-\.Z acknowledges the grant number NCN 2014/13/B/ST9/00902. This material is based upon work supported by the National Aeronautics and Space Administration under Agreement No. NNX15AD94G for the program “Earths in Other Solar Systems”. The results reported herein benefited from collaborations and/or information exchange within NASA’s Nexus for Exoplanet System Science (NExSS) research coordination network sponsored by NASA’s Science Mission Directorate. \ifastroph | 16 | 9 | 1609.05898 |
1609 | 1609.04470_arXiv.txt | Conventional Type Ia supernova (SN Ia) cosmology analyses currently use a simplistic linear regression of magnitude versus color and light curve shape, which does not model intrinsic SN Ia variations and host galaxy dust as physically distinct effects, resulting in low color-magnitude slopes. We construct a probabilistic generative model for the dusty distribution of extinguished absolute magnitudes and apparent colors as the convolution of a intrinsic SN Ia color-magnitude distribution and a host galaxy dust reddening-extinction distribution. If the intrinsic color-magnitude ($M_B$ vs. $B-V$) slope $\beta_\text{int}$ differs from the host galaxy dust law $R_B$, this convolution results in a specific curve of mean extinguished absolute magnitude vs. apparent color. The derivative of this curve smoothly transitions from $\beta_\text{int}$ in the blue tail to $R_B$ in the red tail of the apparent color distribution. The conventional linear fit approximates this effective curve near the average apparent color, resulting in an apparent slope $\beta_\text{app}$ between $\beta_\text{int}$ and $R_B$. We incorporate these effects into a hierarchical Bayesian statistical model for SN Ia light curve measurements, and analyze a dataset of SALT2 optical light curve fits of 248 nearby SN Ia at $z < 0.10$. The conventional linear fit obtains $\beta_\text{app} \approx 3$. Our model finds a $\beta_\text{int} = 2.3 \pm 0.3$ and a distinct dust law of $R_B = 3.8 \pm 0.3$, consistent with the average for Milky Way dust, while correcting a systematic distance bias of $\sim 0.10$ mag in the tails of the apparent color distribution. Finally, we extend our model to examine the SN Ia luminosity-host mass dependence in terms of intrinsic and dust components. | Type Ia supernova (SN Ia) rest-frame optical light curves have been used as cosmological distance indicators to trace the history of cosmic expansion, detect cosmic acceleration \citep{riess98,perlmutter99}, and to constrain the equation-of-state parameter $w$ of dark energy \citep{garnavich98b,wood-vasey07, astier06, kowalski08, hicken09b, kessler09, freedman09, amanullah10,conley11,sullivan11,rest14,scolnic14b, betoule14}. Determining supernova distances with high precision and small systematic error is essential to accurate constraints on the cosmic expansion history and the properties of dark energy. Inferring peak optical absolute magnitudes of SN Ia from distance-independent measures such as their light curve shapes and colors underpins the evidence for cosmic acceleration. Empirical studies show that SN Ia with broader, slower declining optical light curves are more luminous (``broader-brighter'') and that SN Ia with redder colors are dimmer. But the ``redder-dimmer'' color-luminosity relation widely used in cosmological SN Ia analyses masks the fact that it has two separate physical origins. An intrinsic correlation arises from the physics of exploding white dwarfs while interstellar dust in the host galaxy also makes supernovae appear dimmer and redder. The conventional approach of fitting a single (usually linear) function for luminosity vs. color, for a given light curve shape, is too simple. This leads to considerable uncertainty regarding the physical interpretation of the color-luminosity distribution of SN Ia, the confounding of extrinsic host galaxy dust reddening with the intrinsic color variations of SN Ia, and the proper way to use SN Ia color measurements to estimate accurate photometric distances. In this paper, we present a new probabilistic model describing the apparent SN Ia color-magnitude distribution as arising from the combination of intrinsic color-luminosity variations and host galaxy dust reddening and extinction, and apply this to SN Ia data to determine the characteristics of these physical components. Cosmological analyses of high-$z$ SN Ia data depend on empirical correlations originally observed in samples of nearby low-$z$ SN Ia \citep{hamuy96_29sne, riess99, jha06, hicken09a, contreras10, stritzinger11, hicken12}. Light curve fitting methods, including MLCS \citep{riess96, riess98, jha07}, SALT2 \citep{guy07, guy10}, SNooPy \citep{burns11}, and \textsc{BayeSN} \citep{mandel11}, all make use of the optical luminosity-light curve width correlation \citep{phillips93,hamuy96, phillips99}. However, current approaches conceptually differ on how measured apparent colors are used to infer the SN Ia luminosities and thus estimate the photometric distance. Methods such as MLCS, SNooPy, and \textsc{BayeSN} explicitly model the intrinsic SN Ia light curves and the effects of host galaxy dust extinction as separate components. However, the most popular tool for fitting cosmological SN Ia light curves is currently SALT2, a spectral template model that does not attempt to separate intrinsic SN Ia variations from host galaxy dust effects. A longstanding puzzle in the analysis of SN Ia light curves is the nature of their apparent color and brightness variations. In principle, they comprise color and luminosity variations intrinsic to the supernovae, as well as reddening and extinction by interstellar dust along the line of sight in their host galaxies. However, the fact that astronomers only observe the combination of these effects poses a challenging inference problem. The function of dust absorption over wavelength \citep[e.g. CCM,][]{ccm89} is typically parameterized by the ratio of total to selective extinction, $R_V = A_V/(A_B - A_V)$. This ratio normally has an average value of 3.1 for interstellar dust in the Milky Way (MW) Galaxy, although it can vary between 2.1 and 5.8 \citep{draine03}. \citet{schlafly16} find that the MW extinction curve is fairly uniform: with a narrow spread $\sigma(R_V) \approx 0.18$. Similar extinction curves have been found in external galaxies; for example, \citet{finkelman08, finkelman10} found average values of $R_V \approx 2.8$. Early analyses of SN Ia data estimated unphysically low values of $R_V \lesssim 1$ \citep[c.f.][for a review]{branch92}, although these analyses did not take into account empirical correlations between the luminosity, color and light curve shape of the events. \citet*{riess96b} used the first MLCS \citep*{riess96} method to fit SN Ia optical $BVRI$ light curves and, by minimizing the Hubble diagram scatter, they found a dust $R_V = 2.6 \pm 0.3$, consistent with the Milky Way average. They noted that the failure to properly account for intrinsic color-luminosity correlations would cause estimates of the dust extinction-reddening ratio $R_V$ to be biased low. Extending the wavelength range of the observations from the optical to the rest-frame NIR improves the constraints on the dust law and thus helps disentangle intrinsic color variation from dust reddening. For several nearby, highly reddened (peak apparent $B-V \gg 0.6$) SN Ia with optical and NIR light curves, $R_V$ can be fit precisely and unusually low values of 1.5-1.8 have been reported \citep{krisciunas07, elias-rosa06, elias-rosa07, wangx08}. While the origin of these apparently low $R_V$ values is poorly understood \citep{lwang05, goobar08, phillips13,johansson14, amanullah15}, these extremely red and dim objects are not present in the cosmological sample due to selection effects and cuts ($B-V < 0.3$). \citet{freedman09} constructed a SN Ia Hubble diagram using rest-frame $i$-band magnitudes and $B-V$ colors. By minimizing the Hubble residuals $\chi^2$, they estimated $R_V \approx 1.74 \pm 0.27$, and suggested that either dust in SN Ia host galaxies has a substantially different extinction law than Milky Way dust, or that there is significant SN Ia intrinsic color-luminosity dispersion, independent of the light curve shape. The latter hypothesis was supported by \citet{folatelli10} to explain a discrepancy found when analyzing of the nearby Carnegie Supernova Project \citep[CSP,][]{contreras10} SN Ia sample. When examining the optical-NIR colors they inferred $R_V = 3.2 \pm 0.4$ when the extremely red objects ($E(B-V) \gtrsim 1$) were excluded. However, when minimizing the Hubble diagram dispersion, they find low values of $R_V \approx 1-2$. Recent analyses of larger optical-NIR nearby samples found that the majority of SN Ia Ia with low reddening appear extinguished by dust with $R_V$ closer to $3$ \citep{mandel11, phillips12, burns14}. Another strategy for separating intrinsic color variation from dust effects is to measure spectral features that correlate with the intrinsic photometric properties of SN Ia. For example, \citet{foleykasen11} have correlated the velocity of the Si II $\lambda$6355 absorption feature with the peak intrinsic $B-V$ color, and found $R_V \approx 2.5$. As they controlled for optical decline rate, this is evidence for independent intrinsic color variations. \citet{chotard11} modeled the components of the apparent SN Ia spectroscopic and photometric variations depending upon Si II and Ca II H\&K equivalent widths, finding that the remainder is well described by a CCM dust reddening law with $R_V = 2.8 \pm 0.3$, consistent with the MW value. However, the vast majority of the high-$z$ SN Ia observations currently used for cosmological analysis consist of rest-frame optical photometry and lack the high-quality spectra or rest-frame NIR data used by the above analyses. In the following subsections, we describe the conventional method, the Tripp formula, for modeling correlations between SN Ia magnitude, color and light curve shapes from optical light curve data, currently used to estimate cosmological distances. \textsc{SALT2mu} is a generalization of this method to account for scatter around the Tripp formula \citep{marriner11}. We comment on the drawbacks of these approaches, primarily because they are inadequate for properly accounting for the physically distinct factors of intrinsic SN Ia variation and host galaxy dust underlying the data. We introduce a new statistical model, \textsc{Simple-BayeSN}, that we have developed to address these shortcomings. \textsc{Simple-BayeSN} analyzes the peak apparent magnitude, apparent color, and light curve shape obtained from light curve fits to the SN Ia photometric time series. It models the SN Ia data as arising from a probabilistic generative process combining intrinsic SN Ia variations, host galaxy dust effects, and measurement error. \textsc{Simple-BayeSN} uses a hierarchical Bayesian framework to fit the SN Ia data on the Hubble Diagram, while coherently estimating the parameters driving the underlying effects. \subsection{The Tripp Formula} Conventional cosmological SN Ia analysis \citep[e.g.][]{betoule14, rest14} currently proceeds by fitting primarily rest-frame optical light curve data to obtain estimates of peak apparent magnitude $m_s$, apparent color $c_s$ and light curve shape $x_s$ for each SN $s$. A simple linear regression model for the absolute magnitude $M_s$ as a function of the distance-independent light curve observables is constructed using the Tripp formula\footnote{This formula is often written with an arbitrary negative sign preceding $\alpha$ \citep[e.g.][]{guy05, astier06}. When it is regarded as a linear regression model for absolute magnitude versus the covariates $x$ and $c$, it is most natural to precede all the regression coefficients by a positive sign.}: \begin{equation}\label{eqn:tripp1} M_s = m_s - \mu_s = M_0 + \alpha \times x_s + \beta \times c_s + \epsilon_\text{res}^s \end{equation} \citep{tripp98}, where $\mu_s$ is the supernova distance modulus. The global coefficients $(M_0, \alpha, \beta)$ are found by fitting this relation with measurements $(\hat{m}_s, \hat{c}_s, \hat{x}_s)$ on the Hubble diagram, $\mu_s = \mu(z_s)$, for a sample of SN Ia $\{s\}$. The optical ``broader-brighter'' width-luminosity relation \citep{phillips93} is captured by $\alpha$, the ``redder-dimmer'' color-luminosity relation by $\beta$. The expected absolute magnitude at $x_s = c_s = 0$ is $M_0$. This equation would be correct if the color-luminosity relation were entirely due to (small amounts) of dust (and the light curve shape dependence truly linear). In that case $m_s - \beta c_s$ is the reddening-free Wesenheit magnitude \citep{madore1982}. However, in the presence of intrinsic color-luminosity variations, this formula is not fundamental: it is just the simplest linear model for absolute magnitude as a function of the observables, light curve shape and apparent color. The residual scatter\footnote{\label{footnote1}This variance term has been variously called the \emph{intrinsic} scatter or dispersion $\sigma_\text{int}$ \citep[e.g.][]{astier06, conley11, marriner11}. However, in their conventional usage, there is no implication that this scatter is solely attributed to physical properties intrinsic to the SN Ia with host galaxy dust subtracted. \citet{scolnic14} instead refers to it as \emph{residual} scatter: it is the additional variance needed to account for the scatter in the Hubble \emph{residuals}. We adopt the latter usage to avoid confusion, and reserve \emph{intrinsic} within our model to conceptually refer to the latent properties of the SN Ia in the absence of host galaxy dust.} around this model that is unaccounted for by measurement error or peculiar velocities is $\epsilon_\text{res}^s$ with variance $\sigma^2_\text{res}$. This contributes to the uncertainty in absolute magnitude $M_s$, and thus the photometric distance modulus $\mu_s$, of an individual SN. In typical usage, $M_s$ and $m_s$ are peak absolute and apparent magnitudes effectively in rest-frame $B$-band, and the color corresponds to peak apparent $B-V$. Hence $\beta$ is the slope of the change in $B$-magnitude for a unit change in $B-V$. This can be compared to the expected slope for normal Milky Way dust extinction $A_B$ vs. $E(B-V)$ reddening, $R_B \equiv R_V + 1 = 4.1$. \citet{tripp98} and \citet{trippbranch99} originally found $\beta \approx 2$ using peak apparent $B-V$ colors and $\Delta m_{15}(B)$ for light curve shape \citep{phillips93}. Using the first SALT model to determine optical light curve stretch and color, \citet{guy05} and \citet{astier06} also found low values $\beta \approx 1.5-2$. \citet{conley07} fit nearby SN Ia on the Hubble diagram and found that the empirical relation between SN Ia optical luminosity and apparent color, controlling for light curve shape, still required a low value of $\beta \approx 2$. They speculated that this is much less than the normal dust $R_B \approx 4$ either because the dust in SN Ia hosts is nonstandard, or because the estimated $\beta$ may actually be measuring some combination of intrinsic color variations (not accounted for by light curve shape $x_s$) and normal interstellar dust. The SALT2 spectral template \citep{guy07,guy10} is the most popular \citep[and well-tested,][]{mosher14} model currently applied to fit cosmological SN Ia light curve data. Using SALT2 and the Tripp formula to fit SNLS1 and SDSS-II SN Ia data, \citet{guy07} and \citet{kessler09} found $\beta \approx 1.8-2.6$. For SNLS3, \citet{guy10} and \citet{conley11} found $\beta \approx 3.1$. Similarly, the combined SDSS+SNLS3 [JLA] analysis of \citet{betoule14} obtained $\beta = 3.10 \pm 0.08$, significantly less than $R_B = 4.1$. \subsection{Luminosity vs. Color Residual Scatter} \citet{marriner11} introduced a more general formalism (\textsc{SALT2mu}) for accounting for the residual scatter around the linear model, Eq.\ref{eqn:tripp1}, and fitting for its regression coefficients. Sensible methods for fitting Eq. \ref{eqn:tripp1} take into account the fact that the fitted values $(\hat{m}_s, \hat{c}_s, \hat{x}_s)$ are different from the true, latent (unobserved) values $(m_s, c_s, x_s)$ that obey Eq. \ref{eqn:tripp1}. The difference amounts to random measurement error with the SALT2 fit covariance matrix. \citet{marriner11} further supposes an additional source of residual scatter between the measured values and the latent values. For example, the measured color of a SN $s$ is decomposed as \begin{equation}\label{eqn:salt2mu} \hat{c}_s = c_\text{mod}^s + c_r^s + c_n^s, \end{equation} where $c_\text{mod}^s$ is the ``model'' color component that enters into the Tripp \emph{model} (Eq. \ref{eqn:tripp1}), and is linearly correlated with the luminosity, $c_n^s$ is the color measurement error ``noise'' and $c_r^s$ is a random ``residual'' color scatter term \citep{scolnic14}. Similar equations can be written for the scatter in the magnitude and light curve shape components of the data. While the measurement errors are quantified by a covariance matrix estimated from the SALT2 light curve fit, the residual scatter covariance matrix $\bm{\Sigma}_r$ is a priori unknown and poorly constrained by the data, and some choices must be made regarding its entries. To date, it has been used generally with non-zero entries only for residual magnitude $\sigma_{m_r}^2$ and color $\sigma_{c_r}^2$ variances. The special case in which only the residual magnitude variance entry is non-zero corresponds to the conventional assumption that the residual scatter is attributed to unexplained variance in luminosity, $\sigma_\text{res}^2 = \sigma_{m_r}^2$. With an assumed residual matrix, \textsc{SALT2mu} estimates the $M_0$, $\alpha$, $\beta$ coefficients by minimizing the Hubble residual $\chi^2$, modified to incorporate measurement errors and the residual scatter matrix. Applying \textsc{SALT2mu} to SDSS-II data, \citet{marriner11} find that attributing the residual scatter only to color led to a larger estimated $\beta \approx 3.2$, still significantly less than $R_B = 4.1$. \citet{scolnic14} analyzed a combined dataset, consisting of SDSS-II, SNLS3 and nearby samples, to examine the dependence of the estimated $\beta$ on the relative attribution of residual scatter to luminosity or color when analyzing the data within the \textsc{SALT2mu} framework. When ``luminosity variation'' is assumed ($\sigma_{m_r} \gg \sigma_{c_r}$), the estimated $\beta \approx 3.2$ but when ``color variation'' is assumed ($\sigma_{m_r} \ll \sigma_{c_r}$), it increases to $\beta \approx 3.7$, closer to the normal MW average. Simulations from a color variation model, in which $c_\text{mod}$ had a ``dust-like'' distribution (with a ``one-sided'' tail to the red, but a sharp edge to the blue), with a MW dust-like true $\beta = 4.1$, could better match the pattern of Hubble residuals vs. color seen in the data, compared to a luminosity variation simulation with a conventional $\beta = 3.1$. They estimated that a misattribution of the residual scatter to luminosity rather than color variation, could lead to a bias $\Delta \beta \approx -1$ in the recovered slope, and a $4\%$ shift in the inferred $w$. Recently, \citet{scolnic16} presented a method to determine the underlying distribution of the latent $c_\text{mod}$ colors by matching the observed data distribution to realistic forward simulations that incorporate measurement noise, and selection effects, and two SALT2 spectral variation models: luminosity-variation dominated \citep[G10,][]{guy10} and color-variation dominated \citep[C11, based on][]{chotard11}. Applying this to the cosmological SN Ia compilation of \citet{scolnic15}, they uncover a ``dust-like'' underlying color distribution, with a red tail and blue edge, when simulating with a color-dominated C11 model. By matching their simulations to data, they find $\beta = 3.85$ using the color-variation model, and $\beta = 3.1$ with the luminosity-variation model. These analyses suggest that the proper modeling of color-luminosity subcomponents is important for understanding the observed SN Ia color-magnitude distribution, the accurate estimation of distances, and inferences for cosmology. \subsection{Shortcomings of these approaches} The simplicity of the Tripp formula has enabled its widespread application in cosmological SN Ia analyses. However, in its conventional usage, Eq. \ref{eqn:tripp1} is too simplistic. Because neither the magnitude $m_s$ nor the color $c_s$ that enter into it are corrected for host galaxy dust extinction or reddening, the absolute magnitude $M_s = m_s - \mu_s$ is actually the dust-extinguished absolute magnitude $M^\text{ext}_s$, and $c_s$ is the dust-reddened apparent color $c^\text{app}_s$. However, in reality, the extinguished absolute magnitude results from the dimming of the supernova's intrinsic luminosity by dust extinction $M^\text{ext}_s = M^\text{int}_s + A_B^s$. The apparent color results from the dust reddening the supernova's intrinsic color: $c^\text{app}_s = c^\text{int}_s + E(B-V)_s$. The dust extinction is surely correlated with its reddening, and can only be positive. The supernova's intrinsic luminosity may be correlated with its intrinsic color, independently of light curve shape, as speculated by e.g. \citet{conley07} and \citet{freedman09}; at the very least we do not know that this intrinsic correlation is zero, and would like to estimate it. By regressing only the sum $M^\text{ext}_s$ against the sum $c^\text{app}_s$, the Tripp formula tries to capture all the color-magnitude correlations in a single trend with slope $\beta$. As it is a priori highly unlikely that the intrinsic color-magnitude slope would be exactly equal to the dust reddening-extinction law, this parameterization is clearly limited and inadequate. It fails to distinguish between the different physical characteristics of color-luminosity variation intrinsic to the SN Ia vs. reddening-extinction by extrinsic host galaxy dust. The residual matrix framework of \textsc{SALT2mu} adds some additional degrees of freedom to the conventional Tripp formula. By decomposing the apparent color $c_s$ into a model color $c_\text{mod}$ and residual color $c_\text{r}$, and allowing $\sigma_{c_r} > 0$, one can in effect capture additional color-magnitude variations. However, this residual color scatter does not correlate with luminosity; this additional component essentially has its own $\beta_r = 0$. Hence, the residual color scatter by itself would not be able to capture a separate non-zero color-luminosity correlation different from $\beta$. This suggests that the color-magnitude covariance entry in the residual matrix $\bm{\Sigma}_r$ could be made non-zero. This raises two challenges: either a numerical value for this residual covariance would have to be set a priori, or it would have to be inferred jointly with the other parameters. Unfortunately, in the former case, it is unclear what value to set it to a priori; in the latter case, the inference of this covariance would likely be highly degenerate with $\beta$, unless strong priors were set. Systematic uncertainties in the treatment of dust and color of SN Ia Ia have important implications for cosmological inference. If the single slope $\beta$ is actually measuring a combination of intrinsic color-luminosity variation and host galaxy dust reddening-extinction, then different SN Ia subsamples may have different proportions of each. The proportions may even be redshift-dependent, owing to the physical environment of the progenitor systems, host galaxies, or selection effects. Applying one $\beta$ slope across the entire sample would incur complex systematic biases that propagate into cosmological inferences. Resolving the confusion between the intrinsic variation and extrinsic host galaxy dust effects is imperative for the proper analysis of SN Ia observables. \subsection{Simple-BayeSN} To remedy the aforementioned shortcomings of the conventional methodology, we propose a new statistical model, \textsc{Simple-BayeSN}, describing the observed color-magnitude distribution as arising from the probabilistic combination of an intrinsic SN Ia color-magnitude variations and host galaxy dust reddening and extinction. The host galaxy dust is given a physically-motivated distribution, allowing for only positive extinction. The intrinsic color-magnitude slope $\beta_\text{int}$ can be different from the reddening-extinction slope $R_B$. The observed data arises from the combination of these effects with measurement error. By fitting this statistical model to the light curve data, the separate physical characteristics of the intrinsic and dust distributions are coherently inferred. As our model uses the same SN Ia light curve measurements that are conventionally used by the Tripp formula, it can be readily applied to current cosmological SN Ia datasets. We adopt a hierarchical Bayesian, or multi-level modeling, framework to build a structured probability model conceptually describing the multiple random effects that underlie the observed SN Ia. This principled strategy enables us to coherently model and make probabilistic inferences at both the level of an ensemble or population of objects as well as at the level of the constituent individuals \citep{gelman_bda, loredohendry10, loredo12}. Inference with the hierarchical model may be regarded as a probabilistic deconvolution of the observed SN data into the multiple, unobserved, latent random effects generating it \citep{mandel_scmav}. Recent astrophysical and cosmological applications of hierarchical Bayesian modeling include \citet{foster13, brewer14, sanders14, mandel14, dfm14, schneider15, alsing16} and \citet{wolfgang16}. Hierarchical Bayesian statistical modeling was first applied to SN Ia analysis by \citet{mandel09,mandel11}, who constructed the \textsc{BayeSN} model for optical and NIR SN Ia light curves \citep{mandelthesis}. Fundamentally, \textsc{BayeSN} models the photometric time series observations of SN Ia as arising from intrinsic light curves and a host galaxy dust extinction across optical and NIR wavelengths. The training process of \textsc{BayeSN} learns the covariance structure of the intrinsic SN Ia light curve distribution across phase and wavelength, as well as the characteristics of the dust distribution and dust law $R_V$. The latent variables for each SN, and the hyperparameters of the population distributions are inferred coherently from the joint posterior density conditional on the set of light curve data. \citet{mandel11} demonstrated with \textsc{BayeSN} that the combination of optical and NIR observations could significantly improve constraints on SN Ia dust and the precision of photometric distances. The \textsc{Simple-BayeSN} approach distills the core concept of \textsc{BayeSN}: hierarchically modeling the SN Ia data as a combination of intrinsic SN Ia variations, dust effects, and measurement error. However, \textsc{BayeSN} is a complex framework that directly and non-parametrically models the multi-wavelength photometric time series observations. As a simplification, \textsc{Simple-BayeSN} instead employs the outputs from an external light curve model that fits the photometric time series data of individual SN Ia and estimates the three parameters used in conventional analyses: peak apparent magnitude, apparent color, and light curve shape. In this paper, we employ the widely-used SALT2 light curve model, but \textsc{Simple-BayeSN} can generally work with the fit parameters from any external model for apparent SN Ia light curve data. \citet{march11} constructed the first hierarchical Bayesian model for fitting the cosmological SN Ia Hubble diagram with the SALT2 optical light curve fit parameters. They encapsulated the Tripp formula in a hierarchical linear regression with population distributions for light curve shape and apparent color. The regression coefficients and the population hyperparameters are inferred simultaneously with cosmological parameters in the posterior distribution. This concept was further extended recently by \citet[UNITY,][]{rubin15} and \citet[BAHAMAS,][]{shariff16}. However, these Bayesian models inherit the same fundamental, conceptual limitations of the Tripp formula, because they are essentially still regressing extinguished absolute magnitudes directly against apparent light curve parameters, rather than modeling the constituent, and physically distinct, intrinsic and dust components underlying the data. This paper is structured as follows. In \S \ref{sec:motivation}, we construct a probabilistic generative model for the distribution of extinguished absolute magnitudes and apparent colors as a convolution of the intrinsic SN Ia color-magnitude distribution and the host galaxy dust distribution. We describe the features and generic implications of this model. In \S \ref{sec:model}, we encapsulate this generative model in a hierarchical Bayesian framework, \textsc{Simple-BayeSN}, for analyzing SN Ia magnitudes, colors and light curve shape measurements. We demonstrate inference of the parameters of this model from the SN Ia data via maximum likelihood and Gibbs sampling. In \S \ref{sec:application} we apply this model to analyze a data set of SALT-II parameters for a sample of 248 nearby SN Ia ($z < 0.10$). In \S \ref{sec:hostmass}, we demonstrate how the host galaxy stellar mass dependence \citep{pkelly10} can be included in this new framework. We discuss our results in \S \ref{sec:discussion} and conclude in \S \ref{sec:conclusion}. Mathematical and computational details about our methods are described in Appendices \S \ref{sec:app:dustydistr}, \ref{sec:bayesian}, \ref{sec:algorithm}, \& \ref{sec:fit_tripp}. \vfill\break | 16 | 9 | 1609.04470 |
|
1609 | 1609.01125_arXiv.txt | We report the detection of a GeV $\gamma$-ray source which is spatially overlapping and thus very likely associated with the unidentified very-high-energy (VHE) $\gamma$-ray source HESS J1427-608 with the Pass 8 data recorded by the Fermi Large Area Telescope. The photon spectrum of this source is best described by a power-law with an index of $1.85\pm0.17$ in the energy range of $3-500$ GeV, and the measured flux connects smoothly with that of HESS J1427-608 at a few hundred GeV. This source shows no significant extension and time variation. The broadband GeV-TeV emission over four decades of energies can be well fitted by a single power-law function with an index of 2.0, without obvious indication of spectral cutoff toward high energies. Such a result implies that HESS J1427-608 may be a PeV particle accelerator. We discuss possible nature of HESS J1427-608 according to the multi-wavelength spectral fittings. Given the relatively large errors, either a leptonic or a hadronic model can explain the multi-wavelength data from radio to VHE $\gamma$-rays. The inferred magnetic field strength is a few $\mu$G, which is smaller than typical values of supernova remnants (SNRs), and is consistent with some pulsar wind nebulae (PWNe). On the other hand, the flat $\gamma$-ray spectrum is slightly different from typical PWNe but similar to that of some known SNRs. | The High Energy Stereoscopic System (HESS) survey of the inner Galaxy \citep{Aharonian2006a} has discovered a large number of very high energy (VHE; $>$ 100 GeV) sources. About half of them have been firmly identified as supernova remnants (SNRs) and pulsar wind nebulae (PWNe). However, there is still a significant fraction of VHE sources that has not been identified due to the lack of counterparts in the radio/X-ray or GeV $\gamma$-ray band \citep{Aharonian2005,Aharonian2008a}. Therefore, searching for the counterparts of these unidentified VHE sources in longer wavelength bands is very helpful to identify and classify them. In the GeV band, more than seven years survey data from the Fermi Large Area Telescope (Fermi-LAT) provides a good opportunity to hunt for the unidentified VHE sources. Indeed, several unidentified VHE sources have been detected with the Fermi-LAT data \citep{Hui2016}. HESS J1427-608 is one of the unidentified VHE sources discovered by the HESS Galactic Plane Survey \citep{Aharonian2008b}. The TeV image shows that it is only slightly extended with a symmetric Gaussian function with width of $\sigma$ = $3'$. This source is detected with an integral flux above 1 TeV of $F_{\gamma}(>1 {\rm TeV})$ = $4.0 \times 10^{-12} $ erg cm$^{-2}$ s$^{-1}$ and a power-law spectrum with an index of $\Gamma$ = 2.16. No nearby SNR or pulsar was detected in the vicinity of HESS J1427-608 \citep{Green2014}. The X-ray counterpart of it, Suzaku J1427-6051, was reported by \citet{Fujinaga2013}. The X-ray spatial morphology is clearly extended with a Gaussian width of $\sigma$ = $0.9' \pm 0.1'$ and the spectrum is well fitted by a power-law with an index of $\Gamma$ = $3.1^{+0.6}_{-0.5}$, implying non-thermal X-ray continuum emission. The unabsorbed flux in the 2-10 keV band is $F_{\rm X}=(9^{+4}_{-2}) \times 10^{-13}~{\rm erg}~{\rm cm}^{-2}~{\rm s}^{-1}$. \citet{Fujinaga2013} has discussed the possible nature of HESS J1427-608 as a PWN or non-thermal SNR. Each one can explain some of the observational facts but face difficulties. No radio counterpart was detected in the direction of HESS J1427-608. In the second Fermi-LAT catalog \citep[2FGL;][]{Nolan2012}, a GeV source, 2FGL J1427.6-6048c, which is located in the vicinity of HESS J1427-608, was reported. However, the central position of them differs moderately from each other. Furthermore, the power-law spectrum index of 2FGL J1427.6-6048c is $\Gamma$ = 2.7 \citep{Nolan2012}, which can not connect with the TeV spectrum of HESS J1427-608 well. It is very likely that 2FGL J1427.6-6048c is not associated with HESS J1427-608. Meanwhile, it should be noted that no source in the third Fermi-LAT source catalog \citep[3FGL;][]{Acero2015} is associated with 2FGL J1427.6-6048c. In this paper, we carry out a complete analysis of this region using more than 7 years Fermi-LAT Pass 8 data to investigate the $\gamma$-ray emission of HESS J1427-608. In section 2, we describe the data analysis and results, including the spatial, spectral and timing analysis. The possible nature of HESS J1427-608 based on the multi-wavelength spectral energy distribution (SED) fitting is discussed in section 3 and the conclusion of this work is presented in section 4. | We investigate more than seven years data of the Fermi-LAT in the direction of the unidentified TeV source HESS J1427-608. A point source is detected with a significance of $\sim5.5\sigma$. This GeV source is positionally coincident with HESS J1427-608, and the GeV spectrum can connect with the TeV spectrum smoothly. These facts suggest that this source is very likely the GeV counterpart of HESS J1427-608. We collect multi-wavelength data to constrain the radiation model of HESS J1427-608. Both the leptonic and hadronic scenarios can fit the SED. At present it is difficult to exclude either one considering the large uncertainties of the data. A remarkable spectral feature of this source is its $E^{-2}$ spectrum over four decades of energies without obvious cutoff. This is unique among all currently detected $\gamma$-ray sources. If CR protons are responsible for the $\gamma$-ray emission, the highest energy of protons should exceed a few hundred TeV and even PeV. This makes HESS J1427-608 a promising PeV CR accelerator (PeVatron). Unfortunately, the nature of HESS J1427-608 is not clear yet. Either a SNR or a PWN seems to be plausible. The low strength of magnetic field and the absence of shell structure in the X-ray band indicate that HESS J1427-608 may be a PWN rather than a SNR. However, its flat $\gamma$-ray spectrum in a wide energy range is more close to some SNRs but different from typical PWNe. Further multi-wavelength observations, in the radio and VHE $\gamma$-rays, are crucial to understand its nature ultimately. Especially the precise measurements of the $\gamma$-ray energy spectrum up to hundreds of TeV energies by, e.g., the CTA, will crucially test its PeVatron nature. | 16 | 9 | 1609.01125 |
1609 | 1609.06705_arXiv.txt | We present an analysis of the orientation effects in SDSS quasar composite spectra. In a previous work we have shown that the equivalent width $EW$ of the [OIII] $\lambda$5008\AA~ line is a reliable indicator of the inclination of the accretion disk. Here, we have selected a sample of $\sim$15,000 quasars from the SDSS $7^{th}$ Data Release and divided it in sub-samples with different values of \ew. We find inclination effects both on broad and narrow quasars emission lines, among which an increasing broadening from low to high EW for the broad lines and a decreasing importance of the blue component for the narrow lines. These effects are naturally explained with a variation of source inclination from nearly face-on to edge-on, confirming the goodness of \ew as an orientation indicator. Moreover, we suggest that orientation effects could explain, at least partially, the origin of the anticorrelation between [OIII] and FeII intensities, i.e. the well known Eigenvector 1. | The optical-UV emission of Active Galactic Nuclei (AGN) is ascribed to an accretion disk around a supermassive Black Hole (BH). Models developed for such a structure predict that, in order to be radiatively efficient, the disk must be optically thick and geometrically thin \citep{ShakuraSunyaev1973}. In this case the geometry of the emitting region imposes a disk continuum intensity that decreases with $\cos \theta$, $\theta$ being the angle between the disk axis and the observer line of sight, i.e. the source inclination angle. This fact can be hardly directly proven due to the difficulties in intrinsic continuum measurements. The simpler way to test the behaviour of the continuum as a function of the inclination angle is therefore a comparison between this angle-dependent continuum emission and an inclination-independent one. The [OIII] line at $5008$\AA, emitted by the Narrow Line Region (NLR) at hundreds of parsecs from the central black hole, has isotropic characteristics, at least if compared with the emissions coming from accretion disk and BLR, and is considered a good indicator of bolometric luminosity of AGN \citep{Mulchaey94, Heckman2004}. Since line emitting regions are optically thin to line radiation, isotropy depends on their dimensions, i.e. they have to be large enough not to be significantly obscured by opaque structures, such as the accretion disk, the dusty torus and possible nuclear dust lanes. \citet{Mulchaey94} find that the [OIII] emission is isotropic in Seyfert galaxies (but see \citet{Diamond-Stanic2009} and \citet{diSeregoAlighieri97} for different results). As a consequence, the observed \ew, i.e., the ratio between line and local (same wavelength) continuum intensities is expected to be a function of the inclination angle $\theta$. Moreover, the [OIII] line holds a fundamental role in the context of the \emph{Eigenvector1}, the set of spectral properties able to explain most of the variance in optical spectra of quasars \citep{BorosonGreen1992}. The Eigenvector1 is usually referred to as the anticorrelation between [OIII] and FeII intensities, but other spectral properties, such as the H$\beta$ FWHM, contribute to drive the variance of quasars spectra. In a previous work by our group (\citet{Risaliti2011}, hereafter R11) we have studied the distribution of the observed \ew\ for a flux limited sample of $\sim$7,300 quasars at redshift z$<0.8$, obtained from the SDSS DR5 quasar catalog \citep{Schneider2007}. The observed \ew\ distribution peaks at $\sim$10\AA$\;$ and appears to be dominated by the orientation effect at $EW>30$\AA, while at lower wavelengths it mainly resembles the intrinsic distribution of EW (in general, \ew\ is expected to depend on several geometrical and physical properties of the source, such as the dimension and shape of the NLR and the spectral characteristic of the continuum emission). The orientation signature consists in the presence, at high values, of a power-law tail ($\gamma=-3.5$) that can be reproduced only by inclination effects. On the other hand, an examination of the observed distribution of EWs of the main broad lines has revealed that in these cases the inclination effect is much weaker, if not totally absent. This result has been interpreted as a hint at a possible disk-like shape for the BLR: if the BLR and the accretion disk share the same geometrical anisotropy then the EWs of broad lines should not show any angle dependence. Moreover, in such a case, the width of the broad lines should exhibit a trend with \ew\, specifically a broadening towards high \ew\, corresponding to \emph{edge-on} sources; indeed the dependence of broad line linewidths in terms of orientation is well known observationally \citep{WillsBrowne1986}. As a preliminary test, in R11 we plotted the line width of H$\beta$ versus \ew\, and found that despite the expected large dispersion, the average widths of H$\beta$ are larger in quasars with higher \ew\, i.e. more edge-on, a fact which is difficult to explain with other scenarios. Based on these early results, the aim of this work is to search for \ew\-dependent (i.e. orientation-dependent) effects in quasars spectra. For this purpose we have analyzed a large sample of quasars ($\sim 12000$, approximately twice as many as in R11). We divided the sample in narrow bins of \ew\, and performed a detailed spectral analysis of the staked spectra for each interval. The paper is organised as follows: in Section \ref{Sample} we present our new sample and we verify the presence of the relations found in R11; in Section \ref{sec:data_analysis} we illustrate our stacking procedure, and the spectral analysis. In Section \ref{results} we report our results on the analysis of orientation signatures in both [OIII] $\lambda 5008$\AA$\;$ and broad lines (H$\beta$, H$\alpha$ and MgII) profiles, as well as a possible explanation of the \emph{Eigenvector 1} in terms of orientation. In Section \ref{discussion} we discuss our results. | \label{discussion} Our analysis of the EW distribution of [OIII] and of the stacked spectra of quasars with different \ew\ have revealed several relevant properties of the narrow and broad line regions: \begin{enumerate} \item The distribution of \ew\ is well reproduced by an intrinsic log-normal distribution convolved with a high-EW power law tail. The maximum and dispersion of the intrinsic distribution are EW$_{MAX}=1$\AA~ and $\sigma = 9$\AA. The exponent of the power law tail is $\Gamma = 3.5$. \item The distribution of EW of H$\beta$ is well represented by a Gaussian distribution with EW$_{MAX}=58$~\AA~ and $\sigma = 23$~\AA~ and a power law tail with $\Gamma \simeq 7$. \item The [OIII]/H$\beta$ distribution shows the same high end tail as \ew. \item The [OIII] line shows a blue tail whose intensity decreases moving from low to high \ew. The blueshift similarly decreases with EW. \item The width of broad lines increases moving from low to high \ew. \item FeII emission is prominent for low \ew\ and its intensity decreases moving to high EW([OIII]). \item Double peaked broad lines objects are more frequent for high EW([OIII]) with respect to low \ew. \end{enumerate} Here we discuss the physical consequences of our results. \begin{figure} \centering \includegraphics[scale=0.65]{figures/OIII_profile_py.pdf} \caption{[OIII] $\lambda 5008$\AA$\;$ profile for each \ew\ representative spectrum. Profiles are normalized to their peak values.} \label{OIII_profile} \end{figure} \subsection{Distribution of \ew\ and EW([H$\beta$])} The distribution of \ew\ is relevant in two respects: (1) the high-EW tail $\Gamma=3.5$ fully confirms the scenario of an isotropic emission of the [O III]~$\lambda$5008~\AA~ line, with intensity proportional to the illumination from the ionizing source, and a disc-like continuum emission; (2) the ratio between the width of the intrinsic distribution and its peak value is an estimate of the precision of the [O III] luminosity as an indicator of the bolometric luminosity (which is supposed to be dominated by the disk emission). From our results, we conclude that an estimate of the bolometric/disk luminosity based on the [O III] line has an uncertainty of a factor of $\sim$2. The possible reasons for the observed intrinsic dispersion of \ew\ are variations in the covering factor of the Narrow-Line Region clouds as seen from the disc and effects of the dispersion in the optical/UV spectral energy distribution (the emission of the [O III] line is expected to be proportional to the disk emission at the line {\em ionizing} frequency of [O III], i.e. $\sim50$~eV, while the continuum is measured at the {\em emission} frequency). These effects are discussed in Risaliti et al.~(2011). We note that in principle an increase in \ew\ could be due to dust reddening of the disc component. In this case, however, the effect of dust reddening should be seen also on continuum spectra, while we have shown in Section \ref{spectra_stacking} that our selection ensures that only blue objects are present (see also Tab. \ref{tab4}). The distribution of EW(H$\beta$) strongly suggests a disc-like emission of the line. The orientation effects found in the broad emission lines require the BLR to be not only flat, but also optically thick to these lines. This is likely to be the case for the H$\beta$ line: for densities and column densities typical of BLR clouds ($n>10^{9}-10^{10}$~cm$^{-3}$ and $N_{H}>10^{23}$~cm$^{-2}$) the optical depth of the Ly$\alpha$ line is expected to be higher than $10^{4}$, and the optical depth of the Balmer lines start to be significant when $\tau$(Ly$\alpha)$ is higher than a few hundred \citep{OsterbrockFerland2006}. Moreover, the distribution of R=[OIII]/H$\beta$ confirms the suggestion of a disk-like shape for the H$\beta$ emitting region. If the BLR geometry resembles that of the accretion disk then this ratio is a close version of the \ew\, the difference between the two being determined by the larger height scale of the disk of the BLR, probably caused by the presence of turbulence in the gas componing this structure. \begin{figure*} \centering \includegraphics[scale=0.9]{figures/v_OIII_OII_py.pdf} \caption{Velocity shifts for the main (left panel) and blue (right panel) [OIII] components with respect to [OII] velocity, representing the systemic velocity for the host galaxy.} \label{voiii-voii} \end{figure*} \begin{figure} \begin{center} \includegraphics[scale=0.4]{figures/double_ks.pdf} \caption{Cumulative distributions of \ew\ for the whole sample (black, continuous line), the ``unambiguous'' double-peaked quasars (red, dot-dashed), and the ``possible'' double peaked quasars (blue, dotted line). } \label{double} \end{center} \end{figure} \subsection{The [OIII] line profiles} The spectral results on the [OIII] emission line can be interpreted as a simple consequence of the increasing inclination of the sources going from low EWs to high EWs. In Fig. \ref{oiii_vs_oii} [OIII] and [OII] $3727.092, 3729.875$\AA$\;$ profiles are compared for each of the six stacks. The [OIII] profile shows a prominent blue tail decreasing toward high \ew\, while [OII] holds quite steady in all the bins. The [OIII] blue component is due to gas in outflow from the NLR towards a direction mostly perpendicular to the plane of the accretion disk. In this scenario, the angle between the outflow direction and the observer line of sight is the same as the disk inclination angle. The velocity component along the line of sight is therefore \begin{equation} v_{obs}=v_{outflow} \cos \theta \simeq v_{outflow} \frac{EW[OIII]_{INT}}{EW[OIII]_{OBS}}\;\;. \end{equation} The increase in the central velocity shift of the blue [OIII] component with respect to the systemic velocity of the host galaxies ($v_{[OIII]}-v_{[OII]}$) has the same explanation: the shift is more important when the object is face-on because we are observing the outflow exactly along the line of sight (Fig. \ref{voiii-voii} right panel) (the same result was found in \citet{Boroson2011}). A somewhat more surprising result is the measured blueshift in the [OIII] main component. This finding can be explained as an indirect consequence of orientation effects on the global [OIII] profiles used by \citet{Shen2011} for the estimates of the redshifts. Since we use redshifts from this reference, we are obtaining an inclination dependent systematic shift: more face-on sources have a more prominent blue tail, and so a bluer central $\lambda$ in the global profile. This bias is instead negligible in edge-on objects (Fig. \ref{voiii-voii} left panel). To give a more quantitative measurement of the [OIII] profile degree of asymmetry we evaluate an asymmetry index similar to that defined in \citet{Heckman1981} and based on differences between Inter-Percentile Velocities (IPV); the asymmetry index is defined as \begin{equation} A_{IPV_{05-95}} = \frac{v_{50}-v_{05}}{v_{95}-v_{50}}\;, \end{equation} where $v_{05}$, $v_{50}$ and $v_{95}$ are the velocities corresponding to the wavelengths including $5\%$, $50\%$, and $95\%$ of the line total flux. Through this definition we are able to quantify the asymmetry of a line; with a $A_{IPV_{05-95}}>1$ the line is characterized by a blueward asymmetry, while for $A_{IPV_{05-95}}<1$ the line is more prominent in the red part of its profile. The asymmetry index for each stack is reported in Tab. \ref{tab5}. $A_{IPV_{05-95}}$ decreases moving towards high \ew\ in agreement with the result in Fig. \ref{OIII_profile} (i.e. a blue tail becoming less prominent at higher \ew. \begin{figure*} \centering \includegraphics[scale=0.42]{figures/hist_Loiii_L5100.pdf} \caption{$L_{[OIII]}$ and $L_{5100}$ distribution for each stack (the dashed lines represent the mean value for each distribution). The distribution of sources in terms of their continuum at $5100$\AA$\;$ luminosity is stable around $\sim 44.6$, a consequence of the flux limited selection. On the other hand the central value of sources distribution in terms of [OIII] luminosity increases going towards high \ew\ bins. This is due to the flux limit of the sample: when moving towards edge-on positions we are selecting intrinsically more luminous objects.} \label{hist_Loiii_L5100} \end{figure*} \subsection{Broad line profiles} The number of studies supporting a non-spherical shape of the BLR has constantly grown in recent years. \citet{Zhu2009} analyzed the BLR profiles in SDSS quasars and suggests that two components with different geometries and physical conditions are needed to reproduce the observed spectra; others studies also suggest the presence of two components, one of them with a spherical geometry while the outer one disk-shaped \citep{Bon2006}. In the same BLR disk-shaped scenario it has also been suggested that the kinematics of this inner region, consisting of a combination of rotational and turbulent motions, could affect its geometry, with broader lines emitted from more flattened regions \citep{Kollatschny2011}. Moreover, the proximity of the BLR to the obscuring ``torus'' of the Unified Model \citep{Antonucci93} suggests a smooth connection between the two structures, rather than two completely separated regions, as usually described in the standard unification model. Indeed, the BLR could represent a transition region from the outer accretion disk to the dusty region of the torus \citep{Goad2012}. Recently \citet{Pancoast2014} used direct modelling techniques on a sample of AGN for which high quality Reverberation Mapping data were available in order to investigate the geometry and the dynamics of the BLR. They found that the geometry of the BLR, as traced by H$\beta$ emission, is consistent with a thick disk. All these works, despite different aims and explanations, share a common interpretation of the geometry of the BLR. We claim that the dependence of the broad components on \ew\ is an evidence of the disk-like shape of the BLR: moving from low to high EWs (that is from ``face-on'' to edge-on objects) the component of velocity of the BLR in the direction of the observer grows steadily with the cosine of the inclination angle. Taking into account this result leads to a number of improvements in our understanding of the AGN inner regions, starting from the determination of the SMBHs virial masses. As long as we consider the BLR as composed by virialized gas, the SMBH mass can be inferred from the BLR lines width according to the relation \begin{equation} M_{BH}=f \frac{v_{obs}^{2} R_{BLR}}{G}\;\;, \end{equation} where $v$ is the BLR observed line width and $R_{BLR}$ can be obtained from Reverberation Mapping and from the luminosity-$R_{BLR}$ relation for single epoch observations \citep{Kaspi2000, Bentz2013}. An orientation-dependent analysis of emission lines could on one hand improve our knowledge of the morphology of the BLR and so help us in determining more accurately the \emph{virial factor} $f$ \citep{Shen2013}, and on the other hand remove the systematic underestimate of the line widths in non edge-one sources. \begin{figure*} \centering \includegraphics[scale=0.8]{figures/OII_OIII_stacks_py.pdf} \caption{[OIII] and [OII] profiles comparison for each stack; the [OIII] blue component decreases moving from low to high \ew\ stacks, i.e. from face-o to edge-on positions. From the relative position of the [OIII] peak with respect to the [OII] peak we can also estimate the systematic error' in sources redshift: the [OIII] peak is systematically shifted with respect to the [OII] one and the shift is decreasing moving towards higher \ew\ stacks. } \label{oiii_vs_oii} \end{figure*} \subsection{Eigenvector 1} The orientation effects revealed using \ew\ as an inclination indicator provide a possible interpretation of \emph{Eigenvector 1} (EV1) \citep{BorosonGreen1992}, namely the anticorrelation in the intensity of the emissions of FeII and [OIII]. FeII features are BLR lines and so they show the same trend as the other broad lines, i.e. a decreasing intensity and increasing broadening going from face-on (low EWs) to edge-on (high EWs) positions. On the other hand, the [OIII] emission is isotropic. EV1 can then be simply explained in terms of the orientation effects as follows: assuming a disk-like shape for the BLR the intensity of FeII emission lines decreases from face-on to edge-on positions. This effect is clearly present in our stacks: FeII emissions are more prominent in stacks with low \ew\ (Fig. \ref{fig:fits_figures}). FeII emissions seem to disappear moving towards high \ew, rather than decrease in intensity as the other broad lines do. This fact may suggest that the FeII disk-like structure, besides being flatter than the other broad lines emitting regions, is flatter than the continuum emitting region itself, i.e. the accretion disk. Unfortunately the FeII spectrum is characterized by the presence of several, close multiplets and its emissions are by far less intense than those of the other broad lines. This complicates a deeper investigation on this subject. It is possible, in fact, that for the two reasons mentioned above we are simply not able to detect a behaviour of FeII emissions similar to that of the other broad lines. \begin{figure} \centering {\includegraphics[scale=0.25]{figures/fit_hb_oiii_norm_hb_0001_0006.pdf}} {\includegraphics[scale=0.25]{figures/fit_hb_oiii_norm_hb_0012_0025.pdf}} {\includegraphics[scale=0.25]{figures/fit_hb_oiii_norm_hb_0100_0250.pdf}} \caption{\protect Fits of the H$\beta$-[OIII] spectral window for the $(1-6)$\AA, $(12-25)$\AA, $(100-250)$\AA$\;$ stacks respectively. The spectrum is fitted with several functions: for the broad component of permitted lines a double power law convolved with a Gaussian is used. Narrow lines (for both permitted and forbidden lines) are fitted to two Gaussian, the first one accounting for the main component of the line, the other one accounting for the blue tail ascribed to outflowing gas fron the NLR. For the FeII emission several templates are taken into account (see Section \ref{sec:spec_fitting} for details).} \label{fig:fits_figures} \end{figure} | 16 | 9 | 1609.06705 |
1609 | 1609.05706_arXiv.txt | Recently, ultraluminous X-ray source (ULX) M82 X-2 has been identified to be an accreting neutron star, which has a $P=1.37$ s spin period, and is spinning up at a rate $\dot{P}=-2.0\times 10^{-10}~\rm s\,s^{-1}$. Interestingly, its isotropic X-ray luminosity $L_{\rm iso}=1.8\times 10^{40}~\rm erg\,s^{-1}$ during outbursts is 100 times the Eddington limit for a $1.4~\rm M_{\odot}$ neutron star. In this Letter, based on the standard accretion model we attempt to constrain the dipolar magnetic field of the pulsar in ULX M82 X-2. Our calculations indicate that the accretion rate at the magnetospheric radius must be super-Eddington during outbursts. To support such a super-Eddington accretion, a relatively high multipole field ($\ga 10^{13}$ G) near the surface of the accretor is invoked to produce an accreting gas column. However, our constraint shows that the surface dipolar magnetic field of the pulsar should be in the range of $1.0-3.5\times 10^{12}$ G. Therefore, our model supports that the neutron star in ULX M82 X-2 could be a low magnetic field magnetar (proposed by Tong) with a normal dipolar field ($\sim 10^{12}$ G) and relatively strong multipole field. For the large luminosity variations of this source, our scenario can also present a self-consistency interpretation. | Ultraluminous X-ray sources (ULXs) are defined as point extranuclear sources found in nearby galaxies. ULXs are characterized by isotropic X-ray luminosities exceeding $\sim10^{40}~\rm erg\,s^{-1}$ (some papers adopted a luminosity of $10^{39}~\rm erg\,s^{-1}$), which is larger than the Eddington limit of a $10~\rm M_{\odot}$ stellar mass black hole \citep{fabb89}. To interpret the extreme luminosities, some works suggested that ULXs originated from stellar mass black holes by either actually super-Eddington X-ray radiation \citep{bege02} or anisotropic beaming emission \citep{king01,kord02}. Alternatively, the accretors in some ULXs are proposed to be intermediate-mass black holes ($10^{2}-10^{4}~ \rm M_{\odot}$) \citep{colb99,li04}. Since higher masses naturally yield a larger isotropic Eddington limit, theoretical models have favored black hole rather than neutron star X-ray binaries. The dynamical measurements for two ULXs masses support this viewpoint \citep{liu13,motc14}. Recently, the identification of NuSTAR J095551+6940.8 in the nearby galaxy M82 has raised an extreme interest of theoretical researchers \citep{bach14}. The broadband X-ray observations for M82 X-2 (also entitled as X42.3+59, discovered by Kaaret, Simet \& Lang 2006) imply pulsations with an average period of 1.37 s and a 2.5 d sinusoidal modulation. The former comes from the spin of a magnetized neutron star, and the latter is explained as the orbital motion. The isotropic X-ray luminosity in the 0.3 - 10 keV can reach $L_{\rm iso}=1.8\times 10^{40}~\rm erg\,s^{-1}$, which is 100 times the Eddington limit for a $\rm 1.4~M_{\odot}$ neutron star. Meanwhile, the pulsar in M82 X-2 was reported to be spinning up at a rate $\dot{P}=-2.0\times 10^{-10}~\rm s\,s^{-1}$ (in the interval from MJD 56696 to 56701) in a high-mass X-ray binary including a donor star with mass greater than $5.2~\rm M_{\odot}$ \citep{bach14}. Simulations by binary population synthesis show that the contribution of neutron star X-ray binaries for the ULXs population obviously exceeds black hole X-ray binaries \citep{shao15}. It is an intriguing issue how to radiate such an X-ray luminosity for an accreting neutron star. \cite{eksi15} show that the pulsar in M82 X-2 is a magnetar with a dipole magnetic field of $6.7\times 10^{13}$ G, which can reduce the scattering cross-section and enhance the critical luminosity \citep{pacz92}. Based on the observational properties of M82 X-2, the calculation for 30 keV photons performed by \cite{dall15} indicates that a magnetic field $\sim 10^{13}$ G can reduce the scattering cross-section by a factor of 50. \cite{kari16} have simulated the formation of M82 X-2, and concluded that a neutron star with $4\times10^{13}$ G magnetic field accreting from a disc-shape wind produced by a Be-Companion can reproduce the observed parameters. Conversely, \cite{kluz15} favored a neutron star with a weak magnetic field of $10^{9}$ G. They argued that such a magnetic field allows the accretion disc to extend to the surface of the neutron star, providing the spin-up torque fitting the observed properties. \cite{tong15} argued that the accretion column originating from strong multipole field of a low magnetic field magnetar could responsible for the super-Eddington accretion. Recently, \cite{king16} presented an interesting picture, in which M82 X-2 is a beamed X-ray source feeding at super-Eddington rate. In present, the surface dipolar magnetic field and large X-ray luminosity variation in M82 X-2 still remain controversial. Based on the standard accretion model, in this Letter we attempt to constrain the surface dipole magnetic field and the beaming factor of M82 X-2. | Based on the observed properties of M82 X-2, using the standard accretion model we found the pulsar should experienced a super-Eddington accretion at a rate of at least $3.4~\dot{M}_{\rm Edd}$ during outbursts. Strong multipole field ($\ga ~10^{13}$ G) near the surface of the pulsar can result in an accretion gas column, which allow the accretor accreted the material at a super-Eddington rate. However, our calculation show that the upper limit of the dipole magnetic field in the pulsar pole is $3.5\times10^{12}$ G {when $\xi=1.0$. Therefore, we proposed that the pulsar in M82 X-2 could be low magnetic field magnetar. It has a normal dipole field ($\sim 10^{12}$ G) and relatively strong multipole field in the surface of the pulsar \citep{tong15}. At present, three candidates for low magnetic field magnetar including SGR 0418+5729 \citep{rea10}, Swift J1822.3$-$1606 \citep{rea12}, 3XMM J185246.6+003317 \citep{zhou14,rea14} have been detected. Based on the accretion induced the polar magnetic field decay model, the simulation given by \cite{pan16} show that the polar magnetic field decays to $4.5\times10^{13}$ G when the pulsar in M82 X-2 accreted $\sim 0.005~\rm M_{\odot}$, while the strong magnetic field still remain in the out-polar region. According to the observed parameters of M82 X-2, and adopting a typical accretion column factor $l_{0}/d_{0}=30$, the dipole magnetic field in the pulsar pole and the beaming factor at peak luminosity are constrained to be $B_{\rm p}=1.0-3.5\times 10^{12}$ G, and $b= 0.03 - 0.05$, respectively. A high accretion rate would obtain a small minimum magnetic field. Recently, based on the observations in the lowest observed states \citep{brig16,tsyg16}, \cite{chri16} used two independent methods to calculate the magnetic field of the pulsar in M82 X-2, and obtain two similar results of $3.1\times10^{12}$, and $2.3\times10^{12}$ G. Following their assumption, if the pulsar is in the propeller line, and is accreting at Eddington accretion rate, our equation (6) derives a magnetic field of $1.6\times10^{12}$ G (The main difference originates from different magnetic momentum ($\mu$) formulas: we use $\mu=BR^{3}$, while they took $\mu=BR^{3}/2$). However, in our scenario such an accretion rate and a magnetic field can not provide an accretion torque fitting the observation. Due to the uncertainty of $\xi$ factor in the magnetospheric radius equation, it seems that our scenario can not fully rule out the magnetar possibility for M82 X-2. If we take a relatively small $\xi=0.5$ favored by \cite{ghos79a}, it would give rise to an upper limit of $1.2\times10^{13}$ G, which is still less than the quantum critical limit $B_{\rm c}=m_{\rm e}^{2}c^{3}/\hbar e=4.4\times 10^{13}$ G. However, our estimation for the accretion rate strongly depend on the accretion column model. If the pulsar in M82 X-2 is experiencing a hyper-critical accretion at a rate of 100 $\dot{M}_{\rm Edd}$, equation (6) shows that the maximum dipole magnetic field may reach $10^{14}$ G. It is worth noting that it just is an upper limit of the dipole magnetic field. In our opinion, at present it is difficult to make an accurate constraint for the dipole field of the pulsar in M82 X-2. \cite{kong07} detected M82 X-2 turned off twice in 1999, and 2000. Recently, \cite{dall16} reported that this source has not been detected in the first observation performed by \emph{Chandra} HRC. An analysis for \emph{Chandra} fifteen years data indicated that M82 X-2 has a violent X-ray luminosity variations from $10^{38}$ to $10^{40}~\rm erg\,s^{-1}$ \citep{brig16}. These observations indicate that M82 X-2 experienced a large variation in the accretion rate. In our model, only if the accretion rate at $r_{\rm m}$ during outbursts is greater than $3.4~\dot{M}_{\rm Edd}$ (see also equation 7), $r_{\rm m}<r_{\rm co}$, and accretion occurs. Actually, the spin-up rate of M82 X-2 in the interval from MJD 56685.5 to 56692 is $\dot{P}\approx-4.0\times 10^{-11}~\rm s\,s^{-1}$ (see also Figure 2 of Bachetti et al. 2014), which also reveals that the accretion rate of the pulsar decreases by a factor of five. Therefore, the accretion rate should be greater than $\sim0.7~\dot{M}_{\rm Edd}$. When the accretion rate drops to less than $0.7~\dot{M}_{\rm Edd}$, $r_{\rm m}>r_{\rm co}$, the pulsar is in the propeller phase \citep{illa75}, and the accretion ceases or sub-Eddington accretion occurs. The disc inner radius in M82 X-2 has been reported to be $3.5^{+3.0}_{-1.9}\times 10^{9}$ cm \citep[90\% confidence,][]{feng10}. This radius obviously exceed the corotation radius, the standard accretion theory suggested that the accretion should halt because of centrifugal force. If this measurement is confident, meantime the pulsar should be in propeller phase.Some other neutron star X-ray binaries emitted much lower X-ray luminosities, which provided strong evidence of propeller phase \citep{cui97,zhan98}. \cite{dall16} proposed a similar model, while their critical mass inflow rate is $\dot{M}_{\rm Edd}$. The sub-Eddington accretion would cause the source luminosity to decline by a large factor. If the accretion fully stops, much smaller X-ray luminosity may originated from the magnetosphere emission \citep{camp97,camp95}. | 16 | 9 | 1609.05706 |
1609 | 1609.02917_arXiv.txt | Following the recent discovery of the first radial velocity planet in a star still possessing a protoplanetary disc (CI Tau), we examine the origin of the planet's eccentricity (e $\sim 0.3$). We show through long timescale ($10^5$ orbits) simulations that the planetary eccentricity can be pumped by the disc, even when its local surface density is well below the threshold previously derived from short timescale integrations. We show that the disc may be able to excite the planet's orbital eccentricity in $<$ a Myr for the system parameters of CI Tau. We also perform two planet scattering experiments and show that alternatively the observed planet may plausibly have acquired its eccentricity through dynamical scattering of a migrating lower mass planet, which has either been ejected from the system or swallowed by the central star. In the latter case the present location and eccentricity of the observed planet can be recovered if it was previously stalled within the disc's magnetospheric cavity. | The recent discovery of a radial velocity planet in the young, disc bearing star CI Tau \citep{Johns-Krull2016} offers the first opportunity to test theories for the formation and early evolution of hot Jupiters in discs. To date, planet discoveries in discs have derived from direct imaging (e.g \citealt{Chauvin2004,Chauvin2005,Neuhauser2005,Neuhauser2008,Marois2008,Kraus2012,Sallum2015}) due to difficulties in applying transit detection and radial velocity methods in young stars. The presence of discs evidently rules out transit detections (though candidate transit detections have been obtained in disc-less young stars: \citealt{vanEyken2012,Ciardi2015,David2016}). Both the transit and radial velocity techniques are impeded by the extreme variability of young stars \citep{Xiao2012,Stauffer2014}; in particular it is difficult to disentangle companion induced radial velocity variations from the quasi-periodic signals produced by starspots. In the case of CI Tau, however, this effect has been minimised using K band (where starspot activity is reduced); this has allowed the extraction of a radial velocity periodicity ($9$ days) distinct from the photometric period ($7$ days, plausibly ascribed to stellar rotation). The planet parameters in CI Tau (P = $9$ days, Msini = $8.1$ M$_{\rm Jup} $) place it firmly in the "hot Jupiter" category. In contrast to another hot Jupiter recently found around a T Tauri star \citep{Donati2016}, CI Tau also possesses a massive circumstellar disc of $\sim 37$ M$_{\rm Jup}$ as deduced from previous mm observations using PdBI (\citealt{Guilloteau2011}; see also \citealt{AndrewsWilliams2007}); if the inclination inferred from the outer disc ($45$ to $54$ degrees; \citealt{Guilloteau2014}) is also the inclination of the planet then the measured Msini corresponds to a mass of $\sim 10$M$_{\rm Jup}$. Given the impossibility of forming giant planets {\it in situ} in close proximity to the host star\footnote{See \citet{ChiangLaughlin2013,HansenMurray2013} for {\it in situ} formation models for planets considerably less massive than that in CI Tau.}, there is a long-standing debate about the origin of hot Jupiters: whether they arrive in their present locations during the gas rich phase (by disc mediated migration and/or scattering of planetary embryos) or whether instead by dynamical scattering after the disc has dispersed \citep{Lin1996,RasioFord1996}. The recent discovery in CI Tau provides a key demonstration that in at least one object the former is the case. % The relatively high eccentricity ($e = 0.3 \pm 0.16$)\footnote{See Figure 5 of \citealt{Johns-Krull2016} for a plot of the distribution of the possible values.} is however somewhat unexpected in a scenario of purely disc mediated migration, as discs tend to damp planetary eccentricity \citep{Papaloizou2000,TanakaWard2004}\footnote{ Note that {\it stellar tides} raised in the planet are ineffective in modifying the eccentricity of a planet with these orbital parameters on a Myr timescale \citep{BarkerOgilvie2009}}. Although the eccentricity of massive planets can be excited by the disc \citep{Papaloizou2001,Dangelo2006,Bitsch2013}, \citet{Dunhill2013} have argued that this requires that the disc surface density in the vicinity of the planet falls in a restricted range: we will revisit this conclusion through long timescale FARGO3D integrations of disc planet systems in Section 2. Alternatively, such eccentricities can be driven by interactions involving multiple planets \citep{Marzari2010,Moeckel2012,Lega2013}; in the case of CI Tau, the absence of another period in the radial velocity data implies that any perturbing giant planet is no longer in the sub- A.U. region and in Section 3 we explore, through two-planet FARGO3D simulations and through simply parametrised scattering experiments, whether there are orbital histories that can generate significant eccentricity in the observed planet while also removing the perturber from the inner disc. We emphasise that this paper is mainly concerned with the excitation of eccentricity in the CI Tau radial velocity planet and we do not present an exhaustive set of scenarios for the system's prior evolutionary history. In Section 4 we discuss whether the planet is likely to have acquired its eccentricity at its current position and whether its present location - close to but not at the radius of corotation between the disc and the star - is significant. | So far we have considered various eccentricity driving mechanisms assuming these operate close to the planet's present position. Given that it is unlikely that the planet formed at $0.1$ A.U., we need also to consider scenarios for its inward migration from larger radius. While there have been conflicting results as to whether the development of eccentricity inhibits disc mediated migration \citep{Papaloizou2001,Dangelo2006,Duffell2015,Rice2008}, there is some suggestion that eccentricity growth is favoured at low $H/R$ \citep{Armitage2005}. Since discs are generally flared ($H/R$ increases with $R$), this would favour growth at small radii, possibly delaying the excitation of eccentricity until the planet arrives at its present location. Also, our own simulations, which are the only ones to have studied the migration of eccentric planets over such long timescales, however find that eccentric planets {\it can} migrate (see Figure \ref{fig:eccen_fargo}) and in this case the eccentricity may be excited anywhere between its birthplace and current location. We find (in simulations for which $e$ is as large as $0.15$) that the planet migrates at a rate that is consistent with the rate found in simulations involving planets on circular orbits (\citet{Duffell2014} and \citet{Duermann2015}). The migration time can be approximated as $t_{mig} = t_\nu \max(1,\frac{M_p}{\Sigma \pi a^2})$, where $t_\nu$ is the viscous timescale \footnote{ The formula ignores the recent findings that type II migration is a factor of 2-3 faster than the viscous time-scale as the ones presented here are order of magnitude estimates, but it does catch the correct dependence with the disc mass.}. For the massive planet considered here, the second term in brackets is relevant within $\sim 10$ A.U. (i.e. the planetary inertia is important) so that for a steady state disc we have $t_{mig} \sim 10 M_p/\dot M$, independent of radius and viscosity assumptions. For the parameters of CI Tau, this implies $t_{mig} \sim$ a Myr. It is tempting to ascribe some significance to the fact that the radius of the observed planet is only $\sim 30 \%$ beyond the corotation radius between the star and the disc (assuming that the $7$ day photometric period of the star measured by \citealt{Johns-Krull2016} is the star's rotation period). Models of disc braking of young stars suggest that systems evolve to a state of disc locking where the disc is truncated slightly inside the corotation radius. A migrating planet is expected to stall as it enters the magnetospheric cavity \citep{Romanova2006,Papaloizou2007} so that if another planet arrives at small radii through disc migration, a dynamical interaction between the two is assured at this location. In around $40 \%$ of the cases studied in Section 3, the remaining $10$ Jupiter mass planet is scattered outward in the interaction while the lighter sibling planet ends up being swallowed by the star. This provides a plausible explanation for why the planet is located in the vicinity of, but not exactly at, the expected radius of the magnetospheric cavity. Alternatively, if disc mediated migration is effective at moderate eccentricity, then the dynamical interaction can have occurred at a range of radii. While the subsequent migration may be accompanied by a damping of the eccentricity to the equilibrium value excited by the disc, this would take $\sim 10^6\,\mathrm{yr}$ (longer if $\tau_e$ is larger at large $e$, as found by \citealt{Papaloizou2000}). In this case the proximity of the observed planet to the putative magnetospheric cavity is coincidental and the planet is likely still migrating. | 16 | 9 | 1609.02917 |
1609 | 1609.09528_arXiv.txt | Detailed stellar abundances from modern astronomical surveys represent a unique window on the early universe and on the formation of the Milky Way and its satellite galaxies. However, the theoretical tools needed to translate these abundances into meaningful understanding of cosmological structure formation are challenged by uncertainties in their inputs. In this proceedings, we present an overview of our efforts to address this fundamental challenge by creating a flexible pipeline (see Fig.~\ref{fig_pipeline}) for modeling chemical evolution and by quantifying the robustness of our chemical evolution predictions due to uncertainties in nuclear physics, stellar evolution models, and observational inputs. | 16 | 9 | 1609.09528 |
||
1609 | 1609.09658_arXiv.txt | { Fast radio bursts can be caused by some phenomena related to 'new physics'. One of the most prominent candidates of the kind are axion Bose stars which can undergo conversion into photons in magnetospheres of neutron stars. In this short research note this scenario is investigated and important caveats are examined. First, tidal disruption of Bose stars can considerably extend the time span of conversion process, making it impossible to reach necessary $\mathcal{O}$(ms) level. Second, existing observations at two widely separated frequencies imply that axion-related scenarios can hardly explain entirety of FRBs.} \PACS{} \begin{document} | 16 | 9 | 1609.09658 |
||
1609 | 1609.02256_arXiv.txt | Ground Level Enhancements (GLEs) of cosmic-ray intensity occur, on average, once a year. Due to their rareness, studying the solar sources of GLEs is especially important to approach understanding their origin. The SOL2001-12-26 eruptive-flare event responsible for GLE63 seems to be challenging in some aspects. Deficient observations limited its understanding. Analysis of extra observations found for this event provided new results shading light on the flare. This article addresses the observations of this flare with the \textit{Siberian Solar Radio Telescope} (SSRT). Taking advantage of its instrumental characteristics, we analyze the detailed SSRT observations of a major long-duration flare at 5.7 GHz without cleaning the images. The analysis confirms that the source of GLE63 was associated with an event in active region 9742 that comprised two flares. The first flare (04:30\,--\,05:03 UT) reached a GOES importance of about M1.6. Two microwave sources were observed, whose brightness temperatures at 5.7 GHz exceeded 10 MK. The main flare, up to the M7.1 importance, started at 05:04 UT, and occurred in strong magnetic fields. The observed microwave sources reached about 250 MK. They were not static. Having appeared on the weaker-field periphery of the active region, the microwave sources moved toward each other nearly along the magnetic neutral line, approaching a stronger-field core of the active region, and then moved away from the neutral line like expanding ribbons. These motions rule out an association of the non-thermal microwave sources with a single flaring loop. | \label{S-introduction} Solar energetic particles are somehow accelerated in association with solar eruptive events. The highest-energy particles sometimes produce in the Earth's atmosphere considerable fluxes of secondary particles, which are able to cause ground-level enhancements (GLE) of cosmic-ray intensity. GLEs are mainly detected with neutron monitors (see, \textit{e.g.}, \citealp{Cliver2006, Nitta2012}; and references therein). Seventy-two GLE events caused by relativistic solar protons have been observed from 1942 to 2015. Most solar sources of the 51 GLEs registered after 1970 \citep{Kurt2004, Cliver2006, Aschwanden2012, Gopalswamy2012, Gopalswamy2013, Thakur2014} were associated with major flares of the soft X-ray (SXR) GOES X class (40 GLEs, including GLE72 -- see \citealp{Chertok2015}) or M class (GLE28 with an M5, GLE63 with an M7.1, and GLE71 with an M5.1 flare). The GOES importance of the flare associated with GLE24 and those of the far-side sources of GLE23, GLE29, GLE39, GLE50, and GLE61 are uncertain. Atypically favorable conditions probably accounted for GLE33 (C6) and GLE35 (M1.3) associated with moderate flares, weak microwave bursts, and relatively slow coronal mass ejections (CMEs), due to shock-acceleration of a high coronal seed population \citep{Cliver2006}. The rare occurrence of GLEs and their frequent association with both big flares and fast CMEs hampers identifying their origins and makes studying parent solar events highly important. The `big flare syndrome' concept \citep{Kahler1982} explained the correlation between the parameters of near-Earth proton enhancements and flare emission by a general correspondence between the energy release in an eruptive flare and its various manifestations. However, recent studies by \cite{Dierckxsens2015} and \cite{Trottet2015} indicate that both flares and shock waves can accelerate GLE particles. \cite{Grechnev2013b} revealed a scattered correlation between the peak fluxes of $>100$~MeV protons and peak flux densities of 35~GHz bursts, although the proton outcome of four events, including GLE63 and GLE71, was much stronger. Further, \cite{Grechnev2015} found a higher correlation between the total proton and microwave fluences; nevertheless, the proton-abundant events were the outliers. Their superiority could be due to, for example, predominant shock-acceleration or contributions from stronger nearly concurrent far-side events. The subject of the present study is the 26 December 2001 solar event with an SXR peak time at 05:40~UT (all times hereafter refer to UT) responsible for GLE63. The aim is to understand the possible causes of its atypically high proton outcome. Some other aspects of this solar event look also challenging. The SXR emission of this event was atypically long \citep{Aschwanden2012}. Examination of the SXR light curves led \cite{Gopalswamy2012} to set a probable onset time of the associated flare at 05:03~UT. This time is close to the extrapolated CME onset time in the online CME catalog (\url{http://cdaw.gsfc.nasa.gov/CME_list/}; \citealp{Yashiro2004}) and a reported appearance of a type II burst in the range 04:59\,--\,05:02~UT. On the other hand, this burst is clearly visible at 04:53~UT and detectable still earlier in the \textit{Hiraiso Radio Spectrograph} (HiRAS) spectrum (2001122605.gif) at \url{http://sunbase.nict.go.jp/solar/denpa/hirasDB/Events/2001/}. This slowly drifting burst evidences the presence of a moving source at least, ten minutes before the estimated onset time of a fast CME (average speed of 1446~km~s$^{-1}$ according to the CME catalog). Furthermore, it is not clear why the fast CME and a strong shock wave (possibly responsible for the GLE particles) developed in association with a microwave burst, which was not extremely strong. It is not also clear when and where the shock wave appeared. The flare and eruption in this event have been studied incompletely because of limited data. The observations with the \textit{Extreme-ultraviolet Imaging Telescope} (EIT: \citealp{Delaboudiniere1995}), onboard the \textit{Solar and Heliospheric Observatory} (SOHO), had a gap from 04:47 to 05:22~UT. The \textit{Transition Region and Coronal Explorer} (TRACE: \citealp{Handy1999}) did not produce extreme-ultraviolet images in which an eruption could be detected. No SXR images or hard X-ray data are available. The search for extra data revealed that the 26 December 2001 event was observed in microwaves with the \textit{Siberian Solar Radio Telescope} (SSRT: \citealp{Smolkov1986, Grechnev2003}) at 5.7~GHz; the \textit{Nobeyama Radioheliograph} (NoRH; \citealp{Nakajima1994}) at 17 and 34 GHz, and in 1600~\AA\ by TRACE. The analysis of extra observations found for this event has led to new results that we present in three companion papers. This article (Paper~I) addresses the SSRT observations of this flare. The SSRT routinely observes the Sun since 1996, but the difficulties to clean and calibrate the SSRT images of strong flare sources and a rather long time (typically 2\,--\,3 min) required to produce each image restrict the opportunities to use its imaging (2D-mode) observations. The usage of the SSRT data in studies of flares has been limited (\textit{e.g.} \citealp{Altyntsev2002, Altyntsev2007, Altyntsev2016, Meshalkina2012, Alissandrakis2013}). The 1D-mode observations of a higher time resolution have also been involved, especially in studies of microwave sources on short time scales. The flare occurred near the winter solstice that is a most unfavorable season for the observations with the SSRT. At that time, its beam pattern is considerably extended in the north-south direction, and distortions associated with insufficiently accurate knowledge of instrumental characteristics reach maximum. These circumstances hamper efficient cleaning of the images. Nevertheless, a relatively low level of the SSRT beam side lobes due to a large number of antennas arranged in an equidistant array makes raw images (without cleaning) usable for an analysis. In the present article the 26 December 2001 flare is studied from SSRT raw images as well as its 1D response. This has made possible to study the flare development in microwaves, to produce detailed light curves of the total flux and brightness temperature of the microwave sources at 5.7~GHz, and to analyze their spatial evolution. We focus on the opportunities provided by the SSRT data in studies of major flares and pursue the results, which can be obtained from the SSRT observations using the simplest estimations. Some of the results, which do not correspond to conventional properties of microwave flare sources, will be analyzed using observations in different spectral domains. Analyzing the TRACE and NoRH observations along with multi-frequency total flux data, Paper~II \citep{Grechnev2016a} elaborates and clarifies the conclusions drawn in the present article. Based on the results of these two articles, Paper~III (Grechnev \textit{et al.}, 2016b, in preparation) will address the eruption and endeavor to understand the possible causes of the high proton productivity of the 26 December 2001 solar event. | \label{S-discussion} Both SXR and microwave data show that the long-duration flare in AR~9742 consisted of two parts. The first flare started at about 04:30~UT, lasted half an hour, and reached a GOES importance of about M1.6 at 05:03~UT. The main flare, which was much stronger in microwaves, started at 05:04~UT and reached an importance of M7.1 at 05:40~UT. The major microwave burst fully developed and mainly finished within 36~min, before the SXR peak. The decay of the SXR emission possibly lengthened due to subsidiary bursts during 05:40\,--\,06:10~UT and around 06:25~UT. The SXR emission corresponded to the microwave burst via the Neupert effect. The only source of the microwave burst was AR~9742, and the only source of the SXR emission was the flare in this region. Most likely, the solar event in AR~9742 associated with the M7.1 flare was the only source of the near-Earth proton event and GLE63. In the preceding studies of microwave flare emissions, the sources observed by the SSRT at 5.7~GHz were typically associated with loop-top regions (\textit{e.g.} \citealp{Altyntsev2002, Altyntsev2007, Altyntsev2016, Meshalkina2012}, and others). In the 26 December 2001 flare, two distinct microwave sources were observed. They were polarized, comparable in size and brightness temperatures (up to extremely high values), and varied rather similarly. These properties of the two sources and their positions, most likely at different sides of the magnetic neutral line, indicate their association to the conjugate legs of a closed magnetic structure. Observations of flares in thermal emissions (H$\alpha$, ultraviolet, extreme ultraviolet, soft X-rays, \textit{etc.}) typically show complex multi-loop structures. By contrast, microwave and hard X-ray (HXR) images of non-thermal emissions in impulsive flares usually reveal simpler configurations identified with one or two loops (\textit{e.g.}, \citealp{Hanaoka1996, Hanaoka1997, Nishio1997, GrechnevNakajima2002}; and many others). Moreover, an analysis of the microwave morphologies in many near-the-limb events led \cite{TzatzakisNindosAlissandrakis2008} to a conclusion about single-loop microwave configurations existing even in some long-duration major flares. These observational indications resulted in a prevailing concept of a single microwave-emitting flare loop (or, at most, two loops). However, even the inhomogeneous flare-loop model, initially proposed by \cite{Alissandrakis1984} and further developed using powerful modeling tools to account for several inhomogeneities \citep{TzatzakisNindosAlissandrakis2008, KuznetsovNitaFleishman2011}, cannot explain various observations. In particular, \cite{Zimovets2013} demonstrated that, at least, some of the seemingly single loops shown by microwave NoRH images corresponded to multi-loop arcades observed with telescopes of a higher spatial resolution. The concept of a single microwave-emitting flare loop is difficult to reconcile with systematic motions of the microwave sources in Figure~\ref{F-2_sources}a. A continuous relative displacement of the two sources by $\gsim 20^{\prime \prime}$ in the plane of the sky implies their association with different loops or loop systems at different times. In principle, the observed relative motions of the sources could be caused by a varying height of the emitting regions in two legs of a single loop. This effect should also be manifested in the degree of polarization due to a varying magnetic field strength and viewing angle. However, while the variations in the relative distance in Figure~\ref{F-2_sources}a before 05:20~UT are nearly symmetric to those after 05:20~UT, the shear angle in Figure~\ref{F-2_sources}b and polarization in Figure~\ref{F-polariz}a are strongly asymmetric. Thus, while the height variations are possible, the observed motions were most likely determined by the displacements of the footpoint regions of numerous arcade loops. Similar motions are known from HXR observations. From a statistical study of HXR sources, \cite{Bogachev2005} found different types of their motions relative to the magnetic neutral line and interpreted them in terms of the standard flare model. The diverging motion at the decay phase (05:30\,--\,05:45~UT and later) resembles a usual expansion of the flare ribbons, which represent the footpoint regions of numerous loops. This is type~I motion according to \cite{Bogachev2005}. The approach of the two sources toward each other along the neutral line throughout the first flare is also difficult to relate to a single loop; this kind of motion was also observed in HXR (type~II motion in \citealp{Bogachev2005}). The deviations from the systematic motions during subsidiary bursts also imply more complex configurations than a single microwave-emitting flare loop. Additional indications are provided by the degree of polarization. This parameter of the gyrosynchrotron emission is closely related to the magnetic field strength, being not directly dependent on an unknown number of emitting electrons. Keeping this in mind, the varying degree of polarization in Figure~\ref{F-polariz}a is difficult to understand in terms of the single-loop hypothesis. On the other hand, with the limited data we consider, a probable participation of additional loops makes the interpretation of these complex variations ambiguous. We therefore consider, for simplicity, the approximation of a single homogeneous source for each of the two observed microwave sources. To analyze the gyrosynchrotron emission, analytic approximations by \cite{DulkMarsh1982} and \cite{Dulk1985} are widely used. The accuracy of the formulas is reduced at low harmonics of the gyrofrequency; in such situations, we invoke them to obtain rough estimates and to understand major tendencies, and additionally refer to the plots in those articles that present the results of numerical calculations. Note that gyroresonance features in the spectrum are not expected in observations due to inhomogeneities of the magnetic field. The polarization of the east source corresponded to the $x$-mode emission and reached $r_\mathrm{c} \approx -0.75$. This is only possible, if the optical thickness [$\tau$] was small ($\tau < 1$). According to \cite{Dulk1985}, the degree of circular polarization, $r_\mathrm{c}$, in the optically thin limit is \begin{eqnarray} r_\mathrm{c} \approx 1.26\times 10^{0.035\delta}\times 10^{-0.071\cos\theta}\left(\frac{\nu}{\nu_{B}}\right)^{-0.782+0.545\cos\theta} \quad (\tau_{\nu}\ll 1), \nonumber \end{eqnarray} with $\delta$ being a power-law index of the number density spectrum of microwave-emitting electrons, $\theta$ a viewing angle, and $\nu_{B} \approx 2.8 \times 10^6 B$ the electron gyrofrequency in the magnetic field [$B$]. The degree of polarization directly depends on the magnetic field strength. The polarization of the east source strengthened until the decay phase. During the first flare, this occurred presumably due to an increasing contribution of the gyrosynchrotron emission, which became dominant at the end of the first flare (Figure~\ref{F-flare_light_curves}b); and, possibly, due to the motion of the source from a weaker-field periphery of the active region to its stronger-field core. With a degree of polarization of $-(0.20-0.25)$ just before the main flare and a viewing angle around $\theta \approx 60^{\circ}$, corresponding to the position of AR~9742, the magnetic field strength in the east source should not exceed $-100$~G. This conclusion applies to the whole first flare. An increasing degree of the polarization during the major burst suggests a strengthened magnetic field in the east source. Assuming for certainty $\theta = 60^{\circ}$ and $\delta = 2.5-3.5$ (see Paper~II), with $r_\mathrm{c} \approx -0.5$ at the peak of the burst, we estimate the magnetic field strength in the east source to be around $-250$~G at that time. With $r_\mathrm{c} \approx -0.75$ at the end of the major burst, the magnetic field strength in the coronal east source could reach $\approx -540$~G. This magnetic field corresponds to $\nu/\nu_B \approx 4$, beyond the validity range of the \cite{Dulk1985} approximation, but nevertheless consistent with his Figure~3. If the optical thickness of this source was not small enough to satisfy the condition $\tau_{5.7} \ll 1$, then the magnetic field should be somewhat stronger. The magnetic field at the photosphere underneath should be considerably stronger than in the low corona. The magnetic field in this region of the MDI magnetogram on that day at 04:51~UT ranged from $-600$ to $-850$~G. Since AR~9742 was not far from the limb, this magnetogram might be strongly affected by the projection effect. We additionally examined an MDI magnetogram observed two days before, at 04:51~UT on 24 December. The magnetic fields at about this place reached more than $-1000$~G; on the other hand, the active region evolved. Thus, the estimated maximum magnetic field strength of $\approx -540$~G in the east coronal source seems to be plausible. The west source was polarized in the sense of the $o$-mode emission, with a degree, not exceeding $-20\%$. Either its intrinsic emission corresponded to the $x$-mode and was inverted, propagating through a layer of the quasi-transversal magnetic field, or it was initially optically thick. If the SSRT observing frequency, 5.7~GHz, was higher than the peak frequency of this source, then its brightness temperature should depend on the magnetic field strength directly \citep{DulkMarsh1982, Dulk1985}. However, a weak magnetic field of $-(40-45)$~G corresponding to $r_\mathrm{c} = -0.2$ with the same $\delta = 2.5-3.5$ and $\theta = 60^{\circ}$ would contradict a higher brightness temperature of the west source relative to the east source. Hence, the west source was not optically thin. Its peak frequency, $\nu_{\rm peak}$, was either slightly lower than 5.7~GHz (inverted $x$-mode emission), or, most likely, higher (intrinsic $o$-mode emission). The latter option is consistent with an estimated $\nu_{\rm peak} \approx 6.9$~GHz for the total flux in this event (\citealp{Grechnev2013b}; see also Paper~II). Indeed, the total flux is the sum of the emissions from the east source with a $\nu_{\rm peak} < 5.7$~GHz, and the west source, whose peak frequency should be $> 5.7$~GHz, even if the total flux had $\nu_{\rm peak} \geq 5.7$~GHz. The peak frequency can be estimated, referring again to \cite{DulkMarsh1982} and \cite{Dulk1985}, as \begin{eqnarray} \nu_\mathrm{peak} \approx 2.72 \times 10^3 \times 10^{0.27\delta}(\sin\theta)^{0.41+0.03\delta}(NL)^{0.32-0.03\delta} \times B^{0.68+0.03\delta}, \nonumber \end{eqnarray} where $(NL)$ is a column density of emitting electrons. Although it can be different in the two sources located in the conjugate legs of the same closed structure, the dependence of $\nu_\mathrm{peak} \propto (NL)^{0.22-0.25}$ is considerably weaker than $\nu_\mathrm{peak} \propto B^{0.76-0.79}$. While the basic formula might be inaccurate at a low harmonic of the gyrofrequency, a stronger magnetic field seems nevertheless to be a most probable reason for a higher $\nu_\mathrm{peak}$ in the west source. A brightness temperature of $\approx 4.2 \times 10^8$~K estimated in Section~\ref{S-main_properties} for the deconvolved west source near the peak of the burst (Figure~\ref{F-ssrt_img}c) roughly corresponds, with $\delta = 2.5-3.5$ and $\theta = 60^{\circ}$, to the optically thick emission around the fifth harmonic (Figures~3 in \citealp{DulkMarsh1982} and \citealp{Dulk1985}), \textit{i.e.}, 400~G \textit{vs.} $\approx 250$~G in the east source at the same time. All of the estimates, along with a behavior of the polarization during the main flare, indicate that the west source was optically thick and located in a stronger, relative to the east source, magnetic field, which increased in the course of the flare. This conclusion is supported by a higher brightness of the west source throughout the event at both 17 and 34~GHz in the NoRH movie, \url{norh20011226_0505_pfi.mpeg}, available at \url{http://solar.nro.nao.ac.jp/norh/html/event/} entry \url{20011226_0505}. Both sources were optically thin at these two frequencies, and therefore their brightness temperatures directly depended on the magnetic field strength. The magnetogram observed on 26 December shows the photospheric magnetic fields in the west part of AR~9742 to be around 1000~G. A sunspot was there. A nearly radial magnetic field in its central part should be substantially reduced in the line-of-sight magnetogram observed close to the limb. We have not radialized the magnetogram to avoid overestimating the magnetic field in the region under the east source. The MDI magnetogram on 24 December shows the magnetic fields exceeding 2700~G in the central part of the sunspot. Strong magnetic fields were really present on the photosphere approximately under the west source, which was the brightest during the main flare. Some of the estimated magnetic field strengths fall outside the range where the accuracy of the formulas by \cite{DulkMarsh1982} and \cite{Dulk1985} is guaranteed. Nevertheless, our results are supported by the following facts. i)~Our estimates are also consistent with the results of numerical calculations by \cite{DulkMarsh1982} and \cite{Dulk1985} in their Figures~3. ii)~Comparison of the positions of the two sources in Figures \ref{F-ssrt_img}b\,--\,\ref{F-ssrt_img}d with the magnetogram confirms that magnetic fields under the west source were stronger than those under the east source, as considered in this section, while their probable values in the corona correspond to the estimates. iii)~The NoRH movie of the flare observed at 17 and 34~GHz also supports our results. We conclude that the main flare occurred in strong magnetic fields, whose photospheric base, most likely, had a strength of $\gsim 1000$~G. Probably, the west flare ribbon extended into the strongest magnetic fields above the sunspot. However, the spatial resolution of the SSRT and the coalignment accuracy are insufficient to judge to what extent this occurred. This issue will be addressed in Paper~II, which will also analyze the microwave spectrum. As mentioned in Section~\ref{S-introduction}, the onset time of the main flare corresponds to an estimated launch time of the CME. The major phase of the 26 December 2001 GLE63-related event resembles those of the 20 January 2005 event (GLE69; \citealp{Grechnev2008}) and of the 13 December 2006 event (GLE70; \citealp{Grechnev2013a}). Furthermore, \cite{Grechnev2013b} showed flaring in stronger magnetic fields above the sunspot umbrae to be typical of big proton events. Like the previously mentioned events, the flare on 26 December 2001 involved rather strong magnetic fields and occurred, at least, close to a sunspot. The magnetic fields involved in the GLE63-related flare were probably not so strong as in the flares related to GLE69 and GLE70, when the peak frequencies exceeded 25 GHz, and the fluxes at 35~GHz were considerably higher than $10^4$~sfu. Nevertheless, major aspects of these events look qualitatively similar. An additional particularity of the 26 December 2001 event was the very long duration of the flare. The rise phase of the main flare alone lasted 36~min \textit{vs.} 18~min for the flares related to GLE69 and GLE70. According to the Neupert effect \citep{Neupert1968}, this phase corresponds to the effective particle acceleration in a flare. A considerably higher correlation between the fluences of near-Earth proton enhancements, on the one hand, and fluences of the SXR and microwave emissions, on the other hand found by \cite{Grechnev2015} indicates a dependence of the total number of high-energy protons arriving at the Earth orbit on both the intensity and total duration of the acceleration process. The role of the flare duration is obvious, if protons are accelerated simultaneously with electrons in a flare, but it is more difficult to expect such a correspondence, if protons are accelerated by shock waves far away from a flare region. Therefore, a considerably higher correlation between the fluences of protons and flare emissions than between their peak values found for the 26 December 2001 event indicates a significant contribution from flare processes to the acceleration of protons. The discussion of the particle event will be further addressed in Paper~III. \subsection{Conclusion} Our analysis has confirmed that the solar event in active region 9742 associated with the M7.1 flare was the only source of the near-Earth proton event and GLE63. No signs of a concurrent far-side event have been found. The event in AR~9742 consisted of two parts. The first flare (04:30\,--\,05:03~UT) reached a GOES importance of about M1.6. The brightness temperatures at 5.7~GHz exceeded 10~MK. The main flare, up to M7.1 importance, started at 05:04~UT, when a CME was launched. The microwave sources reached about 250~MK. The SSRT data indicate that strong magnetic fields were involved in the main flare. These magnetic fields were probably associated with the sunspot in the west part of AR~9742. The two microwave sources observed at 5.7~GHz initially approached each other along the magnetic neutral line and then moved away from it like expanding ribbons. These motions are difficult to understand in the frame of a single-loop hypothesis. A natural explanation of the observed properties of the microwave sources might be their association to the legs of the flare arcade. To verify this conjecture, microwave data should be compared with the flare arcade or ribbons observed in a different spectral range, where they are clearly visible. These issues will be addressed in Paper~II. The possible causes of the high proton productivity of the 26 December 2001 event will be considered in Paper~III. This is a first detailed study of a major long-duration flare from combined 2D and 1D SSRT data. A relatively low side-lobe level of the SSRT beam and rather large areas of the microwave sources allowed using the images produced by the SSRT without cleaning. The techniques described here provide an opportunity to study important major flares recorded with the SSRT in the past. The analysis has revealed shortcomings of the imaging and calibration software that were not manifested previously because of a deficient experience in handling major flares. Some imperfect techniques and software (\textit{e.g.}, calibration routines) have been improved in the course of our study. The development of some others is in progress. The SSRT routinely carried out imaging observations of the whole Sun at 5.7~GHz based on the initial operating principle from 1996 to July 2013. Currently, the central part of the antenna array is under reconstruction to upgrade the SSRT to the multi-frequency (4\,--\,8~GHz) \textit{Siberian Radio Heliograph} (SRH: \citealp{Lesovoi2012, Lesovoi2014}). The remaining part of the original antenna array keeps on the initial-principle observations. \begin{acks} We thank A.T.~Altyntsev for the idea of this study and useful remarks, and our colleagues for their contribution, efforts, and assistance. The data used here are provided by the SSRT team in Badary. S.A.~Anfinogentov has substantially contributed to the collaborative development of the SSRT raw-data processing and calibrating software and assisted in computations. S.V.~Lesovoy developed the data acquisition system and the routine imaging software. We thank him and A.M.~Uralov for fruitful discussions. We appreciate the memories of T.A.~Treskov, one of the major developers of the SSRT, whose ideas helped us to implement the techniques described here, and N.N.~Kardapolova, who managed the SSRT observations for many years. We thank the reviewer for useful remarks. We are grateful to the instrumental teams of SOHO/MDI (ESA and NASA), GOES, USAF RSTN Network, and Nobeyama Radioheliograph. A.K. was supported by the Russian Foundation of Basic Research under grants 15-32-20504 mol-a-ved and 15-02-01089. \end{acks} \medskip \noindent {\footnotesize \textbf{Disclosure of Potential Conflicts of Interest} \quad The authors declare that they have no conflicts of interest.} | 16 | 9 | 1609.02256 |
1609 | 1609.07159_arXiv.txt | In the second paper of the series, we have modeled low frequency carbon radio recombination lines (CRRL) from the interstellar medium. Anticipating the LOw Frequency ARray (LOFAR) survey of Galactic CRRLs, we focus our study on the physical conditions of the diffuse cold neutral medium (CNM). We have used the improved departure coefficients computed in the first paper of the series to calculate line-to-continuum ratios. The results show that the line width and integrated optical depths of CRRL are sensitive probes of the electron density, gas temperature, and the emission measure of the cloud. Furthermore, the ratio of CRRL to the [CII] at 158~$\mu$m line is a strong function of the temperature and density of diffuse clouds. Guided by our calculations, we analyze CRRL observations and illustrate their use with data from the literature. | The interstellar medium (ISM) plays a central role in the evolution of galaxies. The formation of new stars slowly consumes the ISM, locking it up for millions to billions of years while stars, as they age, return much of their mass increased in metallicity, back to the ISM. Stars also inject radiative and kinetic energy into the ISM and this controls the physical characteristics (density, temperature and pressure) as well as the dynamics of the gas as revealed in observed spectra. This interplay of stars and surrounding gas leads to the presence of distinct phases (e.g. \citealt{field1969,mckee1977}). Diffuse atomic clouds (the Cold Neutral Medium, CNM) have densities of about $50~\mathrm{cm^{-3}}$ and temperatures of about $80~\mathrm{K}$, where atomic hydrogen is largely neutral but carbon is singly ionized by photons with energies between $11.2~\mathrm{eV}$ and $13.6~\mathrm{eV}$. The warmer ($\sim8000~\mathrm{K}$) and more tenuous ($\sim0.5~\mathrm{cm^{-3}}$) intercloud phase [the Warm Neutral medium (WNM) and Warm Ionized Medium (WIM)] is heated and ionized by FUV and EUV photons escaping from HII regions \citep{wolfire2003}. While these phases are often considered to be in thermal equilibrium and in pressure balance, the observed large turbulent width and presence of gas at thermally unstable, intermediate temperatures may indicate that kinetic energy input is important. Thermally unstable gas could indicate that the gas does not have sufficient time to cool between subsequent passages of a shock or after intermittent dissipation of turbulence (e.g. \citealt{kim2011}). In addition, the ISM also hosts molecular clouds, where hydrogen is in the form of $\mathrm{H_2}$ and self-gravity plays an important role. All of these phases are directly tied to key questions on the origin and evolution of the ISM, including energetics of the CNM, WNM and the WIM; the evolutionary relationship of atomic and molecular gas; the relationship of these ISM phases with newly formed stars; and the conversion of their radiative and kinetic power into thermal and turbulent energy of the ISM (e.g. \citealt{cox2005,elmegreen2004, scalo2004,mckee2007}). The diffuse interstellar medium has been long studied using, in particular, the 21 cm hyperfine transition of neutral atomic hydrogen (e.g. \citealt{kulkarni1987,heilesandtroland2003a}). These observations have revealed the prevalence of a two phase structure in the interstellar medium of cold clouds embedded in a warm intercloud medium. However, it has been notoriously difficult to determine the physical characteristics (density, temperature) of these structures in the ISM as HI by itself does not provide a good probe. Optical and UV observations of atomic lines can provide the physical conditions but are by necessity limited to pinpoint experiments towards bright background sources. However, with the opening up of the low frequency radio sky with modern interferometers such as the Low Frequency ARray for Radioastronomy (LOFAR, \citealt{vhaarlem2013}), Murchison Wide field Array \citep{tingay2013}, Long Wavelength Array \citep{ellingson2013} and, in the future, the Square Kilometer Array (SKA), systematic surveys of low frequency ($\nu \lesssim 300~\mathrm{MHz}$) Carbon Radio Recombination Lines (CRRLs) have come in reach and these surveys can be expected to quantitatively measure the conditions in the emitting gas \citep{oonk2015a}. Carbon has a lower ionization potential (11.2 eV) than hydrogen and can be ionized by radiation fields in regions where hydrogen is largely neutral. Recombination of carbon ions with electrons to high Rydberg states will lead to CRRLs in the sub-millimeter to decameter range. CRRLs have been observed in the interstellar medium of our Galaxy towards two types of clouds: diffuse clouds (e.g.: \citealt{konovalenko1981, erickson1995, roshi2002, stepkin2007,oonk2014}) and photodissociation regions (PDRs), the boundaries of HII regions and their parent molecular clouds (e.g.: \citealt{natta1994, wyrowski1997, quireza2006}). Recently, \citet{morabito2014} discovered extragalactic CRRLs associated with the nucleus of the nearby starburst galaxy, M82. Theoretical models for CRRLs were first developed by \citet{watson1980} and \citet{walmsley1982}, including the effects of dielectronic recombination \footnote{As in \citet{salgado2015}, following common usage in the astronomical literature, we refer to this process as dielectronic recombination rather than the more appropriate dielectronic capture.} with the simultaneous excitation of the ${^2}P_{3/2}$ fine-structure level and later extended by \citet{ponomarev1992} and by \citet{payne1994}. However, these studies were hampered by the limited computer resources available at that time. In the coming years, we will use LOFAR to carry out a full northern hemisphere survey of CRRL emitting clouds in the Milky Way. This will allow us to study the thermal balance, chemical enrichment and ionization rate of the cold neutral medium from degree-scales down to scales corresponding to individual clouds and filaments in our Galaxy. Furthermore, following the first detection of low-frequency CRRLs in an extragalactic source (M82; \citealt{morabito2014}) we will also use LOFAR to perform the first flux limited survey of CRRLs in extragalactic sources. Given the renewed observational interest in CRRLs, a new theoretical effort seems warranted. In the first paper of this series, (\citealt{salgado2015}, hereafter Paper~I), we studied the level population of hydrogenic atoms including the effects of dielectronic recombination in carbon atoms. The level population of atoms, however, is not the only process that influences the strength of an observed line as radiative transfer effects can alter the strength/depth of an observed line. In this paper, we use the results of Paper~I to develop CRRLs as a tool to derive the physical conditions in the emitting gas. In this, we will focus on cold diffuse clouds as these are expected to dominate the low frequency CRRL sky. The paper is organized as follows: in Section \ref{section_radtransf} we review radiative transfer theory in the context of radio recombination lines. We review the line broadening mechanisms of CRRLs in Section \ref{section_lineprofile}. In Section \ref{section_results}, we present the results of our models and compare them with observations from the literature and provide guidelines to analyze such observations. Finally, in Section \ref{section_conclusions}, we summarize our results and provide the conclusions of our work. | In this paper we have analyzed carbon radio recombination line observations. Anticipating the LOFAR CRRL survey, we focus our study in the low frequency regime, corresponding to transitions between lines with high principal quantum number. We have studied the radiative transfer of recombination lines and the line broadening mechanisms in the most general form. Our results show that line widths provide constraints on the physical properties of the gas. At high frequencies the observed line widths provide limits on the gas temperature and on the turbulent velocity of the cloud. At low frequencies, observed line widths provide constraints on the electron density of the intervening cloud and on the radiation field that the cloud is embedded in. Using the departure coefficients obtained in Paper~I, we analyzed the behavior of the lines under the physical conditions of the diffuse ISM. Integrated optical depths provide constraints on the electron density, electron temperature and the emission measure or size of the cloud. The use of CRRLs together with [CII] at 158~$\mu\mathrm{m}$ can constrain the temperature. As an illustration of the use of our models, we have analyzed existing data in low frequency CRRLs towards Cas~A and the inner galaxy to derive physical parameters of the absorbing/emitting clouds \citep{payne1994,stepkin2007,erickson1995}. Our models predict that detailed studies of CRRLs should be possible with currently available instrumentation. By using realistic estimates for the properties of the diffuse ISM we obtain optical depths that are within the capabilities of LOFAR and of the future Square Kilometer Array \citep{oonk2015a}. Given the clumpy nature of the ISM, we encourage observations with high angular resolution. Observations with large beams are biased towards line of sights with large optical depth and narrow lines, and these happen to be clouds of low density for a given temperature. High spectral resolution is also encouraged in order to distinguish multiple components along the line of sights. Once the temperature and the density have been determined, the observed intensities yield the C$+$ column density which can be combined with the HI column density from 21 cm observations to determine the gas phase carbon abundance. The main conclusions of our work are: 1) CRRLs provide a powerful probe of the physical conditions of diffuse interstellar clouds. 2) Meaningful constraints on gas properties can be derived from combining information on the location of the transition from emission to absorption, $\alpha$-to-$\beta$ ratios and $\alpha$-line ratios spread in frequency. Further limits are provided by the low frequency line width. 3) Comparison of CRRLs with [CII] 158 $\mu \mathrm{m}$ line measured by COBE \citep{bennett1994}, BICE \citep{nakagawa1998} and Herschel (GOT C+;\citealt{pineda2013}); in addition to new observations with the German Receiver for Astronomy at Terahertz Frequencies (GREAT; \citealt{heyminck2012}) on board of SOFIA, will provide important constraints primarily on the temperature, but also aid in further constraining the density and size of diffuse clouds. | 16 | 9 | 1609.07159 |
1609 | 1609.05476_arXiv.txt | We apply the methodology developed in \cite{Li2014,Li2015} to BOSS DR12 galaxies and derive cosmological constraints from the redshift dependence of the Alcock-Paczynski (AP) effect. The apparent anisotropy in the distribution of observed galaxies arise from two main sources, the redshift-space distortion (RSD) effect due to the galaxy peculiar velocities, and the geometric distortion when incorrect cosmological models are assumed for transforming redshift to comoving distance, known as the AP effect. Anisotropies produced by the RSD effect are, although large, maintaining a nearly uniform magnitude over a large range of redshift, while the degree of anisotropies from the AP effect varies with redshift by much larger magnitude. We split the DR12 galaxies into six redshift bins, measure the 2-point correlation function in each bin, and assess the redshift evolution of anisotropies. We obtain constraints of $\Omega_m=0.290 \pm 0.053,\ \ w = -1.07 \pm 0.15$, which are comparable with the current constraints from other cosmological probes such as type Ia supernovae, cosmic microwave background, and baryon acoustic oscillation (BAO). Combining these cosmological probes with our method yield tight constraints of $ \Omega_m = 0.301 \pm 0.006,\ w=-1.054 \pm 0.025$. Our method is complementary to the other large scale structure probes like BAO and topology. We expect this technique will play an important role in deriving cosmological constraints from large scale structure surveys. | The current standard model of cosmology has been highly successful at reproducing the Universe on large scales. % From the temperature fluctuations in the cosmic microwave background (CMB), to the late time clustering of galaxies, the vacuum energy dominated cold dark matter model ($\Lambda$CDM) fits the data surprisingly well \citep{Planck2015,Anderson2013}. This result is all the more impressive considering both the underlying assumptions, such as homogeneity, isotropy, scale invariance of the primordial fluctuations, and the minimal set of cosmological parameters that are required. Nonetheless, these models produce the unsatisfactory prospect that we must include within our ontology both a vacuum energy that is much smaller than that predicted from quantum mechanics, or alternatively a new scalar field (dark energy) that has negative pressure \citep{SW1989,Riess1998,Perl1999,PR2003,Li2011}, and a new matter component, which is not contained within the standard $SU(3)\times SU(2) \times U(1)$ formulation of particle physics. With an over-abundance of models for both dark energy-like accelerated expansion and dark matter, it is crucial to obtain precise and model-independent measurements of the cosmic evolution, usually referred to as background observables. Two such observables are the angular diameter distance, $D_A$, and the Hubble factor, $H$. If these quantities can be measured at various redshifts and to a high degree of accuracy then our ability to differentiate between various competing models will be greatly increased. In the last few years there has been increasing interest in using the Alcock-Paczynski (AP) effect \citep{AP1979} in the large-scale clustering of galaxies to obtain constraints on $D_A$ and $H$ \citep{Guzzo2008,topology}. Assuming an incorrect cosmological model for the coordinate transformation from redshift space to comoving space produces residual geometric distortions. These distortions are induced by the fact that measured distances along and perpendicular to the line of sight are fundamentally different. Measuring the ratio of galaxy clustering in the radial and transverse directions provides a probe of this AP effect. There have been several methods proposed for applying the AP test to the large scale structure (LSS). The most widely adopted one uses anisotropic clustering \citep{Ballinger1996,Matsubara1996}, which has been used for the 2 degree Field Quasar Survey \citep{Outram2004}, the WiggleZ dark energy survey \citep{Blake2011}, the Sloan Digital Sky Survey-I/II (SDSS-I/II) Luminous Red Galaxy (LRG) survey \citep{Eisenstein et al. 2011,ChuangWang2012}, and the SDSS-III Baryon Oscillation Spectroscopic Survey (BOSS) \citep{Reid2012,Beutler2013,Linder2013,2014arXiv1407.2257S, 2014ApJ...781...96L, Alam2016, Beutler2016, Sanchez2016} The main caveat of this method is that, because the radial distances of galaxies are inferred from redshifts, AP tests are inevitably limited by redshift-space distortions (RSD) \citep{Ballinger1996}, which leads to apparent anisotropy even if the adopted cosmology is correct. The RSDs must be accurately modeled for the 2-point statistics of galaxy clustering. \cite{Marinoni2010} proposed using the symmetry properties of galaxy pairs. Unfortunately this method is also seriously limited by RSD. The peculiar velocity distorts the redshift and changes the apparent tilt angles of galaxy pairs. The effect depends on both redshift and underlying cosmology, and is rather difficult to model accurately \citep{Jennings2011}. \cite{Ryden1995} and \cite{LavausWandelt1995} proposed another method using the apparent stretching of voids. This approach has the advantage that the void regions are easier to model compared with dense regions, but has limitations in that it utilizes only low density regions of the LSS and requires large samples. \cite{Li2014} proposed another method utilizing the redshift dependence of AP effect to overcome the RSD problem. The anisotropies produced by RSD effect are, although very large, close to uniform in magnitude over a large range of redshift. Conversely, if cosmological parameters are incorrectly chosen, the LSS appear anisotropic and the degree of anisotropy varies with redshift. We used the {\it galaxy density gradient field} to characterize the anisotropies in LSS and tested the idea on Horizon Run 3 (HR3) N-body simulations \citep{horizonrun}, demonstrating that the method leads to unbiased estimation of the density parameter $\Omega_m$ and the dark energy equation of state (EoS) $w$. The same topic was revisited in \cite{Li2015}, but using the {\it galaxy two-point correlation function} (2pCF) as the statistical tool. The 2pCF as a function of angle, $\xi(\mu)$, is measured at different redshifts. Similar to \cite{Li2014}, we found that the RSD effect, although significantly distorting $\xi(\mu)$, exhibits much less redshift evolution compared to the amount of change in $\xi(\mu)$ due to incorrectly adopted cosmologies. When incorrect cosmological parameters are adopted, the shape of $\xi(\mu)$ appears anisotropic due to the AP effect, and the amplitude is shifted by the change in comoving volume; both effects have significant redshift dependence. We test the method using the 2pCF on mock surveys drawn from HR3 and find the constraints obtained are tighter than those from the methodology of \cite{Li2014}. The change of the comoving volume size is another consequence of an incorrectly adopted cosmology, and has motivated investigations constraining cosmological parameters from number counting of galaxy clusters \citep{PS1974,VL1996}. An obstacle in using the comoving volume for cosmological tests is the evolution of the number of target objects. The essential need for reducing the evolution effects in applying the test led \cite{topology} to propose a new method using the topology of LSS. Since the topology is a measure of intrinsic connectivity of structures, it is expected to be insensitive to non-linear gravitational evolution, type of density tracers, and RSD on large scales. This method has been applied to the WiggleZ Dark Energy Survey data by \cite{WiggleZtopoloy}, and to simulated BOSS samples \citep{Speare2015}. In this paper we apply our methodology to SDSS-III BOSS Data Release 12 (DR12) galaxies \citep{Reidetal:2016}. We take the 2pCF as statistical tool characterizing the anisotropic clustering and follow the procedure of \cite{Li2015} to conduct the analysis. We assume a flat Universe and constrain parameters of $\Omega_m$ and $w$. 2pCF is a mature statistic in cosmology and its optimal estimation and statistical properties are well understood. Compared with the density gradient field statistic, it leads to tighter constraints and is less affected by survey geometry. The outline of this paper is as follows. In Sec. 2 we describe the observational data used in this paper. In Sec. 3 we discuss the N-body simulations and mock galaxy catalogues that are used in this analysis. In Sec. 4 we briefly review the nature and consequences of the AP effect when performing coordinate transforms in a cosmological context. In Sec. 5 and Sec. 6, we describe our analysis method and present the cosmological constraints obtained from BOSS DR12 galaxies. We conclude in Sec. 7. | We apply the methodology developed in \cite{Li2014,Li2015} to BOSS DR12 galaxies. In LSS surveys, the observed galaxy distribution appears anisotropic due to two reasons: the contamination of galaxy redshifts due to the galaxy peculiar velocities, known as the RSD effect, and the error in the distance due to inaccurate cosmological parameters. % \cite{Li2014} reported that anisotropies produced by RSD effect are, although large, close to uniform in magnitude over a large range of redshift, while the degree of anisotropies introduced by AP varies with redshift. Thus we can use the redshift dependence of the anisotropic clustering of galaxies to constrain cosmological parameters without being much affected by RSD. As in \cite{Li2015}, we investigate the redshift-dependence of clustering anisotropy by examining the 2pCF. The 2pCF measured along different angular directions was characterized by the function $\hat \xi_{\Delta s}(\mu)$. When the cosmological parameters governing the expansion history of the universe are incorrectly chosen, the shape of this function evolves with redshift. We split the DR12 galaxies into six redshift bins, measure the 2pCF in each redshift bin, and search for the underlying true values of $\Omega_m$ and $w$ of our Universe % by requiring minimal redshift evolution of $\hat \xi_{\Delta s}(\mu)$ across the six redshift bins. We obtain tight constraints of $\Omega_m=0.290 \pm 0.053,\ \ w = -1.07 \pm 0.15$ % from our method alone. The constraints on $\Omega_m$ and $w$ from our AP method are comparable with or tighter % than the other cosmological probes of SNIa, CMB, BAO, and $H_0$. For the direction of degeneracy, % our method is similar to SNIa and orthogonal to CMB and BAO+$H_0$. Combining the results of our method with those of other cosmological probes, we obtain tight constraints $\Omega_m = 0.301 \pm 0.006,\ w=-1.054 \pm 0.025$. \subsection{Comparison with other LSS probes} Our method uses the anisotropic galaxy on scales from 6 to 40 $h^{-1}$Mpc. Constrains on cosmological parameters are obtained from the redshift dependence of $D_A(z) H(z)$. As a comparison, the BAO method uses the BAO feature in the clustering of galaxies on scales of 100-150 $h^{-1}$Mpc created by the oscillation of the baryon-photon plasma in the early Universe. Measuring the BAO feature in 1D or 2D yields measurements of $D_V$ or $D_A$ and $H$ at some representative redshifts. The AP methods proposed to date, such as those using galaxy pairs and voids, measure the rate of geometric distortion and are sensitive to $D_A(z) H(z)$, while our method uses its the redshift dependence. These methods can be combined together to fully utilize the physics of AP test. Also, reducing the RSD effect through the redshift dependence could be applicable to these methods. The topology method proposed by \cite{topology} uses the redshift evolution of the volume effect and is sensitive to the quantity $D_A(z)^2 / H(z)$. Combing this method with ours can yield separate constraints on $D_A$ and $H$ for the same observational sample. Constraints from the other statistical measures, such as the distribution function of size or richness of LSS, can be also combined \citep{Park2012,Park2015}. Recently, \cite{MS2016} developed a novel method constraining cosmological parameters based on the high-level similarity of the emission measure in the cluster outskirts. In incorrect cosmologies, the emission measure from clusters exhibits redshift dependence. Utilizing a sample of 320 galaxy clusters ($0.056<z<1.24$) observed with Chandra, they achieve tight cosmological constraints comparable to ours. The idea of this novel technique is to some extent similar to our method (seeking for the conservation of geometric quantity with redshift), and could have promising future. The above geometric methods can be combined with probes of RSD \citep{Guzzo2008,Blake2011a,Beutler2012,Reid2012b,Samushia2012,GM2016b,Li2016} to have a more complete study of LSS galaxy clustering. See \cite{DHW2013} for a review of more LSS probes of dark energy. \subsection{Room for improvements} This paper is the first application of this redshift dependent AP test to observed LSS data. There remains considerable opportunity for improving the analysis methodology, e.g., the optimized schemes of redshift binning and optimized choices of the scales of clustering (the values $s_{\rm min}$ and $s_{\rm max}$). In this paper, the 2pCF is characterized by $\hat \xi_{\Delta s}(\mu)$. It should be more advantageous to use the redshift evolution of 2pCF in two dimensions, i.e., $\hat \xi(s,\mu)$, to capture the full information. It is also necessary to combine the information in the higher-order statistic beyond the two point functions. As pointed out in Sec. \ref{sec:caveats}, more theoretical and numerical studies on the redshift evolution of the RSD effect will remove the remaining small uncertainties in our results. To avoid the difficulty of modeling galaxy bias we dropped the information of strength of clustering through normalizing the amplitude of $\xi$. In the case that we have good knowledge of galaxy bias, one can utilize the amplitude of $\xi$ to probe the redshift dependence of volume effect and obtain much tighter constraints. \cite{Li2015} showed that the area of constraining regions in $\Omega_m$ and $w$ space from the redshift dependence of the volume effect is 3.5 times smaller than that from the redshift dependence of the AP effect alone. In this analysis, by reducing the RSD effect, we are able to use the galaxy clustering down to 6 $h^{-1}$Mpc. This is a major advance in extracting the cosmological information on small scales where galaxy clustering is strong and there are a lot of independent structures. To make sure that the derived cosmological constraint is robust, in the procedure of systematics correction, one should construct many mock surveys in which the RSD and other systematic effects are reliably modeled. It would also be helpful to investigate the effect of galaxy properties. For example, in case of a dense survey containing galaxies with various properties, one can split the galaxy sample into subsamples with different galaxy properties, derive cosmological constraints from these subsamples, and check the consistency of the results. \subsection{Synergies with future observations} The constraining power of our method is proportional not only to the size, but also to the redshift coverage of the galaxy sample. In this analysis we utilize 1\,133\,326 BOSS DR12 galaxies covering $0.15< z < 0.693$. Future redshift surveys such as eBOSS \citep{eBOSS}, EUCLID \citep{EUCLID}, and DESI \citep{DESI} will have larger sample sizes and wider redshift coverage. Our method is expected to yield much tighter cosmological constraints and can be applied to constrain wider classes of cosmological models when these data are available. In this analysis we assume a flat Universe to constrain $\Omega_m$ and $w$. Our method should be sensitive to all other cosmological parameters governing the cosmic expansion history, e.g., the curvature of the Universe, and the other dark energy parameters. Overall, the future of our method is extremely promising. With the progressive development of LSS experiments, we expect it to play an important role in deriving cosmological constraints from LSS surveys. | 16 | 9 | 1609.05476 |
1609 | 1609.02060_arXiv.txt | Motivated by the ever increasing pursuit of science with the transient sky (dubbed Time Domain Astronomy or TDA), we are fabricating and will commission a new deployable tertiary mirror for the Keck I telescope (K1DM3) at the W.M. Keck Observatory. This paper presents the detailed design of K1DM3 with emphasis on the opto-mechanics. This project has presented several design challenges. Foremost are the competing requirements to avoid vignetting the light path when retracted against a sufficiently rigid system for high-precision and repeatable pointing. The design utilizes an actuated swing arm to retract the mirror or deploy it into a kinematic coupling. The K1DM3 project has also required the design and development of custom connections to provide power, communications, and compressed air to the system. This NSF-MRI funded project is planned to be commissioned in Spring 2017. | \label{sec:intro} % A major thrust of astronomy in the 21st century is to study, observationally and with theoretical inquiry, time-variable phenomena in the night sky. This area is broadly referred to as time domain astronomy (TDA) and its high scientific priority was established by the Astro2010 report (National Research Council, 2010). Their highest recommendation for large telescope ground-based observing, for example, was to build the Large Synoptic Survey Telescope (LSST). In advance of that ambitious project, several projects are using wide-field cameras to image large areas of the sky at high cadence. This includes the partially NSF-sponsored Palomar Transient Factory (PTF) and its NSF/MSIP funded follow-on (ZTF), and the Pan-STARRS surveys which repeatedly image the full northern sky, finding hundreds of new transient phenomena on every clear night. These surveys are discovering thousands of supernovae, immense samples of asteroids and near-Earth objects, variable stars of diverse nature, flaring phenomena, and other exotic sources. These advances in TDA observing at optical wavelengths follow decades of TDA science performed at higher energies from space. Indeed, the first astronomical sources detected with $\gamma$-rays were themselves transient phenomena: the so-called $\gamma$-ray bursts (GRBs). Satellites like NASA's Swift and Fermi monitor $\approx \pi$ steradians of the sky, scanning for transient and variable high-energy events. The focus of most previous and on-going TDA projects has been wide-field imaging of the sky in search of rare and new classes of events. To fully explore and exploit the astrophysics of newly discovered sources, however, one must establish the redshift and/or the type of object responsible. Optical and infrared wavelengths remain the most powerful and efficient passbands to perform the required spectroscopy. This is the primary role of large, ground-based observatories in TDA science. Recognizing their value, several 8\,m-class observatories have established very effective observing strategies (generally at great expense) to perform such science. Both the Gemini and European Southern Observatories (ESO) designed their largest telescopes with systems that could rapidly feed any of the available foci. Furthermore, they designed queue operations to enable rapid responses to targets-of-opportunity (ToOs) and programs that repeatedly observe a source for short intervals at high cadence. Successes of this model include time-resolved spectroscopy of varying absorption lines from a GRB afterglow on minute time-scales and high-cadence monitoring of the Galactic center to recover high fidelity orbital parameters. The W.M. Keck Observatory (WMKO) boasts twin 10\,m telescopes, currently the largest aperture, fully-operational optical/IR telescopes. Over the course of the past ~20 years, we have successfully instrumented each telescope with high-throughput imagers and spectrometers spanning wavelengths from the atmospheric cutoff to several microns. The Keck I (K1) telescope hosts the Nasmyth-mounted HIRES spectrograph, one of the primary tools for obtaining high-resolution visible wavelength spectra from TDA observations. K1 also hosts the Nasmyth-mounted Keck I adaptive optics (AO) system with a high-performance laser guide star (LGS) system, and now hosts the near-IR integral field spectrograph OSIRIS, a key tool for synoptic observations of the Galactic center. Two K1 instruments are used at Cassegrain: the LRIS multi-object visible wavelength spectrograph and the near-IR multi-object spectrograph and imager MOSFIRE. This is a unique instrument suite, especially within the U.S. community: the HIRES spectrometer is the only echelle spectrometer on a large aperture telescope in the northern hemisphere; LRIS provides extremely sensitive spectroscopy especially at blue ($< 4000$\AA) and red ($> 8000$\AA) wavelengths, and MOSFIRE represents a unique capability for multi-slit, near-IR spectroscopy in the northern hemisphere. By its nature, TDA science demands a more nimble and flexible approach to observations than the traditionally, classically-scheduled observing which has been the standard at WMKO. Most TDA programs require observations made with a specific instrument at specific times, while classical scheduling on a telescope with multiple instrument configurations may mean that the desired instrument will not be available for the TDA program. In the current configuration of the Keck telescopes a removable module, called the tertiary module, which contains the telescope tertiary mirror (M3), is used to support observations with Nasmyth and bent Cassegrain mounted instruments. % The desired instrument along the elevation axis ring is selected by rotating the tertiary mirror around the telescope optical axis. To install and use a Cassegrain mounted instrument, the tertiary module must be removed from the telescope which is a process typically requires daycrew. The K1 deployable tertiary (K1DM3) will directly address these shortcomings, enabling WMKO to fully and vigorously participate in TDA science with its unique set of K1 instruments. As the driving paradigm in observational astronomy shifts from a passive, static sky to one that displays dramatic changes on a nightly basis, it is critical to enhance our technical capabilities in this arena. The K1DM3 will increase the flexibility for ToO and Cadence observations with the Nasmyth, bent Cassegrain, and whichever Cassegrain instrument is installed in the telescope, without requiring any configuration change other than rotating the tertiary mirror to the appropriate focal station or retracting the mirror from the telescope beam. The K1DM3 will also reduce the time required for telescope reconfigurations by eliminating the need to remove or install the tertiary mirror module. We proposed successfully to the National Science Foundation (NSF) Major Research Instrumentation (MRI) Program in 2013 for funding of the K1DM3 project. We received the full award requested and the total project budget (\$2.1M USD) includes a 30\% cost-share from WMKO, the UC Observatories, and the University of California Santa Cruz. The project entered its Detailed Design phase in October 2015 and this paper presents the Detailed Design. The K1DM3 device will enable astronomers to swap between any of the foci on Keck 1 in under 2 minutes, both to monitor varying sources (e.g. stars orbiting the Galactic center) and rapidly fading sources (e.g. supernovae, flares, gamma-ray bursts). The design consists of a passive wiffle tree axial support system and a flexure-rod lateral support system with a 4.7\,arcminute field-of-view mirror. The mirror assembly is inserted into the light path with an actuation system and it relies on kinematic couplings for achieving repeatable, precise positioning. The actuation system may rotate (partially when retracted or fully when deployed) on two bearings mechanized with a pair of drive motors. It is our goal to commission K1DM3 at WMKO by March 2017. Figure~\ref{fig:K1DM3_overview} shows the overall configuration of the K1DM3 module and the module installed in the tertiary tower. The K1DM3 module consists of a light-weighted fixed outer drum and a moveable inner drum. The inner drum is supported at each end by 4-point contact ball bearings. The lower bearing has a ring gear that is driven with a pinion gear by a servo motor system. An absolute position encoder is used to measure the position of the rotating drum. The tertiary mirror is supported by axial and lateral supports attached to a whiffle tree structure. This whiffle tree connects the mirror and support structure to a swing arm system. In turn, this swing arm moves the mirror between the deployed and retracted positions, driven by two linear actuators. The top of the drum supports the swing arm in the deployed position through a bipod structure with two defining points (at the right side of the figure) and a third defining point at the hinge point of the swing arm (the third defining point is not visible in the figure). The swing arm is locked in the deployed position by a set of 4 clamping mechanisms. No power is required to maintain the mirror in either the deployed or retracted positions. In order to set the swing arm into the kinematics, the deployment process will be performed at the elevation angle where the kinematics are oriented normal to gravity (\depang). We will retract the mirror at specific rotation angles, one of two positions where power and ethernet is supplied to the swing arm. Full rotation of the module drum is possible with the mirror deployed. Interference with items at the top of the tertiary tower limit the rotation when K1DM3 is retracted. When the mirror is deployed, there are six positions used to direct the light to one of the two Nasmyth focal stations or one of the four bent Cassegrain positions. Each of these deployed positions is held by a detent mechanism engaging a v-groove. The detent mechanism is engaged by a pneumatic cylinder and retracted by a spring. The K1DM3 module is inserted into the tertiary tower from the telescope's Cassegrain platform and moved through the tertiary tower to its operating position on a pair of rails. Guide rollers mounted on the outer drum support the module on the tracks. When the module is installed in the tower it is held in position using three defining point mechanisms equipped with kinematic mounting points that are engaged and disengaged by three air motors. The kinematic mounting points ensure repeatable positioning. K1DM3 is designed to be coated with Al at WMKO. \begin{figure}[h!] \begin{center} \vskip -0.1in \includegraphics[width=3.0in]{K1DM3_overview.pdf} \includegraphics[width=3.0in]{K1DM3_in_tower.pdf} \vskip -0.10in \caption[(left) Schematic of the K1DM3 system. (right) K1DM3 module installed in the Keck~I tertiary tower.]{\footnotesize (left) Schematic of the K1DM3 system. (right) K1DM3 module installed in the Keck~I tertiary tower. }\label{fig:K1DM3_overview} \vskip -0.1in \end{center} \end{figure} | 16 | 9 | 1609.02060 |
|
1609 | 1609.00545_arXiv.txt | {We study the star formation quenching mechanism in cluster galaxies by fitting the spectral energy distribution of the \textit{Herschel} Reference Survey, a complete volume-limited $K$-band-selected sample of nearby galaxies including objects in different density regions, from the core of the Virgo cluster to the general field. The spectral energy distributions of the target galaxies are fitted using the CIGALE SED modelling code. The truncated activity of cluster galaxies is parametrised using a specific star formation history with two free parameters, the quenching age $QA$ and the quenching factor $QF$. These two parameters are crucial for the identification of the quenching mechanism which acts on long timescales if starvation while rapid and efficient if ram pressure. To be sensitive to an abrupt and recent variation of the star formation activity, we combine in a new way twenty UV to far infrared photometric bands with three age-sensitive Balmer line absorption indices extracted from available medium-resolution ($R$ $\sim$ 1000) integrated spectroscopy and with H$\alpha$ narrow band imaging data. The use of a truncated star formation history significantly increases the quality of the fit in HI-deficient galaxies of the sample, thus in those objects whose atomic gas content has been removed during the interaction with the hostile cluster environment. The typical quenching age of the perturbed late-type galaxies is $QA$ $\lesssim$ 300 Myr whenever the activity of star formation is reduced by 50\% $<$ $QF$ $\leq$ 80\% and $QA$ $\lesssim$ 500 Myr for $QF$ $>$ 80\%, while that of the quiescent early-type objects is $QA$ $\simeq$ 1-3 Gyr. The fraction of late-type galaxies with a star formation activity reduced by $QF$ $>$ 80\% ~ and with an HI-deficiency parameter $HI-def$ $>$ 0.4 drops by a factor of $\sim$ 5 from the inner half virial radius of the Virgo cluster ($R/R_{vir}$ $<$ 0.5), where the hot diffuse X-ray emitting gas of the cluster is located, to the outer regions ($R/R_{vir}$ $>$ 4). The efficient quenching of the star formation activity observed in Virgo suggests that the dominant stripping process is ram pressure. We discuss the implication of this result in the cosmological context of galaxy evolution. } {} {} {} {} {} | Environment plays a major role in shaping galaxy evolution. Since the seminal work of Dressler (1980) it became evident that galaxies in rich environments are systematically different than those located in the field. Dense environments are dominated by early-type galaxies, while low density regions by spirals and irregulars (e.g. Dressler 1980; Whitmore et al. 1993; Dressler et al. 1997). It has also been shown that massive local clusters are currently accreting gas-rich, star forming systems (e.g. Tully \& Shaya 1984; Colless \& Dunn 1996) whose physical properties systematically change once they reach the densest regions. Indeed, it is now widely recognised that late-type galaxies located in rich clusters are generally HI-deficient (Haynes et al. 1984; Gavazzi 1987; Cayatte et al. 1990; Solanes et al. 2001; Gavazzi et al. 2005, 2006a). There is a growing evidence indicating that they also lack of molecular gas (Fumagalli et al. 2009; Boselli et al. 2014b) and dust (Cortese et al. 2012a) with respect to similar objects in the field. The angular resolution of the HI, CO, and far infrared images now available thanks to interferometric observations in the radio domain and to the superior quality of the instruments on board of \textit{Herschel} revealed that cluster galaxies have truncated gaseous and dust discs (Fumagalli et al. 2009, Boselli et al. 2014b, Cortese et al. 2012a, Davis et al. 2013), suggesting that the mechanism responsible for their stripping acts outside in. The activity of star formation of the late-type galaxies in clusters is also systematically reduced with respect to that of field objects (Kennicutt 1983; Gavazzi et al. 1998, 2002a, 2006b) and it is limited to the inner disc (Koopmann \& Kenney 2004; Boselli \& Gavazzi 2006; Cortese et al. 2012b; Fossati et al. 2013). The correlation between the molecular gas content and the activity of star formation (e.g. Bigiel et al. 2008), generally called Schmidt law (Schmidt 1959; Kennicutt 1998a), can easily explain the observed truncation of the star forming discs of these perturbed objects. Different mechanisms have been proposed in the literature to explain the transformation of galaxies in rich environments and the formation of the red sequence (e.g. Boselli \& Gavazzi 2006, 2014). They include the gravitational interaction between galaxies (Merritt 1983) or that with the potential well of the cluster as a whole (Byrd \& Valtonen 1990), or their combined effect generally called ``galaxy harassment'' (Moore et al. 1998). Other possible mechanisms are those related to the interaction of the galaxy interstellar medium with the hot ($T$ $\sim$ 10$^7$-10$^8$ K) and dense ($\rho_{ICM}$ $\sim$ 10$^{-3}$ cm$^{-3}$) diffuse gas trapped within the potential well of the clusters observable in X-rays (Sarazin et al. 1986). These include the ram pressure (Gunn \& Gott 1972) and the viscous stripping (Nulsen 1982) exerted by the intracluster medium on galaxies moving at high velocity ($\sim$ 1000 km s$^{-1}$) within the cluster, or the thermal heating of the galaxy ISM once in contact with the hot X-ray emitting gas of the cluster (Cowie \& Songaila 1977). Since a large fraction of cluster galaxies is accreted via small groups, the perturbing mechanisms can start to shape galaxy evolution well before the galaxy is within the massive cluster (pre-processing, e.g. Dressler 2004). Finally, it is possible that under some circumstances the ICM only perturbs the hot gas in the galaxy halo unaffecting the cold component on the disc. The cold component, however, not replenished by fresh infalling material, is later exhausted by the star formation activity of the galaxy itself (starvation; Larson et al. 1980). The identification of the perturbing mechanism responsible for the quenching of the star formation activity of galaxies in high density regions is becoming one of the major challanges of modern extragalactic astronomy. The identification of the dominant perturbing mechanism, which is expected to change with the mass of galaxies and with the properties of the overdensity region, is crucial for cosmological simulations and semi analytical models of galaxy evolution. Different mechanisms may have different effects on the morphology and internal dynamics of the stellar component, making discs thicker and increasing the bulge-to-disc ratio (gravitational perturbations), while other are expected to decrease the surface brightness of the perturbed disc (starvation). At present, models and simulations often overestimate the fraction of quiescent galaxies along the red sequence (Kang \& van den Bosch 2008; Font et al. 2008; Kimm et al. 2009; Fontanot et al. 2009; Guo et al. 2011; Weinmann et al. 2011; Wang et al. 2012; Hirschmann et al. 2014; see however Henriques et al. 2015), suggesting that the physical prescriptions used to reproduce the environmental quenching is still poorly understood. An accurate identification of the dominant mechanism is also crucial for the physical understanding of the stripping process, for the calibration of tuned hydrodynamic simulations, and for the charactrisation of the effects that the mechanism has on perturbed galaxies and on the stripped material. Critical for constraining the perturbing mechanism is the identification with observations of its acting radius within high-density regions and of the quantification of the timescale necessary to significantly remove the gas and affect the star formation process (efficiency). The analysis of nearby and high redshift clusters still gives discordant results. The detailed study of representative objects in nearby clusters such as Virgo and Coma based on dynamical modelling of their HI and CO gas kinematics (e.g. Vollmer et al. 2004), of the radial variation of their star formation properties derived from integral field unit (IFU) optical spectroscopy (e.g. Crowl \& Kenney 2008), or by the comparison of multifrequency data with tuned chemo-spectrophotometric models of galaxy evolution (e.g. Boselli et al. 2006) indicate a recent and rapid truncation of their star formation activity consistent with a ram pressure stripping scenario. The analysis of large statistical samples of galaxies extracted from several surveys such as the SDSS, GAMA, or GALEX, or targeted observations of nearby clusters and groups combined with the results of cosmological simulations or semi analytical models of galaxy evolution rather suggest a slow and long quenching process typical of starvation (McGee et al. 2009; Wolf et al. 2009; von der Linden et al. 2010; De Lucia et al. 2012; Wheeler et al. 2014; Taranu et al. 2014; Haines et al. 2015; Paccagnella et al. 2016). Other works suggest a bimodal evolution, with an inefficient quenching at early epochs, when galaxies become satellites of more massive halos, and then a rapid quenching when these still relative small systems are accreted in massive clusters (pre-processing; e.g. Wetzel et al. 2012, 2013; Muzzin et al. 2012; Wijesinghe et al. 2012). The identification of the perturbing mechanism requires an accurate reconstruction of the star formation history of galaxies in different environments. This is generally done using photometric and spectroscopic data to characterise the stellar emission in the UV to near-infrared spectral domain. The typical colours of galaxies in these bands are sensitive to their underlying stellar populations, and become redder whenever the star formation activity decreses because perturbed by the surrounding environment. The reconstruction of the star formation history of perturbed galaxies using these sets of data, however, is limited by the fact that other physical mechanisms such as dust attenuation and metallicity might affect the colours of galaxies in a similar way. To overcome this problem and limit any possible degeneracy between the effects of dust attenuation and aging of the stellar populations astronomers developed UV to far-infrared spectral energy distribution (SED) fitting codes. These codes, by measuring the total energy emitted by dust in the far-infrared, are able to quantify in a self consistent way the dust attenuation on the stellar emission, and are thus now widely used to study the star formation history of different samples of galaxies (e.g. GRASIL: Silva et al. 1998; MAGPHYS: da Cunha et al. 2008). In the last years a huge number of multifrequency data, from the UV to the radio centimetre, have been gathered for a complete volume-limited sample of nearby galaxies, the \textit{Herschel} Reference Survey (HRS). This sample includes galaxies in different density regions, from the core of the Virgo cluster to the general field, and is thus perfectly suited for environmental studies (Boselli et al. 2010a). These data are perfectly suited to be fitted with CIGALE, a modelling code designed to reconstruct the star formation history of galaxies through an analysis of their UV to far infrared SED (Noll et al. 2009). In this work we improve this SED fitting code to combine in an original manner twenty photometric bands imaging data with three age sensitive spectroscopic indices (Balmer absorption lines) derived from integrated spectroscopy (Boselli et al. 2013) and narrow band H$\alpha$ imaging data (Boselli et al. 2015) with the aim of characterising the quenching star formation episode. We already explored the use of the SED fitting analysis to reproduce the variations on the star formation history of cluster galaxies using only broad band imaging photometry in Ciesla et al. (2016). The present work is thus a further developement of this technique. To date the combination of photometric and spectroscopic data in the SED fitting analysis has been limited to the stellar emission (Pacifici et al. 2012, 2015; Newman et al. 2014; Thomas et al. 2016; Chevallard \& Charlot 2016; Lopez Fernandez et al. 2016). Our work is the first where this technique is extended to a much wider spectral domain, from the UV to the far-infrared, where the far-infrared emission is used through an energy balance to break any possible degeneracy due to dust attenuation in the observed stellar emission. The paper is structured as follow: we describe the sample in sect. 2, the multifrequency data in sect. 3, and the SED fitting technique and procedure in sect. 4. Section 4 also includes several tests done on mock samples to check the reliability of the output parameters. In sect. 5 we apply the SED fitting code to a dozen of representative galaxies for which independent results are available in the literature, and in sect. 6 to the whole HRS sample. The analysis is done in sect. 7 and the discussion in sect. 8. In Appendix A we discuss the effects of the adoption of different star formation histories on the derived quenching parameters. For a fair comparison with models and simulations we recall that the typical distance of the main body of the Virgo cluster is 17 Mpc (Gavazzi et al. 1999; Mei et al. 2007) and its total dynamical mass $M_{200}$ (1.4-4.2) $\times$ 10$^{14}$ M$_{\odot}$ (McLaughlin 1999; Urban et al. 2011; Nulsen \& Bohringer 1995; Schindler et al. 1999). | We study the star formation history of the \textit{Herschel} Reference Survey, a $K$-band-selected, volume-limited complete sample of nearby galaxies using the CIGALE SED fitting code. The sample includes objects in the Virgo cluster and is thus perfectly suited for understanding the role of the environment on galaxy evolution. To quantify the perturbation induced by the interaction with the hostile environment on the star formation activity of cluster galaxies, we adopt a truncated star formation history where the secular evolution is parametrised using the chemo-spectrophotometric physically justified models of Boissier \& Prantzos (2000). To constrain any possible abrupt variation of the star formation activity of the Virgo cluster galaxies, we combine the UV to far infrared photometric data (20 bands) with age sensitive Balmer absorption line indices extracted from medium resolution ($R$ $\sim$ 1000) integrated spectroscopy (3 bands) and H$\alpha$ imaging data (1 band). H$\alpha$ fluxes are compulsory to constraint variations occurred at very recent epochs ($\lesssim$ 100 Myr). The best fit to the data gives the quenching factor $QF$ and the quenching age $QA$. The first parameter quantifies how much the star formation activity has been reduced during the interaction ($QF$=0 for unperturbed systems, $QF$ = 1 when the star formation activity has been completely stopped), while $QA$ gives the lookback time of the epoch of the quenching episode. We checked the reliability of the output parameters using mock catalogues of simulated galaxies and different star formation histories. The analysis of the sample brought to the following results:\\ 1) The quality of the SED fitting significantly increases using a truncated star formation history in all the HI-deficient galaxies of the sample, where the interaction with the hostile environment removed a large fraction of the cold gas necessary to feed star formation.\\ 2) In these HI-deficient objects the activity of star formation is reduced by a factor of 50\% $\leq$ $QF$ $<$ 80\% on timescales of $\simeq$ 135 Myr, and $QF$ $\geq$ 80\% on timescales of $\simeq$ 250 Myr, while it is fully stopped after $\simeq$ 1.3 Gyr in early-type galaxies. These timescales are a factor of $\simeq$ 2 longer if a smoothly declining quenching process is assumed.\\ 3) The fraction of quenched late-type galaxies ($QF$ $\geq$ 80\%), as the fraction of HI-deficient objects, decreases by a factor of $\simeq$ 5 from the core of the Virgo cluster ($R$/$R_{vir}$ $\leq$ 0.5) to the cluster periphery ($R$/$R_{vir}$ $\geq$ 4). The disagreement with the results obtained from the analysis of large samples such as the SDSS (e.g. Wetzel et al. 2012, 2013), which indicate longer timescales with a delayed-then-rapid quenching process, is only apparent since our timescales are representative of the rapid decrease of the star formation activity occured when galaxies enter the rich Virgo cluster, while those give in the literature of the time since a galaxy became a satellite of a more massive halo. Furthermore, our are the timescales necessary to reduce the star formation activity by a factor of $\gtrsim$ 50-80\%, while those given in the literature to fully stop the activity of the perturbed galaxies. All these results are consistent with a rapid quenching of the star formation activity of the late-type galaxies recently accreted on the Virgo cluster as predicted by ram pressure stripping models. These results discard inefficient mechanisms such as starvation, which require very long ($\gtrsim$ 6 Gyr) timescales to significantly quench the star formation activity of the perturbed galaxies. | 16 | 9 | 1609.00545 |
1609 | 1609.07824_arXiv.txt | We present wide field $JHK_{S}$ photometry of 16 Galactic globular clusters located towards the Galactic bulge, calibrated on the 2MASS photometric system. Differential reddening corrections and statistical field star decontamination are employed for all of these clusters before fitting fiducial sequences to the cluster red giant branches (RGBs). Observed values and uncertainties are reported for several photometric features, including the magnitude of the RGB bump, tip, the horizontal branch (HB) and the slope of the upper RGB. The latest spectroscopically determined chemical abundances are used to build distance- and reddening-independent relations between observed photometric features and cluster metallicity, optimizing the sample size and metallicity baseline of these relations by supplementing our sample with results from the literature. We find that the magnitude difference between the HB and the RGB bump can be used to predict metallicities, in terms of both iron abundance $\mathit{[Fe/H]}$ and global metallicity $\mathit{[M/H]}$, with a precision of better than 0.1 dex in all three near-IR bandpasses for relatively metal-rich ($[M/H]$$\gtrsim$-1) clusters. Meanwhile, both the slope of the upper RGB and the magnitude difference between the RGB tip and bump are useful metallicity indicators over the entire sampled metallicity range (-2$\lesssim$$[M/H]$$\lesssim$0) with a precision of 0.2 dex or better, despite model predictions that the RGB slope may become unreliable at high (near-solar) metallicities. Our results agree with previous calibrations in light of the relevant uncertainties, and we discuss implications for clusters with controversial metallicities as well as directions for further investigation. | Galactic globular clusters (GGCs) play a crucial role in constraining stellar evolutionary models as well as Galactic chemical evolution. Recently, many of these clusters have been the subject of large-scale photometric surveys using deep, high resolution multi-colour space-based observations \citep{piotto,sarajedini,piotto14}. However, GGCs located towards the Galactic bulge, despite their importance as the most metal-rich (and in some cases, massive) members of the GGC system, have been generally excluded from these surveys due to severe total and differential extinction at optical wavelengths. For this reason, infrared wavelengths, where the effects of extinction are greatly reduced ($A_{K}$$\sim$$0.12A_{V}$; \citealt{casagrande}), are ideal for photometric investigations of such clusters. The \textit{Vista Variables in the Via Lactea} (VVV), an ESO public survey, has observed a 562 sq.~degree field including the Galactic bulge and a portion of the disk in $YZJHK_{S}$ filters down to $K_{S}$$\sim$20, and thus presents an ideal opportunity to study the GGCs located in the survey area. Since the advent of near-IR arrays, a wealth of effort has been devoted to studying GGCs in the near-infrared largely by Valenti, Ferraro and collaborators (e.g.~\citealt{ferraro00,v04obs,v04abs}, Valenti et al.~2010, hereafter \citealt{v10}; also see \citealt{chun} and references therein), in addition to the earlier studies of \citet{chobump} and \citet{jura2mass} which employed photometry from the Two Micron All Sky Survey (2MASS; \citealt{skrutskie}). An important goal of these investigations was the construction of relations between observable features in cluster near-IR colour-magnitude diagrams (CMDs) and their chemical abundances, as these relations can then be applied to obtain photometric metallicity estimates. With an eye towards future application for distant and/or heavily extincted stellar systems, we revisit these calibrations. This is advantageous in light of not only the quality of the VVV photometry, but more importantly its wide-field nature, facilitating a statistical assessment of contamination by field stars (see Sect.~\ref{decontbigsect}), leveraged together with improved spectroscopic abundances (see Sect.~\ref{badfehsect}) and reddening maps (e.g.~\citealt{javier12,gonzalez,cohen6544}; see Sect.~\ref{diffredsect}). Here we analyse an initial subset of GGCs within the VVV survey area which have spectroscopically measured $\mathit{[Fe/H]}$ values, with the goal of constructing updated distance- and reddening-independent relations between photometric features observable on the cluster giant and horizontal branches and their metallicities. The resulting relations between distance- and reddening-independent photometric features measured from near-IR cluster CMDs versus cluster metallicities are further optimized by concatenating the results presented here with those available in the literature. In the next section, we present the details of our observations and data processing, including corrections for differential reddening and field star contamination, and the resulting cluster CMDs. In Sect.~3, we describe our methodology for measuring cluster photometric features as well as their uncertainties, and in Sect.~4 we use these measurements, along with literature values, to construct relations which can be used to estimate metallicities of old stellar populations photometrically. In the final section we summarize our results, discussing implications for clusters with controversial metallicity values. | \label{discusssect} Our results shown in Figs.~\ref{slopefig}, \ref{bumphbfig} and \ref{bumptipfig} suggest that the three photometric metallicity indicators $slope_{JK}$, $\Delta$$m^{HB}_{RGBB}$ and $\Delta$$m^{RGBB}_{TRGB}$ each have their respective advantages in different metallicity regimes. At relatively high ($[M/H]$$\gtrsim$-1) metallicities, $\Delta$$m^{HB}_{RGBB}$ yields the best overall precision with an rms deviation of $<$0.1 dex from our linear fit in all three $JHK_{S}$ filters, although both $slope_{JK}$ and $\Delta$$m^{RGBB}_{TRGB}$ do nearly as well, with rms deviations of $\sim$0.15 dex. However, moving to lower metallicities, $\Delta$$m^{HB}_{RGBB}$ becomes difficult to apply for two reasons. First, clusters which are more metal-rich tend to have HB magnitudes which can be more reliably measured in the near-IR due to the increased horizontality of the HB in near-IR CMDs. For example, a peak in the cluster LF corresponding to the HB could not be reliably detected for clusters with $\mathit{[Fe/H]}$\citep{c09}$<$-1.2, similar to the results of \citet{cohenispi}. Second, the RGB bump becomes less prominent with decreasing metallicity \citep{natafbump}, also hindering the use of $\Delta$$m^{RGBB}_{TRGB}$. Therefore, despite its somewhat poorer rms deviation of $\sim$0.2 dex, $slope_{JK}$ may be the best option at lower metallicities, particularly for relatively sparse stellar populations where either the RGBB and/or the TRGB location cannot be reliably measured. We can apply the calibrations listed in Table \ref{relationtab} to obtain purely photometric metallicity estimates for our target clusters. For each cluster, there are a total of seven calibrations available to calculate $\mathit{[Fe/H]}$ or $\mathit{[M/H]}$, and we exclude those for which data are not available in individual cases (i.e. the HB magnitudes for metal-poor clusters). The resulting mean photometric $\mathit{[Fe/H]}$ and $\mathit{[M/H]}$, weighted by the inverse quadrature sum of the observational uncertainty and the calibration rms, are given in Table \ref{photcaltab} along with the number of relations from Table \ref{relationtab} available. These photometric metallicity estimates are compared in Fig.~\ref{photcalfig} with $\mathit{[Fe/H]}$ values from \citet{brunolores},\citet{mauro14}, the HiRes spectroscopic values in Table \ref{fehtab}, and the \citet{h96} catalog. \begin{table} \centering \caption{Photometric Metallicity Estimates for Target Clusters} \begin{tabular}{lccc} \hline Cluster & $\mathit{[Fe/H]}$ & $\mathit{[M/H]}$ & N(relations) \\ \hline NGC6380 & -0.81$\pm$0.05 & -0.55$\pm$0.05 & 7 \\ NGC6401 & -1.34$\pm$0.21 & -1.03$\pm$0.21 & 1 \\ NGC6440 & -0.46$\pm$0.06 & -0.23$\pm$0.06 & 7 \\ NGC6441 & -0.44$\pm$0.05 & -0.21$\pm$0.05 & 7 \\ NGC6453 & -1.80$\pm$0.12 & -1.49$\pm$0.12 & 4 \\ NGC6522 & -1.38$\pm$0.11 & -1.10$\pm$0.11 & 4 \\ NGC6528 & -0.08$\pm$0.05 & 0.10$\pm$0.05 & 7 \\ NGC6544 & -1.69$\pm$0.12 & -1.39$\pm$0.12 & 4 \\ NGC6553 & -0.23$\pm$0.05 & -0.03$\pm$0.05 & 7 \\ NGC6558 & -1.18$\pm$0.22 & -0.87$\pm$0.22 & 1 \\ NGC6569 & -1.00$\pm$0.05 & -0.71$\pm$0.05 & 7 \\ NGC6624 & -0.58$\pm$0.05 & -0.34$\pm$0.05 & 7 \\ M28 & -1.17$\pm$0.11 & -0.90$\pm$0.11 & 4 \\ M69 & -0.60$\pm$0.05 & -0.37$\pm$0.05 & 7 \\ NGC6638 & -1.09$\pm$0.07 & -0.79$\pm$0.07 & 7 \\ NGC6642 & -1.55$\pm$0.05 & -1.20$\pm$0.05 & 7 \\ M22 & -1.71$\pm$0.11 & -1.41$\pm$0.11 & 4 \\ \hline \end{tabular} \label{photcaltab} \end{table} In general, our linear fits in Figs.~\ref{slopefig}-\ref{bumptipfig} favor recent spectroscopic metallicities over those listed in the compilation of \citet{c09} for metal-rich clusters (NGC 6380, 6440, 6528, 6569 and to a lesser extent NGC 6624). For the six target clusters with $\mathit{[Fe/H]}$$\lesssim$-0.7, the agreement with the \citet{h96} catalog is particularly good, with a mean offset of -0.03$\pm$0.02 dex. For NGC 6528, arguably the most metal-rich GGC, our calibrations give $\mathit{[Fe/H]}$ between the the lower values reported by \citet{zoccali04}, \citet{origlia05} and \citet{sobeck06} and the super-solar value of \citet{carretta01}, in good agreement with the low-resolution spectra of \citet{dias14}. In a global sense, our photometric metallicity calibrations agree best with the CaII triplet values of \citet{mauro14} compared to the other sets of literature metallicity values. For example, our photometric metallicities imply a decrease in $\mathit{[Fe/H]}$ of $\sim$0.3-0.4 dex for NGC 6380 and NGC 6569 compared to their \citet{c09} values. For NGC 6380, this is also in agreement with the \citet{h96} catalog, while NGC 6569 is a significant outlier in the CaII triplet calibration of \citet{mauro14}, who found $\mathit{[Fe/H]}$=-1.18$\pm$0.11, in better ($\sim$2$\sigma$) accord with our photometric metallicities than any spectroscopic results. For NGC 6401 and NGC 6558, our photometric values are $\sim$0.2 dex lower than those measured from spectroscopy. However, because these clusters lack a detectable RGBB or red HB, our photometric metallicity estimate is based only on the RGB slope, so the uncertainties are relatively large. Furthermore, since these clusters are relatively sparse and projected on the Galactic bulge, their TRGB magnitudes remain uncertain at the $\gtrsim$0.3 mag level (see Appendix \ref{tipdetailsect}). This implies yet a larger corresponding uncertainty of the RGB slope and hence the photometric metallicity should our chosen TRGB candidate be proven incorrect. For the remainder of metal-intermediate blue HB clusters in the VVV sample, we also find metallicities $>$0.2-0.3 dex lower than those of \citet{c09}. For NGC 6544, the \citet{c09} value is supported by \citet{mauro14}, although the discrepancy between their result and ours is only marginally significant in light of the large uncertainties. Meanwhile, our photometric $\mathit{[Fe/H]}$ value for NGC 6453 rests fairly heavily on the assumed TRGB magnitude, and in fact our photometric $\mathit{[M/H]}$ value is in good agreement with \citet{brunolores} if this cluster has a relatively low level of $\alpha$-enhancement as suggested by their fits to synthetic spectra. Lastly, for NGC 6642, the value given by our photometric calibrations agrees with \citet{minniti95} to within the uncertainties, and a value as low as $\mathit{[Fe/H]}$=-1.8 was suggested by \citet{balbinot} based on isochrone fitting to space-based optical photometry. \begin{figure*} \includegraphics[width=0.99\textwidth]{f16.eps}% \caption{Comparison between our photometric $\mathit{[Fe/H]}$ estimates given in Table \ref{photcaltab} and $\mathit{[Fe/H]}$ values from (left to right) \citet{brunolores}, \citet{mauro14}, the HiRes values in Table \ref{fehtab} and the \citet{h96} catalog. In each plot, the dashed line indicates equality. Because \citet{h96} give weights rather than formal uncertainties on their $\mathit{[Fe/H]}$ values, the size of the plotting symbol is proportional to the weight given to the $\mathit{[Fe/H]}$ value for each cluster in the rightmost panel by \citet{h96}.} \label{photcalfig} \end{figure*} Our results underscore the need for high-resolution multi-object spectroscopy of poorly studied bulge GGCs. The tendency of the \citet{c09} compilation to overestimate the $\mathit{[Fe/H]}$ of metal-rich bulge clusters could be simply an artefact of high field star densities in the original integrated light studies compiled by \citet{c09} and/or the use of a super-solar metallicity for NGC 6528 to convert previous metallicity scales to their UVES scale. In either case, the GGC metallicity scale at high metallicities remains poorly constrained, and detailed spectroscopic analyses of large samples of cluster stars (for example, to assess contamination by AGB members) are crucial for accurate and self-consistent determinations of $\mathit{[Fe/H]}$ as well as $[\alpha/Fe]$. This would be a valuable step towards testing GGC evolutionary models at near-solar metallicities and improving our relations to allow a deeper understanding of distant, composite and/or heavily extincted stellar populations. | 16 | 9 | 1609.07824 |
1609 | 1609.00403_arXiv.txt | We have measured the energies of the strongest 1s--2$\ell\ (\ell=\text{s,p})$ transitions in He- through Ne-like silicon and sulfur ions to an accuracy of $<1\,\mathrm{eV}$ using Lawrence Livermore National Laboratory's electron beam ion traps, EBIT-I and SuperEBIT, and the NASA/GSFC EBIT Calorimeter Spectrometer (ECS). We identify and measure the energies of 18 and 21 X-ray features from silicon and sulfur, respectively. The results are compared to new Flexible Atomic Code calculations and to semi-relativistic Hartree Fock calculations by Palmeri et al.\ (2008). These results will be especially useful for wind diagnostics in high mass X-ray binaries, such as Vela X-1 and Cygnus X-1, where high-resolution spectral measurements using \textsl{Chandra}'s high energy transmission grating has made it possible to measure Doppler shifts of $100\,\mathrm{km\,s}^{-1}$. The accuracy of our measurements is consistent with that needed to analyze \textsl{Chandra} observations, exceeding \textsl{Chandra}'s $100\,\mathrm{km\,s}^{-1}$ limit. Hence, the results presented here not only provide benchmarks for theory, but also accurate rest energies that can be used to determine the bulk motion of material in astrophysical sources. We show the usefulness of our results by applying them to redetermine Doppler shifts from \textsl{Chandra} observations of Vela X-1. | Prominent absorption and emission X-ray features from highly charged silicon and sulfur ions have been detected and measured in a medley of celestial sources, including solar flares \citep{neupert1971a}, other stellar coronae \citep[e.g.,][]{kastner2002a,huenemoerder2013a}, various types of Active Galactic Nuclei \citep[e.g.,][]{leejc2001a, kaspi2002a,kinkhabwala:2002fx,holczer2007a,holczer2012a,reeves2013a}, and high-mass X-ray binaries \citep[HMXB; e.g.,][]{sako02a,boroson2003a,watanabe06a, chang07a, hanke08a,miskovicova:16a}. HMXBs, although well studied and cataloged, are not yet fully understood. In general, they consist of a massive O- or B-type star in orbit with a compact object, either a black hole or neutron star. X-ray emission or absorption features from these sources are generated when the luminous ($10^{36}\ldots10^{38}\,\mathrm{erg}\,\mathrm{s}^{-1}$) X-ray continuum from the accreting compact object irradiates, ionizes, and fluoresces the stellar wind material ejected from the companion star. Because the stellar wind of the massive companion is radiation driven, the ionizing nature of the X-ray continuum affects not only the wind structure, but also the mass loss rate of the companion star. Hence, K$\alpha$ transitions originating in the wind have not only been used to determine the ion structure and motion of the wind, but also provide insight into the mass loss rate of the companion star and the strength of the X-ray continuum. For example, in the case of \object{Vela X-1}, \citet{sako02a}, \citet{schulz02a}, \citet{goldstein04a}, and \citet{watanabe06a} report high resolution X-ray emission spectra from 2p$\rightarrow$1s, i.e., K$\alpha$, transitions from both L- and K-shell silicon and sulfur ions. \citet{sako02a} identify resolved line emission from O-like \ion{Si}{7} through H-like \ion{Si}{14}, and an unresolved feature identified as \ion{Si}{2}--\ion{Si}{6}. \citet{goldstein04a} find the motion of different ions of the same element to be non-uniform, based on the limited quality of their used reference wavelengths. \citet{watanabe06a} build a three dimensional Monte-Carlo radiative transfer model and report a mass loss rate for the companion star and the structure of the wind, although they do not analyze the line emission from the L-shell silicon ions, but only from H-like Si. In the case of \object{Cygnus X-1}, the K$\alpha$ absorption features in L-and K-shell ions of silicon and sulfur have been measured and used to diagnose the nature of the stellar wind \citep{hanke08a,hell2013a,miskovicova:16a}. Specifically, these features have been shown to be produced by ``clumps'' of onion-structured material, where the inner layers are colder, denser, and less ionized, moving in and out of the observational line of sight. In the case of multielectron L-shell ions of silicon and sulfur, the utility of the associated X-ray line diagnostics is limited and often precluded by the relatively poor accuracy of the atomic reference data. Accurate calculations of the atomic structure of these ions is challenging because correlation effects among multiple electrons must be taken into account. Historically, Hartree-Fock calculations of \citet{house69a} were used to interpret high resolution solar spectra \citep{fritz1967a}, and more recently have been used to analyze data from both Vela X-1 \citep{schulz02a,goldstein04a} and Cygnus X-1 \citep{hankediss}. However, \citet{house69a} only provide simplified data listing only a single transition for each ion. To provide a more complete and accurate data set, more sophisticated calculations have been completed using more advanced atomic models. For example, \citet{behar02a} used the Hebrew University Lawrence Livermore Atomic Code \citep[HULLAC;][and references therein]{klapisch1971a,klapisch2006a} to calculate transition energies and line strengths for the strongest K-shell transitions in He- through F-like silicon and sulfur ions. At present, the most complete calculation is provided by \citet[][P08]{palmeri08a}, who use a semi-relativistic Hartree-Fock code to calculate level energies, transition wavelengths, and radiative decay rates for $\sim$1400 K-shell transitions in silicon and sulfur ions. The variation among the inner-shell transition energies calculated with various codes is $\sim$2--5\,eV, i.e., on the order of several 100\,$\mathrm{km}\,\mathrm{s}^{-1}$ for the diagnostically important L-shell silicon K$\alpha$ lines. This variation is comparable to the expected Doppler shift of the L-shell silicon K$\alpha$ lines \citep{watanabe06a,liedahl08a,miller05a,miller12a,miskovicova:16a}, and significantly larger than the systematic wavelength error of \textsl{Chandra}'s High Energy Transition Grating Spectrometer (HETGS), which is on the order of $100\,\mathrm{km}\,\mathrm{s}^{-1}$ \citep{marshall2004a,canizares05a,chandrapog2015}. Hence, the main systematic uncertainty in the determination of Doppler shifts from X-ray lines is our knowledge of atomic physics. This has been pointed out before in studies of the K-shell lines in L-shell oxygen ions \citep{schmidt2004a,gu2005a}. When comparing atomic databases commonly used to interpret both Solar and extra-Solar X-ray spectra, the data from P08 are found in the Universal Atomic DataBase (uaDB) accompanying XSTAR \citep{bautista01a}; however, they are not included in either the atomic physics for astrophysics database, AtomDB v2 \citep{foster12a} or the CHIANTI atomic physics database \citep{dere97a,landi2013a}. AtomDB v2 only includes K-shell transitions in helium-like and hydrogen-like ions; CHIANTI only includes H-like, He-like, and Li-like transitions. There is one previous measurement available for L-shell transitions in Be- through F-like Si and S ions. \citet{faenov94a} measured transitions produced in a $\mathrm{CO}_2$ laser-produced plasma. They also provided a comparison to their own theoretical calculations. The density of this plasma is significantly higher than typical densities in an astrophysical environment. The spectra reported by \citet{faenov94a} therefore comprise mainly dielectronic satellites (see their Tables~I and~II) and are only of limited applicability for our purpose. \enlargethispage{\baselineskip} Here, we report results of measurements of the 1s--$2\ell\ (\ell=\mathrm{s, p})$ K-shell energies in He- to Ne-like ions of silicon and sulfur in a coronal plasma produced with the Lawrence Livermore National Laboratory electron beam ion traps (Section~\ref{sec:measurement}). To gauge the systematic uncertainty inherent to calculations of many-body atomic systems, we compare these measurement results (Section~\ref{sec:fitmethod}) to line energy calculations performed with two popular atomic codes used for line identification, namely our own calculations with the Flexible Atomic Code \citep[FAC;][]{gu04b,gu08a} (Section~\ref{sec:lineIDFAC}) and the tables of P08 (Section~\ref{sec:facvspalm}). In addition, we list the centers of major line blends as a reference for observations with moderate resolution and derive new Doppler shifts for Vela~X-1 based on our laboratory measured values (Section~\ref{sec:blends}). We summarize our results in Section~\ref{sec:conclusion}. | \label{sec:conclusion} The K$\alpha$ emission line energies from Si$^{4+}$ through Si$^{12+}$ and S$^{6+}$ through S$^{14+}$ have been measured using the ECS calorimeter at the LLNL EBIT facility. The results have been compared to our own FAC calculations and earlier calculations of \citet{palmeri08a}, \citet{behar02a}, and \citet{house69a}. The newly available data (Table~\ref{tab:center}) can directly be applied to resolve astrophysical problems such as, e.g., wind diagnostics in high mass X-ray binary systems like Vela X-1 \citep{liedahl08a} and Cyg X-1 \citep{miskovicova:16a}. The 90\% confidence limits of $\lesssim0.5$\,eV on the measured line centers presented here correspond to Doppler shifts of less than 90\,$\mathrm{km\,s}^{-1}$. These measurements, therefore, provide line centers with an accuracy slightly better than the uncertainty of $\sim100\,\mathrm{km\,s}^{-1}$ on the \textsl{Chandra} HETG \citep{marshall2004a,canizares05a,chandrapog2015}. When future missions with higher effective area make high-resolution spectra of point as well as extended celestial sources more commonly available, we expect to see these lines to be resolved in a variety of sources. Our results will then be especially useful for extended sources like supernova remnants which have yet to be observed in high resolution. | 16 | 9 | 1609.00403 |
1609 | 1609.05226_arXiv.txt | We investigate test-particle diffusion in dynamical turbulence based on a numerical approach presented before. For the turbulence we employ the nonlinear anisotropic dynamical turbulence model which takes into account wave propagation effects as well as damping effects. We compute numerically diffusion coefficients of energetic particles along and across the mean magnetic field. We focus on turbulence and particle parameters which should be relevant for the solar system and compare our findings with different interplanetary observations. We vary different parameters such as the dissipation range spectral index, the ratio of the turbulence bendover scales, and the magnetic field strength in order to explore the relevance of the different parameters. We show that the bendover scales as well as the magnetic field ratio have a strong influence on diffusion coefficients whereas the influence of the dissipation range spectral index is weak. The best agreement with solar wind observations can be found for equal bendover scales and a magnetic field ratio of $\delta B / B_0 = 0.75$. | It is well-known that magnetic turbulence influences the motion of electrically charged energetic particles such as cosmic rays. Turbulence in general has different properties such as the spectrum describing how the magnetic energy is distributed among different length scales. Another fundamental aspect of turbulence is spectral anisotropy describing how magnetic turbulence varies in different directions of space. Diffusion of particles along the mean magnetic field, for instance, is controlled by gyro-resonant interactions (see, e.g., Schlickeiser 2002 and Shalchi 2009 for reviews). Therefore, the spectrum of turbulence at a certain scale or wavenumber determines the diffusion coefficient of the energetic particles with a certain energy. It should be emphasized, however, that nonlinear effects can be important for parallel diffusion and non-resonant interactions can influence the diffusion parameter in certain parameter regimes (see Shalchi 2009 for a review). Spectral anisotropy can also have an effect but this effect is weaker than originally thought (see Hussein et al. 2015). For perpendicular diffusion, however, the details of the turbulence seem to be less important because the perpendicular diffusion coefficient depends only on the so-called Kubo number and the parallel diffusion coefficient (see Shalchi 2015). Due to the latter dependence, however, the perpendicular diffusion parameter indirectly also depends on spectrum and spectral anisotropy. Another important turbulence property is the dynamics describing the characteristic time scales over which the turbulent magnetic field decorrelates. Different approaches have been proposed in the past to model the turbulence dynamics. Some attempts are based on plasma wave propagation models in which the propagation effect itself is taken into account as well as various damping effects (see again Schlickeiser 2002 for a review). Or there is the important work of Bieber et al. (1994) in which simple models have been proposed to approximate the temporal decorrelation of turbulence, namely the so-called {\it damping model of dynamical turbulence} and the {\it random sweeping model}. In the recent years scientists achieved a more complete understanding of the turbulence time scales. Therefore a more advanced model for the turbulence dynamics has been proposed in Shalchi et al. (2006). This model is called the {\it Nonlinear Anisotropic Dynamical Turbulence (NADT) model} and takes into account wave propagation effects as well as damping effects. It is the aim of this article to simulate energetic particle motion in this type of turbulence and to explore the influence of different turbulence parameters. It was shown in different papers that dynamical turbulence effects can have a strong influence on the transport of energetic particles. This concerns parallel diffusion (see, e.g., Bieber et al. 1994) but also perpendicular diffusion (see, e.g., Shalchi et al. 2006). Such previous investigations were based on quasilinear and nonlinear calculations. These days, however, one can also obtain diffusion parameters from test-particle simulations. Previous work of this type was mostly done for magnetostatic turbulence (see, e.g., Giacalone \& Jokipii 1999, Qin et al. 2002a, and Qin et al. 2002b) or undamped propagating plasma waves (see, e.g., Micha\l ek \& Ostrowski 1996 and Tautz \& Shalchi 2013). In Hussein \& Shalchi (2016) we have started to simulate test-particle transport in the dynamical turbulence models used in Bieber et al. (1994), namely in the {\it damping model of dynamical turbulence} and the {\it random sweeping model}. It was shown in Hussein \& Shalchi (2016) that for certain turbulence parameters we can indeed reproduce different solar wind observations. It is the purpose of the current paper to simulate particle transport in the more realistic NADT model and to compute the parallel mean free path $\lambda_{\parallel}$, the perpendicular mean free path $\lambda_{\perp}$, and the ratio of the two mean free paths $\lambda_{\perp} / \lambda_{\parallel}$. As in Hussein \& Shalchi (2016) our findings are compared with the Palmer (1982) consensus range, observations of Jovian electrons (see Chenette et al. 1977), and Ulysses measurements of Galactic protons (see Burger et al. 2000). We also explore how the different turbulence parameters influence the different diffusion parameters. The reminder of the paper is organized as follows. In Section 2 we explain the physics of turbulence in general but we focus on the NADT model used in the current paper. The methodology which is used to perform particle transport simulations in dynamical turbulence is explained in Section 3. In Section 4 we show our numerical results obtained for parallel and perpendicular diffusion coefficients and we compare them with different solar wind observations. In Section 5 we conclude and summarize. | The current paper is a sequel of Hussein \& Shalchi (2016) where we have started to perform test-particle simulations for dynamical turbulence. The turbulence dynamics can have a strong influence on particle diffusion coefficients at low particle rigidities. In the previous work, we have employed two models for dynamical turbulence, namely the {\it damping model of dynamical turbulence} and the {\it random sweeping model}. Both models were originally proposed in the pioneering work of Bieber et al. (1994). It is the purpose of the current paper to replace the aforementioned dynamical turbulence models by the so-called {\it Nonlinear Anisotropic Dynamical Turbulence (NADT) model} of Shalchi et al. (2006) which takes into account wave propagation effects as well as damping effects. Furthermore, we perform a detailed parameter study in order to explore the influence of the magnetic field ratio $\delta B / B_0$, the turbulence scale ratio $l_{2D} / l_{slab}$, the dissipation range spectral index $p$, the dissipation wavenumber $k_d$, as well as the inertial range spectral index $s$ on the parallel mean free path $\lambda_{\parallel}$, the perpendicular mean free path $\lambda_{\perp}$, and the ratio of the two diffusion parameters $\lambda_{\perp}/\lambda_{\parallel}$. Our findings are shown in Figs. \ref{paracompdb1}-\ref{perp_prll_index} and the corresponding parameter values are listed in Tables \ref{simvalues} and \ref{runs}. We found that the influence of the dissipation range spectral index is minor. The influence of the inertial range spectral index and the dissipation scales are negligible as well. The magnetic field ratio, on the other hand, has a strong influence on both diffusion coefficients and their ratio. We found best agreement with the Palmer (1982) consensus range for $\delta B / B_0 = 0.75$ (corresponding to approximately $\delta B^2 / B_0^2 = 0.6$) which is between the values $\delta B / B_0 = 0.5$ and $\delta B / B_0 = 1$ usually used for this type of work. We also found that the ratio of the bendover scales $l_{2D} / l_{slab}$ has an influence on the parallel mean free path and a very strong influence on the perpendicular diffusion coefficient. This was predictable because analytical treatments of the transport (see, e.g., Shalchi 2015) show the importance of the so-called {\it Kubo number} on the perpendicular motion of energetic particles. The latter number depends on the magnetic field ratio as well as the turbulence scales. The main conclusion of the current paper is that we can indeed reproduce different solar wind observations performed for energetic particles interacting with magnetic turbulence if we employ the NADT model. More detailed turbulence measurements would show what the exact value of the different parameters used in the current paper are. Then one could draw more conclusions concerning the validity of the employed turbulence model. | 16 | 9 | 1609.05226 |
1609 | 1609.06225_arXiv.txt | We report the discovery of a $\sim$ $3\degr\llap{.}4$-wide region of high-energy emission in data from the \emph{Fermi} LAT satellite. The centroid of the emission is located in the Southern Hemisphere sky, a few degrees away from the plane of the Galaxy at the Galactic coordinates l=350$\degr\llap{.}$6, b=-4$\degr\llap{.}$7. It shows a hard spectrum that is compatible with a simple power-law, $\frac{dN}{dE}\propto E^{-\Gamma}$, in the energy range 0.7--500 GeV, with a spectral index $\Gamma = 1.68 \pm 0.04_{\mbox{\tiny stat}} \pm 0.1_{\mbox{\tiny sys}}$. The integrated source photon flux above 0.7 GeV is $(4.71 \pm 0.49_{\mbox{\tiny stat}}\pm 2.13_{\mbox{\tiny sys}}) \times 10^{-9}$ cm$^{-2}$ s$^{-1}$. We discuss several hypotheses for the nature of the source, particularly that the emission comes from the shell of an unknown supernova remnant. | \label{sec:intro} The current generation of gamma-ray observatories have revealed a large variety of Galactic and extragalactic sources of high-energy (MeV to GeV) and very high-energy (TeV) emission. In particular the all-sky survey of the \emph{Fermi} Large Area Telescope (LAT) \citep{2009ApJ...697.1071A} has had considerable impact on the science of high-energy astrophysics. Gamma-ray emitting objects observed by the LAT include supernova remnants (SNRs), and their surroundings, pulsar wind nebulae (PWN), pulsars, binary systems, novae, and supermassive black holes \citep[e.g.,][]{2010ApJ...710L..92A,2013ApJ...773...77A,2010ApJS..187..460A,2009ApJ...701L.123A,2014Sci...345..554A,2015ApJ...810...14A}. Uncovering the origin of gamma-ray emission in all these systems is important to understand the physical processes behind the acceleration of particles. Its study may also help reveal the origin of Galactic cosmic rays with energies up to $\sim$PeV. Several SNRs have been established as sources of high-energy cosmic rays \citep[e.g.,][]{2013Sci...339..807A} but the search for the Galactic PeVatron continues and many of the observed sources remain unidentified. SNRs also show a variety of gamma-ray spectra and features that are not fully understood, which points to the need for more studies and theoretical work on the problem of particle acceleration and transport \citep[e.g.,][]{2009ApJ...706L...1A,2011ApJ...735..120Y,2012JCAP...07..038C,2014BrJPh..44..415B,2016ApJS..224....8A}. Many known Galactic extended sources at GeV and TeV energies are likely PWN or SNRs \citep{2015ApJS..218...23A,2013arXiv1307.4690C}. Well studied examples of SNR shells that emit gamma-rays are RX J1713.7-3946 \citep{2011ApJ...734...28A} and RX J0852.0-4622 \citep{2011ApJ...740L..51T}. These sources show a hard GeV spectrum which can be explained by inverse Compton (IC) scattering of soft photons by high-energy leptons, although the origin of the emission is still debated. Most gamma-ray sources are found close to the Galactic Equator, but this is not always the case. A notable exception is the unidentified source HESS J1507-622 \citep{2011A&A...525A..45H} that lies about 3$\degr\llap{.}$5 from the plane of the Galaxy and has no clear X-ray counterpart, which is surprising given the low absorption expected at its location. This also prevents a clear determination of its distance, a key parameter to understand its nature. Some authors have suggested that this and other unidentified gamma-ray sources could be ancient PWN with no counterpart at other wavelengths \citep{2013ApJ...773..139V}. Here, we report the discovery of an extended source of gamma-rays seen in LAT data, which shows some of the features of the sources discussed above. The source shows a hard photon spectrum. When modeled with a uniform disc template, the center of the emission is located at the Galactic coordinates $l=350\degr\llap{.}6$, $b=-4\degr\llap{.}7$ and its best-fit radius is $\sim 1\degr\llap{.}7$. Previously, several gamma-ray sources were found in the region as well as hints of the presence of extended high-energy emission. The Third EGRET Catalog of High-Energy Gamma-Ray Sources \citep{1999ApJS..123...79H} shows a relatively low-significance unidentified source, 3EG J1744-3934 which can be consistent with an extended source and also with a collection of multiple sources. It is located at the Galactic coordinates $l=350\degr\llap{.}81$, $b=-5\degr\llap{.}38$. A power-law spectral index of $\sim 2.4$ is reported for this EGRET source. In the \emph{Fermi} LAT 4-year Point Source Catalog \citep[3FGL,][]{2015ApJS..218...23A}, the unidentified sources 3FGL J1733.5-3941 and 3FGL J1748.5-3912 lie close to the boundary of the new source reported here and show soft spectra. The source 3FGL J1747.6-4037 is also near the boundary of the new extended source. It is identified as the gamma-ray counterpart of the millisecond pulsar PSR J1747-4036 with a dispersion measure that implies a source distance of 3.4 kpc. The pulsar's spindown luminosity is $1.16\times 10^{35}$ erg s$^{-1}$ \citep[uncorrected for the Shklovskii effect;][]{2012ApJ...748L...2K,2013ApJS..208...17A}. It is a faint gamma-ray source and the LAT's spectrum shows no evidence of spectral curvature. A $\sim31\arcmin$-wide SNR, G351.0-5.4, was recently discovered in radio observations of the region \citep{2014A&A...568A.107D}. The SNR has a low surface radio brightness, its centre is located at the Galactic coordinates $l=351\degr\llap{.}06$, $b=-5\degr\llap{.}49$ and has no optical counterpart, which led the authors to classify it as an old SNR. No PWN is known within the extent of the region. \cite{2014A&A...568A.107D} analyzed Pass 7 LAT data around the SNR and found a slightly spatially extended feature above the background with a 1.5$\sigma$ confidence, although no hints of emission within the radio contours of the SNR were found. Finally, the unidentified point-like sources 2FHL J1741.2-4021, reported in the Second \emph{Fermi} LAT Catalog of High-Energy Sources (2FHL) detected above 50 GeV \citep{0067-0049-222-1-5}, is found within the region, and 3FHL J1733.4-3942 reported in the Third Catalog of Hard \emph{Fermi} LAT Sources \citep[3FHL,][]{0067-0049-232-2-18}, was seen at the edge of the new gamma-ray source reported here. In this paper LAT data selection, analysis and results are described in Section \ref{sec:data}. In Section \ref{sec:discussion} possible interpretations of the nature of this source are given. | \label{sec:discussion} It is reasonable to attribute the origin of the gamma-rays to the shell of a previously unknown SNR. If located at the distance to the pulsar PSR J1747-4036 (3.4 kpc) the object size would be around 200 pc and the center of the disc would be located around 250 pc from the plane of the Galaxy. SNRs with diameters of the order of a few degrees are known and a significant number have been seen as extended objects by the LAT \citep{2016ApJS..224....8A}. Well-known examples are the Cygnus Loop \citep{2011ApJ...741...44K} and RX J0852.0-4622 \citep{2011ApJ...740L..51T}. However, not many SNRs are known to have a diameter greater than 100 pc, although large objects could not have detected due to observational selection effects \citep{1991PASP..103..209G}. Dynamic considerations of the expansion and evolution of a SNR can give a physical justification for an upper limit in their size \citep[e.g.,][]{1988ApJ...334..252C,1993ApJ...417..187S}. It is possible then that if this new source is a SNR it is much closer than 3.4 kpc and hence unrelated to PSR J1747-4036. This pulsar is classified as a millisecond pulsar \citep[e.g.,][]{2013ApJS..208...17A}, and thus if related to the G350.6-4.7, the SNR would also be too old to show a GeV spectrum as the one found here \citep[see e.g.,][]{2016ApJS..224....8A}. The Cygnus Loop SNR is located at about 540 pc \citep{2005AJ....129.2268B} and has a diameter of $\sim35$ pc. If the same physical diameter is assumed for G350.6-4.7, the resulting distance would be similar. However, the LAT spectrum of the Cygnus Loop \citep{2011ApJ...741...44K} is quite different to that reported here for G350.6-4.7 so these two objects likely have very different properties. On the other hand, with a diameter of $\sim2\degr$ (or $\sim$26 pc), RX J0852.0-4622 shows a GeV spectrum that is similar in shape to that of G350.6-4.7 \citep{2011ApJ...740L..51T}. Both of these previously known SNRs have very clear counterparts at other wavelengths. A recently discovered source of gamma-ray emission that shows very similar features to those of G350.6-4.7 at high energies is reported at the location of the SNR G150.3+4.5 in the 2FHL catalog \citep{0067-0049-222-1-5}. The gamma ray emission was described by a disc template of radius $1^{\circ}.27$ and it shows a hard spectrum (power law index $1.66\pm 0.20$). SNR G150.3+4.5, which is thought to be associated to this gamma-ray source, was itself recently discovered \citep{2014A&A...567A..59G}. It is possible that the gamma-rays from G350.6-4.7 are produced in the shell of a new SNR. There might also be a contribution to the gamma-ray emission from the recently discovered SNR G351.0-5.4. However, this SNR has a small size compared to the gamma-ray source G350.6-4.7 and it is likely old \citep{2014A&A...568A.107D}. There are candidate SNR shells that are detected at TeV energies and have no known counterparts at other wavelengths, such as HESS J1912+101 \citep{2015arXiv150903872P}, located at the coordinates $(l,b)=44\degr\llap{.}39,-0\degr\llap{.}07$. This source has no known counterpart at GeV energies and one of the reasons could be that the source has a hard GeV spectrum that is below the LAT sensitivity. It is possible that \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ G350.6-4.7 is an example of this kind of TeV shells but for which the LAT telescope is able to detect the hard GeV spectrum. Follow up radio, X-ray and particularly TeV observations of G350.6-4.7 are necessary to understand its nature. It might be natural to assume from the spectral shape of G350.6-4.7 that the gamma-rays are produced by IC scattering of high-energy electrons. Fig. \ref{fig2} shows the SED of the source with a leptonic model calculated with the naima modeling package \citep{2015arXiv150903319Z}. The particle distribution is a power-law with a cutoff in energy ($\epsilon$) of the form $\frac{dN_e}{d\epsilon}\propto \epsilon^{-s}\cdot \mbox{e}^{-\epsilon/\epsilon_c}$. The particle cutoff energy is of course not constrained by the LAT data. Two models resulting for $\epsilon_c=35$ TeV and $\epsilon_c=80$ TeV, with the same particle spectral index $s=2.4$ and total energy content of \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ $1.3\times10^{49}$ erg, are shown with the SED. The particles are assumed to be uniformly distributed in a spherical volume of radius 15 pc located at a distance of 500 pc from Earth. Since the true location of G350.6-4.7 is unknown, for simpicity, the photon field used in the model to serve as seed for upscattering is the Cosmic Microwave Background (CMB). The value for the magnetic field in the model is 1 $\mu$G, which is a typical value in the Galaxy. Based on the source size and the other parameters, the resulting electron energy density at the source would be 19 eV/cm$^{3}$. \begin{figure} \includegraphics[width=\columnwidth]{leptonic} \caption{The spectral energy distribution of G350.6-4.7 is shown here above 0.7 GeV. The squares are the LAT data points with statistical errors and the shaded region represents the 1$\sigma$ fit uncertainty. Two versions of a simple leptonic model with electron cutoff energies of 35 TeV (dotted line) and 80 TeV (dashed line) are also shown. The solid line is the approximate H.E.S.S. sensitivity for a 100-hr observation of an extended source as explained in the text.} \label{fig2} \end{figure} For $\epsilon_c=35$ TeV, the peak of the synchrotron spectral energy distribution is seen at a photon energy of $20$ eV. The IC cooling time of a high-energy lepton with Lorentz factor $\gamma=7\times10^7$ (which corresponds to a particle energy \ \ \ \ \ \ \ \ \ \ \ \ \ $\epsilon \sim 35$ TeV) interacting with the CMB is \citep{1994hea2.book.....L} \begin{equation} \tau = \frac{\epsilon}{\frac{4}{3}\sigma_T c \gamma^2 U_{\mbox{\tiny CMB}}} = \frac{2.3\times10^{12}\,\,\mbox{yr}}{\gamma} \approx 33 \,\,\mbox{kyr.} \label{ictime} \end{equation} If the source age is not higher than this value it is likely that it could be detected at TeV energies for the scenarios shown in Fig. \ref{fig2}. The approximate sensitivity for a 100-hr observation by H.E.S.S., which was taken from the point source sensitivity plot in \cite{2013APh....43..348F} and degraded by a factor of 10 to account for the fact that the source is extended as indicated by the observatory external proposal guidelines\footnote{https://www.mpi-hd.mpg.de/hfm/HESS/pages/home/proposals/}, can be seen in the SED plot also. Another scenario for the origin of the gamma-rays is hadronic emission from cosmic rays accelerated in the shell of a SNR. The hadronic gamma ray flux above a photon energy of 1 GeV from a SNR located at a distance $d$ by a population of cosmic rays with a differential energy spectrum of the form $\epsilon^{-2.1}$ is given by \citep{1994A&A...287..959D} \begin{equation} F \approx 1.8\times 10^{-7}\,\theta\, \left(\frac{E_{\mbox{\tiny SN}}}{10^{51} \,\mbox{erg}}\right)\,\left(\frac{d}{1\,\mbox{kpc}}\right)^{-2} \, \left(\frac{n}{1 \,\mbox{cm}^{-3}}\right)\, \,\,\mbox{cm}^{-2}\,\mbox{s}^{-1}. \label{pp} \end{equation} Here, $\theta$ is the fraction of the total supernova explosion energy ($E_{\mbox{\tiny SN}}=10^{51}$ erg) converted to cosmic ray energy and it is taken as 0.1 as usual \citep[e.g.,][]{2013A&A...553A..34D}, and $n$ is the target gas density. Using these values and the measured flux this translates to $n_1/d_{1}^2 \approx 0.2$, where the distance $d_1$ is in units of 1 kpc and $n_1$ is in units of 1 cm$^{-3}$. For $d_1=0.5$ this implies $n_1=0.05$. Galactic gas distribution models and constraints on the cosmic ray energetics for off-plane gamma ray sources such as HESS J1507-622 \citep{2011AdSpR..47..640D}, observed 3$\degr\llap{.}$5 away from the plane of the Galaxy, predict an ambient density of $n_1\sim 0.55$ for a source distance of 500 pc. Using this density value and distance, the hadronic scenario for \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ G350.6-4.7 requires a relatively low cosmic ray energy fraction of $\theta = 0.01$. On the other hand, a value closer to $\theta \sim 0.1$ would be allowed if the ambient density at the source is much lower than the one predicted by \cite{2011AdSpR..47..640D}, or the source distance is $\sim 1.6$ kpc. But for a higher distance of 2 kpc and $\theta=0.1$ the required density is $n_1\sim 0.8$ which might be high for the corresponding off-plane distance of \ \ \ \ \ \ \ \ \ \ \ \ \ 160 pc \citep{2011AdSpR..47..640D}. In a hadronic scenario, the source distance is then likely < 2 kpc, although this of course would change for a more energetic progenitor SN event or a different cosmic-ray acceleration efficiency. The 70 Month Swift-BAT All-sky Hard X-Ray Survey \citep{2013ApJS..207...19B}, with a flux limit of \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ $1.3\times10^{-11}$ erg s$^{-1}$ cm$^{-2}$ over 90\% of the sky, reveals several X-ray sources classified as low-mass binaries in the region. This flux upper limit is not costraining for the models discussed here. A PWN has a characteristic spectral energy distribution with two humps, one in X-rays attributed to synchrotron radiation of very-high energy leptons and another in the gamma-ray range produced by IC scattering of ambient photon fields by energetic leptons \citep[e.g.,][]{2009ApJ...703.2051G}. Assuming similar energetics and properties to those of a relatively extended PWN at GeV energies, Vela X \citep{2013ApJ...774..110G}, which is 290 pc away \citep{2001ApJ...561..930C}, the source G350.6-4.7 would be closer or physically larger, but the LAT source associated to the Vela X PWN shows a relatively soft spectrum (unlike that of G350.6-4.7) and bright radio emission \citep{2013ApJ...774..110G}. For a similar or shorter distance, the radio emission from the PWN in G350.6-4.7 would be brighter but has not been seen. Thus the PWN scenario is less likely for G350.6-4.7. If there is no bright X-ray counterpart for G350.6-4.7, a possibility for the origin of the source is the IC scattering of energetic leptons in a relic PWN. The X-ray emission of evolved (relic) PWN is very low or absent leaving a VHE source with no counterpart. It is expected that the older a PWN becomes the lower its X-ray fluxes are, and the peak of the synchrotron hump is displaced to lower and lower energies while keeping the power-law component in the GeV band hard \citep{2012arXiv1202.1455M}. The details of a PWN evolution are expected to depend on the associated pulsar and the magnetic field, but are not well known. A typical prediction for the spectral energy distribution of an evolved 24 kyr-PWN taken from \cite{2013ApJ...773..139V} is shown in Fig. \ref{fig3} with the measured LAT spectral energy distribution of G350.6-4.7. The model was taken from that calculated for HESS J1507-622 in an ancient PWN scenario and it is scaled-up by a factor of 1.5. HESS J1507-622 is a peculiar source discovered in a H.E.S.S. Galactic Plane Survey \citep{2011A&A...525A..45H}, located relatively far from the Galactic plane ($\sim3\degr\llap{.}5$) and with no clear X-ray counterpart. However, for this model to be valid, G350.6-4.7 would have to be relatively close due to its large extent. It should be kept in mind that the relic PWN scenario is not well tested and HESS J1507-622 remains after all unidentified. Generally speaking, this is a plausible scenario for the origin of the high-energy photons if indeed there is no source counterpart at X-ray energies, although the particular model shown in Fig. \ref{fig3} fails to reproduce the shape of the spectral energy distribution. On the other hand there are several established SNRs showing gamma-rays of hadronic and leptonic origin. It is not yet known if there is a counterpart for G350.6-4.7 at lower energies but studies in the future might confirm the SNR origin of the emission. \begin{figure} \includegraphics[width=\columnwidth]{pwn} \caption{The same data shown in Fig. \ref{fig2} with a scaled-up relic PWN model (dashed lines) for the source HESS J1507-622.} \label{fig3} \end{figure} Finally, the region around G350.6-4.7 is located within the Fermi Bubbles, a pair of very large structures seen in gamma-rays by the \emph{Fermi} satellite \citep{2010ApJ...724.1044S}. The spectrum of the bubbles above 1 GeV is described by a power-law with an exponential cutoff with index $\sim1.87$ and cutoff energy of $\sim 113$ GeV, showing no spectral variations across the bubbles \citep{2014ApJ...793...64A}. It was shown in Section \ref{subsec:spec} that the spectrum of G350.6-4.7 is different and thus a possible relation between them is unlikely. Extended gamma-ray sources such as G350.6-4.7 offer an exciting possibility to carry out spatially-resolved spectral modeling with the LAT. This is relevant to understand particle diffusion in some sources interacting with ambient material as well as to probe the local acceleration properties of a source. It becomes important to constrain its distance. The necessary future studies to unveil the nature of G350.6-4.7 will surely benefit from observations at other wavelengths. | 16 | 9 | 1609.06225 |
1609 | 1609.04016_arXiv.txt | The purported spiral host galaxy of GRB\,020819B at $z=0.41$ has been seminal in establishing our view of the diversity of long-duration gamma-ray burst environments: optical spectroscopy of this host provided evidence that GRBs can form even at high metallicities, while millimetric observations suggested that GRBs may preferentially form in regions with minimal molecular gas. We report new observations from VLT (MUSE and X-shooter) which demonstrate that the purported host is an unrelated foreground galaxy. The probable radio afterglow is coincident with a compact, highly star-forming, dusty galaxy at $z=1.9621$. The revised redshift naturally explains the apparent nondetection of CO(3-2) line emission at the afterglow site from ALMA. There is no evidence that molecular gas properties in GRB host galaxies are unusual, and limited evidence that GRBs can form readily at super-Solar metallicity. | \label{sec:intro} The majority of long-duration gamma-ray bursts (GRBs) in the low-redshift universe originate from low-mass, low-metallicity, irregular dwarf galaxies \citepeg{LeFloch+2003,Fruchter+2006,Modjaz+2008,Graham+2013,Perley+2016}. GRB\,020819B\footnote{Occasionally designated as simply GRB\,020819.} stands out as the most notable exception. This burst was detected by the High Energy Transient Explorer (HETE-2) two years prior to the launch of Swift; follow-up using the Very Large Array identified a fading radio source (the probable afterglow) positionally coincident with the outskirts of a large spiral galaxy at $z=0.41$ (\citealt{Jakobsson+2005}; J05). The spiral galaxy is massive \citep{KupcuYoldas+2010}, and metal-rich (\citealt{Levesque+2010}; L10). Although the location of the radio afterglow is far (3$\arcsec$, or 16 kpc in projection) from the spiral's nucleus and a ``blob'' of optical emission distinct from the rest of the galaxy is visible at this location, spectroscopy reported by L10 established that this blob was at the same redshift as the spiral and that its metallicity was above Solar (12+log[O/H] = $9.0\pm0.1$, on the \citealt{KD02} diagnostic). This represented perhaps the clearest demonstration that GRBs can successfully form even in metal-rich environments. \begin{figure} \centerline{ \includegraphics[width=3.3in,angle=0]{020819image.eps}} \caption{Optical (GMOS $r$-band; grayscale) and H$\alpha$ (MUSE; contours) images of the region around GRB\,020819B. The radio afterglow position is marked as a blue circle (uncertainty radius 0.5$\arcsec$) and is consistent with the location of a marginally-extended ``blob'' of optical emission. Green lines indicate the inferred slit positions for the observations of L10 (solid: 2008, dashed: 2009). No significant H$\alpha$ emission at $z=0.41$ is present at the afterglow site in the MUSE image, or in the LRIS or X-shooter spectra.} \label{fig:imaging} \end{figure} The properties of this galaxy have also attracted follow-up at longer wavelengths. It is one of the few GRB hosts to be observed with the Atacama Large Millimetre Observatory (ALMA) to date, as part of the study of \citealt{Hatsukade+2014} (H14). While a source at the afterglow site is well-detected in the millimetre continuum ($F_\nu = 0.14$ mJy at 1.2\,mm), indicating that this region is rich in dust, \emph{no} molecular gas emission was detected in spectroscopic-mode observations covering the expected wavelength of the CO(3-2) emission line at $z=0.41$. The inferred gas-to-dust ratio is lower than for any well-studied class of galaxy, suggesting that GRBs may prefer regions of very low molecular gas density. This could originate from a different mode of star-formation in these regions, possibly involving the direct collapse of atomic gas \citep{Michalowski+2015}. No optical counterpart was identified for this burst and therefore no absorption redshift was obtained, so its association with the massive spiral hinges on two important assumptions. First, the variable radio source reported in the error circle by J05 must indeed be the GRB afterglow. Second, this radio source must have originated from within the spiral galaxy. These two assumptions are justified primarily by statistical arguments. The probability of an unrelated fading radio source appearing in the HETE-2 X-ray error circle and the probability of a source aligning with an unrelated massive foreground galaxy are both low ($\sim$\,$10^{-2}$), but not so low as to be implausible. The considerable importance given to this GRB merits further scrutiny of this issue. In this Letter, we re-analyze the archival data originally presented by L10 and H14 and also present extensive new observations of this system, including near-infrared spectroscopy of the blob at the afterglow site. Our observations demonstrate that the reported GRB counterpart lies in a background galaxy, unrelated to the $z=0.41$ spiral widely assumed by previous authors to be the host. Our observations and basic results are presented in \S 2. We reassess the probabilistic arguments used to associate the GRB with the radio counterpart, and the radio counterpart with the candidate host galaxies, in \S 3. Our conclusions, including a summary of the properties of the high-redshift galaxy that represents the probable host, are then presented in \S 4. | \label{sec:conclusions} A brief summary of the available multiwavelength constraints of the new, probable host galaxy is provided in Table \ref{tab:hostproperties}. % The H$\alpha$ luminosity and the millimetre flux (fit against a variety of starburst galaxy templates; \citealt{Silva+1998,Michalowski+2010}) both imply an SFR of between 30--120\,$M_\odot$\,yr$^{-1}$, consistent with a luminous infrared galaxy (LIRG). LIRGs are uncommon among GRB hosts generally, but have been observed to host heavily obscured GRBs before \citepeg{LeFLoch+2006,Perley+2013b}. Gravitational lensing by the foreground galaxy could conceivably contribute to the apparently high luminosity. % The foreground galaxy's halo mass is unknown, but for $M_{\rm encl}=10^{12}$\,$M_\odot$ the Einstein radius would overlap the burst location, magnifying both the GRB and its host by a potentially large factor. This would provide a natural explanation for the otherwise-coincidental foreground alignment and decrease the inferred luminosity and SFR of the high-$z$ galaxy somewhat. Direct evidence for lensing in the form of multiple images or episodes is, however, lacking. The metal dependence of GRB formation has been hotly debated for the past decade. As the GRB host with the highest well-determined metallicity (both overall and site-specific), GRB\,020819B has been instrumental in establishing a case for a GRB progenitor that is able to function at metallicities above $Z_\odot$. Our discovery that the nearby spiral is not the host therefore has important implications for the GRB metallicity dependence within the super-Solar regime. While there are other cases of super-Solar metallicity hosts \citepeg{Elliott+2013,Kruehler+2015}, most of these originate at high redshifts where offsets and metallicity gradients are difficult to measure. Low-$z$ examples are sparing, and generally also not secure: GRB\,050826, for example (12+log[O/H]=8.83; \citealt{Levesque+2010b}), also lacks an afterglow absorption spectrum or an associated supernova. Swift has now detected over 1000 bursts, some of which ($\sim$1\%; \citealt{Cobb+2008}) are destined to closely align with foreground galaxies. While it may yet be the case that GRBs can and do form above Solar metallicity, the frequency of such events is (at minimum) likely to be lower than previously believed. We suggest that future studies of this topic approach low-$z$ host associations not confirmed via supernova or afterglow spectroscopy with increased caution. More recently, the role of (deficient) molecular gas in GRB (progenitor) formation has also aroused significant interest and debate. % Direct evidence for a molecular gas deficiency via millimetric spectral-line observations is extremely sparse, based almost entirely on the singular example of the GRB 020819B site reported by H14. In fact, the other host galaxy observed by H14 (GRB\,051022) had a low but not anomalous gas-to-dust ratio; likewise, the gas mass in the host of GRB 080517 \citep{Stanway+2015} is normal, although it is being converted into stars very rapidly. With the GRB\,020819B result explained as a redshift mismatch, there is no direct evidence that the large-scale molecular gas properties around GRB sites are unusual. \begin{table} \begin{minipage}{90mm} \caption{Properties of the Probable Host Galaxy} \label{tab:hostproperties} \begin{tabular}{lll} \hline Property & Value & Reference \\ \hline Redshift & 1.9621 $\pm$ 0.0001 & \\ $B$ mag & 26.10 $\pm$ 0.50 $^{a}$ & J05 \\ $R$ mag & 23.98 $\pm$ 0.06 & J05 \\ $K$ mag & 20.80 $\pm$ 0.30 & J05 \\ H$\alpha$ flux & 14.9 $\pm$ 1.2 $^{b}$ & \\ H$\beta$ flux & 3.0 $\pm$ 0.9 & \\ \oii$\lambda$3727 flux & 7.2 $\pm$ 2.0 & \\ \oiii$\lambda$4959 flux & 3.2 $\pm$ 0.6 & \\ \oiii$\lambda$5007 flux & 9.7 $\pm$ 0.7 & \\ \nii$\lambda$6583 flux & 2.9 $\pm$ 1.7 & \\ 1.2 mm flux density & 140 $\pm$ 30 $\mu$Jy & H14 \\ 3 GHz flux density & 31 $\pm$ 8 $\mu$Jy & \citealt{Greiner+2016} \\ Velocity width ($\sigma$) & 190$\pm$20 km\,s$^{-1}$ & \\ \hline \end{tabular} \\ $^{a}$ Magnitudes are Vega, uncorrected for foreground extinction. \\ $^{b}$ Units of all line fluxes are 10$^{-17}$ erg cm$^{-2}$s$^{-1}$ \\ \end{minipage} \end{table} \vskip 0.02cm | 16 | 9 | 1609.04016 |
1609 | 1609.06539_arXiv.txt | {Powerful radio galaxies are often associated with gas-rich galaxy mergers. These mergers may provide the fuel to trigger starburst and active galactic nuclear (AGN) activity. In this Research Note, we study the host galaxies of three seemingly young or re-started radio sources that drive fast outflows of cool neutral hydrogen (\HI) gas, namely 3C\,293, 3C\,305 and 4C\,12.50 (PKS\,1345+12). Our aim is to link the feedback processes in the central kpc-scale region with new information on the distribution of stars and gas at scales of the galaxy. For this, we use deep optical V-band imaging of the host galaxies, complemented with \HI\ emission-line observations to study their gaseous environments. We find prominent optical tidal features in all three radio galaxies, which confirm previous claims that 3C\,293, 3C\,305 and 4C\,12.50 have been involved in a recent galaxy merger or interaction. Our data show the complex morphology of the host galaxies and identify the companion galaxies that are likely involved in the merger or interaction. The radio sources appear to be (re-)triggered at a different stage of the merger; 4C\,12.50 is a pre-coalescent and possibly multiple merger, 3C\,293 is a post-coalescent merger that is undergoing a minor interaction with a close satellite galaxy, while 3C\,305 appears to be shaped by an interaction with a gas-rich companion. For 3C\,293 and 3C\,305, we do not detect \HI\ beyond the inner $\sim$30-45 kpc region, which shows that the bulk of the cold gas is concentrated within the host galaxy, rather than along the widespread tidal features.} | \label{sec:intro} Deep optical broadband imaging has revealed that powerful radio galaxies are often associated with gas-rich galaxy mergers or interactions \citep[][see also work by \citealt{hec86}, \citealt{smi89}, \citealt{roc00}, \citealt{sab13}]{ram12}. Furthermore, powerful radio galaxies often show young stellar populations and contain dust masses that link them to merger activity \citep{tad11,tad14a}. These gas-rich mergers and interactions are believed to deposit the cold material that is needed to fuel both starburst activity \citep[e.g.,][]{dim07} and the powerful radio sources \citep[e.g.,][]{har07}. Although the majority of early-type galaxies outside clusters contain at least modest amounts of cold gas \citep[few $\times$ 10$^{6-9}$ M$_{\odot}$;][]{mor06,oos10,ser12}, the host galaxies of small radio sources can be particularly rich in \HI\ \citep{emo07}. Moreover, young or recently re-started radio sources often contain large amounts of \HI\ gas in their central regions \citep{pil03,ver03,gup06a,gup06,cha11,cha13,ger14}. On average, these central regions appear much richer in \HI\ than those occupied by extended radio sources \citep{gup06,cha13,ger14}. Moreover, the \HI\ profiles associated with young radio sources more often show blue wings, which suggests that these sources drive gas outflows as they clear their way through the rich ambient medium of the host galaxy \citep{cha11,ger15}. Particularly interesting are young or re-started radio sources that have been observed to drive very fast ($\sim$1000 \kms) outflows of cool neutral and cold molecular gas from the central kpc-scale region \citep{oos00,mor03,mor05,mor05sample,das12,mor13sci,mor13,mah13}. The energy released by these jet-driven outflows approaches values needed to clear the central region of gas and quench star formation \citep{mor13,mah15}, certainly when taking into account that part of the kinetic energy of the jets is put into heating of the cold gas \citep{gui12,har12,tad14,das14,lan15}. Thus, the jet-induced feedback may influence the evolution of these galaxies. In this paper, we further investigate the link between the nuclear feedback phenomena and large-scale merger processes for three of these seemingly young or re-started radio galaxies with fast outflows of cold gas, namely 3C\,293, 3C\,305 and 4C\,12.50 (PKS\,1345+12). The cold outflows in these systems were identified through blueshifted \HI\ seen in absorption against the strong radio continuum \citep{mor03,mor05,mor13sci,mah13}. Previous optical imaging revealed evidence for tidal debris from a galaxy merger or interaction in these systems \citep{hec86}. We here present more detailed optical imaging, combined with optical spectroscopy and deep \HI\ observations. Our aim is to identify the stage of merger or interaction in these systems. With this, we complement the studies of the feedback processes in the kpc-scale central region with new information on the distribution of stars and gas at scales of the host-galaxy environment. Throughout this paper we will assume $H_{0} = 71$\,\kms\,Mpc$^{-1}$, $\Omega_{\rm M} = 0.27$ and $\Omega_{\Lambda} = 0.73$. | \begin{figure} \centering \includegraphics[width=0.47\textwidth]{AA_2016_28592_fig3.pdf} \caption{Optical V-band imaging of 4C\,12.50 (PKS\,1345+12). The various plots show the same image of 4C\,12.50, but with different intensity-scaling and zooming. Potential companion galaxies discussed in the text are marked with arrows in the left plot.} \label{fig:PKS1345} \end{figure} The deep optical imaging that we presented of 3C\,293, 3C\,305 4C\,12.50 reveals galaxy-morphologies that agree with those previously shown by \citealt{hec86} (also \citealt{hec82}, \citealt{bre84}). However, our data show in more detail a multitude of subtle features. Specifically, combined with our optical spectroscopy and \HI\ imaging, we obtain the following new insights on these three radio galaxies:\\ \vspace{-2mm}\\ $\bullet$ 3C\,293: We confirm the presence and redshift of a close companion at $\sim$13 kpc distance, beyond which a stellar tail with a 'zig-zag' morphology ends in two possible starforming regions.\\ \vspace{-2mm}\\ $\bullet$ 3C\,305: The major axis of the host galaxy seems to turn by $\sim$75$^{\circ}$ from the faint outer envelope with the two tidal arms to the bright inner region. The inner region shows arc- or spiral-like isophotes. There is a tentative indication that a faint \HI\ feature stretches from an \HI-rich companion towards 3C\,305, along the direction of the radio axis.\\ \vspace{-2mm}\\ $\bullet$ 4C\,12.50: A double tidal-tail stretches in the direction of a bright companion galaxy. On the opposite side, a faint but broad fan of emission stretches up to $\sim$85 kpc from the galaxy.\\ \vspace{-2mm}\\ Our results also reveal that, for 3C\,293 and 3C\,305, no \HI\ has been detected along the stellar tidal debris beyond the central region where the \HI\ absorption-signal dominates. However, we note that sensitive \HI\ observations with a spatial resolution that better matches the extended optical features are needed to better study any potential widespread \HI\ gas in emission. \subsection{Merger state and AGN triggering} Our results indicate that all three systems have been involved in a recent galaxy merger or interaction, which is consistent with the presence of young or post-starburst stellar populations across these systems \citep{tad05}. 3C\,305 was classified by \citet{hec86} as a post-merger system. However, contrary to the claim by \citet{hec86} that there are no close companion galaxies that could have provoked the observed morphological disturbances, we show that 3C\,305 appears to be involved in a galaxy-encounter with a gas-rich companion. This resembles the case of NGC\,6872, where two tidal arms were formed by the low-inclination, prograde passage of companion IC\,4970 \citep[][see also \citealt{mih93}]{hor07}. If a similar passage created the tidal arms in 3C\,305, and assuming that the companion has an average velocity of $\sim$200 \kms\ with respect to 3C\,305, then the time since closest approach must have occurred $\sim$650\,Myr ago. This time-scale is similar to the 0.5\,$-$\,1 Gyr age of the young stellar population across 3C\,305 \citep{tad05}. It could suggest that a galaxy-wide starburst was triggered at the time of closest approach between the two systems, which is consistent with simulations \citep[e.g.,][]{pei10}. Additional observations are needed to investigate this. For 3C\,293, the distorted optical morphology and single nucleus \citep{flo06} suggest that this is a post-merger system which is currently experiencing a minor interaction with the close companion 13\,kpc towards the west-south-west. 4C\,12.50 shows clear evidence from its double nucleus and multiple tidal features that it is currently involved in an ongoing major --possibly multiple-- merger event. This is consistent with its high IR luminosity ($L_{\rm IR} \sim 2 \times 10^{12} L_{\odot}$, that is, in the regime of ultra-luminous infrared galaxies; \citealt{san88}). These results indicate that 3C\,305, 3C\,293 and 4C\,12.50 are likely at different stages in the merger process. Interestingly, this also implies that the radio sources in 4C\,12.50, 3C\,305 and 3C\,293 could have been triggered at different stages during the merger. After all, 4C\,12.50, 3C\,305 and 3C\,293 are classified as a Gigahertz Peaked Spectrum (GPS), Compact Steep Spectrum (CSS) and Steep Spectrum Core (SSC) source, respectively, which likely represent young or re-started radio sources \citep[$\le$10$^{6}$ yr;][]{fan95,ode98,shu12,col16}. This different stage of triggering would be consistent with optical and IR studies, which indicate that a radio source can in general occur at any time during the merger process, either before or after the individual nuclei coalesce \citep{ram11,tad11,dic12}. We caution that, while both 3C\,293 and 4C\,12.50 have radio cores with an estimated age of $\sim$3\,$\times$\,10$^{4}$ yr \citep{aku96,ode00}, they also contain extended radio lobes \citep[e.g.,][]{bes04,sta05}. Although we assume that these outer lobes represent a previous episode of activity, we cannot rule out that the radio cores are older than they appear from their spectral-index measurements \citep[see][]{blu00}, for example, because they have been frustrated by a dense inter-stellar medium \citep[e.g.,][]{ode91}. If the radio cores are actively feeding the much larger and older radio structures, then the moment of triggering is uncertain. However, we find this scenario less likely, because 3C\,293, 3C\,305 and 4C\,12.50 are among the most powerful steep-spectrum sources at low-$z$ and display among the most extreme jet-ISM interactions known in the low-$z$ Universe. Such features are often associated with young or re-started radio sources that are trying to plough their way through the ambient ISM \citep[Sect.\,\ref{sec:intro}; see also][]{hol08,hol11}. Concluding, our results support the notion that galaxy mergers and interactions are the likely mechanisms for accumulating cold gas in the central few kpc of these systems, and possibly also for (re-)triggering the powerful radio sources that violently interact with this dense ISM. | 16 | 9 | 1609.06539 |
1609 | 1609.06680_arXiv.txt | We develop a nonlinear semi-parametric Gaussian process model to estimate periods of Miras with sparsely sampled light curves. The model uses a sinusoidal basis for the periodic variation and a Gaussian process for the stochastic changes. We use maximum likelihood to estimate the period and the parameters of the Gaussian process, while integrating out the effects of other nuisance parameters in the model with respect to a suitable prior distribution obtained from earlier studies. Since the likelihood is highly multimodal for period, we implement a hybrid method that applies the quasi-Newton algorithm for Gaussian process parameters and search the period/frequency parameter space over a dense grid. A large-scale, high-fidelity simulation is conducted to mimic the sampling quality of Mira light curves obtained by the M33 Synoptic Stellar Survey. The simulated data set is publicly available and can serve as a testbed for future evaluation of different period estimation methods. The semi-parametric model outperforms an existing algorithm on this simulated test data set as measured by period recovery rate and quality of the resulting Period-Luminosity relations. | The determination of reliable periods for variable stars has been an area of interest in astronomy for at least four centuries, since the discovery of the variability of Mira ($o$~Ceti) by Fabricius in 1596 and the first attempts to determine its period by Holwarda \& Bouillaud in the mid-1600s. The availability of electronic computers for astronomical research half a century ago enabled the development of many algorithms to estimate periods quickly and reliably, such as \citet{Lafler1965,Lomb1976,Scargle1982}. The aforementioned algorithms work best in the case of periodic variations with constant amplitude and Mira variables present several challenges in this regard. While their periods of pulsation are stable except for a few intriguing cases \citep{Templeton2005}, Mira light curves can exhibit widely varying amplitudes from cycle to cycle \citep[see, for example, the historical light curve of Mira compiled by][]{Templeton2009}. In the case of C-rich Miras, the stochastic changes in mean magnitude across cycles \citep[e.g.,][]{Marsakova1999} only complicate the problem further. The wide variety of light curves for long-period variables, already recognized by \citet{Campbell1925} and \citet{Ludendorff1928}, may complicate the identification of Miras among other stars. Lastly, from a purely practical standpoint, it is simpler to obtain light curves spanning several cycles for RR Lyraes or Cepheids (with periods ranging from $\sim 0.5$ to $\sim 100$~d) than for Miras (with periods ranging from $\sim 100$ to $\sim 1500$~d). Despite these challenges, the identification and determination of robust periods for Miras --- especially in the regime of sparsely sampled, low signal-to-noise light curves --- would be very beneficial for the determination of distances to galaxies of any type. Thanks to the unprecedented temporal coverage of the Large Magellanic Cloud (LMC) by microlensing surveys, the availability of large samples of extremely well-observed Miras has led to a thorough characterization of their period-luminosity relations at various wavelengths \citep{Wood1999,Ita2004,Soszynski2007}. The dispersion of the $K$-band period-luminosity relation \citep[$\sigma=0.13$~mag]{Glass2003}, is quite comparable to that of Cepheids at the same wavelength \citep[$\sigma=0.09$~mag]{Macri2015} and makes them competitive distance indicators. The third phase of the OGLE survey \citep{Udalski2008} imaged most of the LMC with little interruption over 7.5 years and resulted in the discovery of 1663 Miras \citep{Soszynski2009} with a median of 466 photometric measurements per object. The temporal sampling of these light curves and their photometric precision are exceptional relative to typical astronomical surveys and make period estimation relatively easy. In comparison, a similar span of observations of M33 by the DIRECT \citep{Macri2001} and M33SSS projects \citep{Pellerin2011} in the $I$-band consists of a median number of 44 somewhat noisy measurements, heavily concentrated in a few observing seasons. Representative Mira light curves from the OGLE \& DIRECT/M33SSS surveys are shown in Fig.~\ref{fig:example.mira.lc}. There are several reasons for the striking difference in quality between these two data sets. The LMC Miras are among the brightest objects in the OGLE fields, whereas their M33 counterparts are among the faintest in the aforementioned surveys of this galaxy. While the effective exposure times of all these surveys are quite comparable, after taking into account differences in collecting area of their respective telescopes, M33 lies approximately 6.2~mag farther in terms of its $I$-band apparent distance modulus. Furthermore, the main goal of the OGLE project (detection of microlensing events) requires a very dense temporal sampling of the survey fields; this is achieved by using a dedicated telescope and is helped by the fact that the LMC is observable nearly all year long from the site. In contrast, the observations of M33 were carried out using shared facilities (available only a few nights per month) with the primary purpose of studying Cepheids and eclipsing binaries (which do not require exceptionally dense temporal sampling), and the galaxy is only observable all night long for $\sim 1/3$ of the year. Standard period estimation algorithms, which work well for high signal-to-noise, well sampled light curves such as those obtained by OGLE, will fail on more typical data sets represented by the M33 observations. The purpose of this work is to develop and test a methodology for estimating periods for sparsely sampled, noisy, quasi-periodic light curves such as those of Miras observed in M33 by the aforementioned projects. \begin{figure}[htbp] \label{fig:example.mira.lc} \includegraphics[width=0.49\textwidth]{fig01a.eps} \includegraphics[width=0.49\textwidth]{fig01b.eps} \caption{Representative Mira light curves observed by OGLE-III in the Large Magellanic Cloud (top) and DIRECT/M33SSS in M33 (bottom).} \vspace*{-12pt} \end{figure} The rest of the paper is organized as follows. In \S\ref{sec.background} we review several existing period estimation methods. In \S\ref{sec.model} we introduce a new semi-parametric (SP) model for Mira variables which uses a Gaussian process to account for deviations from strict periodicity. We use maximum likelihood to estimate the period and the parameters of the Gaussian process, while other nuisance parameters in the model are integrated out with respect to some prior distributions using earlier studies. Since the likelihood is highly multimodal for the period/frequency parameter, we implement a hybrid method that applies the quasi-Newton algorithm for Gaussian process parameters and a grid search for the period/frequency parameter. In order to assess the effectiveness of the SP model, in \S\ref{sec.construct.test} we carefully construct a simulated data set by fitting smooth functions to the light curves of well-observed OGLE LMC Miras and resampling them at the cadence, noise level, and completeness limits of the aforementioned M33 observations. Using the simulated data, in \S\ref{sec.evaluation} we compare the performance of existing period estimation methods to our SP model. We find that our proposed model shows an improvement over the generalized Lomb-Scargle (GLS) model under various metrics. In \S\ref{sec.discussion}, we conclude and discuss some future applications. Simulated light curves for reproducing the results in the paper and performance benchmarking are made publicly available as supplementary material. | \label{sec.discussion} In this paper, we developed a nonlinear SP Gaussian process model for estimating the periods of sparsely sampled quasi-periodic light curves, motivated by the desire to detect Miras in an existing set of observations of M33. We conducted a large-scale high-fidelity simulation of Mira light curves as observed by the DIRECT/M33SSS surveys to compare our model with the GLS method. Our model shows improved accuracy under various metrics. The simulation data set is provided as a testbed for future comparison with other methods. The SP model will be used in a companion paper to search for Miras in M33, estimate their periods, and study the resulting PLRs. \ \par SH was partially supported by Texas A\&M University-NSFC Joint Research Program. WY \& LMM acknowledge financial support from the NSF through AST grant \#1211603 and from the Mitchell Institute for Fundamental Physics and Astronomy at Texas A\&M University. JZH was partially supported by NSF grant DMS-1208952. The authors acknowledge the Texas A\&M University Brazos HPC cluster that contributed to the research reported here. | 16 | 9 | 1609.06680 |
1609 | 1609.08654_arXiv.txt | Strongly magnetized accreting stars are often hypothesized to be in `spin equilibrium' with their surrounding accretion flows, which requires that the accretion rate changes more slowly than it takes the star to reach spin equilibrium. This is not true for most magnetically accreting stars, which have strongly variable accretion outbursts on time-scales much shorter than the time it would take to reach spin equilibrium. This paper examines how accretion outbursts affect the time a star takes to reach spin equilibrium and its final equilibrium spin period. I consider several different models for angular momentum loss -- either carried away in an outflow, lost to a stellar wind, or transferred back to the accretion disc (the `trapped disc'). For transient sources, the outflow scenario leads to significantly longer times to reach spin equilibrium ($\sim$10x), and shorter equilibrium spin periods than would be expected from spin equilibrium arguments, while the `trapped disc' does not. The results suggest that disc trapping plays a significant role in the spin evolution of strongly magnetic stars, with some caveats for young stellar objects. | The spin rate of a star is strongly affected by the presence of a magnetic field. Neutron stars, for example, show a clear inverse correlation between magnetic field strength and spin rate: the fastest millisecond pulsars ($P_{\rm spin} \sim 0.002$--$0.005~{\rm s}$) have a typical field strength of $B\sim 10^8~{\rm G}$, while magnetars with $B\sim 10^{14}~{\rm G}$ have typical spin periods of a few seconds. The influence of the magnetic field is even stronger when such stars are accreting gas. Although accreted gas adds considerable angular momentum to the star, accreting magnetized stars generally spin well below their break-up velocity (sometimes many orders of magnitude slower), indicating that the presence of the magnetic field is able to regulate the transport of angular momentum between the star and surrounding gas. The stellar magnetic field in fact strongly affects the dynamics of the accreting gas, and couples the star to its surrounding environment. Close to the star, matter is forced to flow along field lines on to the magnetic poles. At the boundary of this region (typically called the {\it magnetospheric} or {\it Alfv\'en radius}, \rin), the magnetic field can in turn be significantly distorted by the gas, which exerts a torque on the star. The sign of the torque depends on the relative location between \rin\ and the {\it co-rotation radius}, $r_{\rm c} \equiv (GM/\Omega^2_*)^{1/3}$, or the location where a Keplerian disc co-rotates with the star. If $r_{\rm m} > r_{\rm c}$, the star spins faster than the inner disc, so that field lines coupling the two will gradually spin down the star. The location of the magnetospheric radius itself is chiefly determined by the stellar magnetic field and accretion rate, although it is also sensitive to the detailed interaction between the gas and the magnetic field. This basic picture leads naturally to the concept of `spin equilibrium' (or `disc locking' in young stars), whereby the star's spin rate gradually adjusts itself until the net torque on the star is roughly zero and the accretion flow is truncated near the co-rotation radius, $r_{\rm m} \simeq r_{\rm c}$. In this way the star's dipolar magnetic field can be estimated, provided the spin and accretion rate are known. Assuming spin equilibrium is reached requires assuming a steady mass accretion rate -- i.e. that the timescale on which the accretion rate changes is generally much longer than the `spin equilibrium time' (\teq), defined as the time the star takes to reach its `spin equilibrium period' (\peq). It is not clear that this assumption is widely valid for accreting magnetized stars, either compact stars (magnetized white dwarfs and neutron stars) or young stellar objects (YSOs). Most magnetized compact stars in binary systems are transient, showing short accretion outbursts followed by long periods of quiescence. In weak-field accreting neutron stars, the low-mass X-ray binaries (LMXBs), the observed duty cycle is on average 3 per cent, but can be well below 1 per cent when allowing for the limited observing baseline \citep{2015ApJ...805...87Y}. The luminosity difference in LMXBs between outburst and quiescence can span many orders of magnitude, suggesting a huge change in accretion rate. High magnetic field transient neutron stars can also show strong variability. In one particular class of system, Be X-ray binaries (neutron stars that accrete from the wind or disc surrounding a Be star), the duty cycles are $\sim 5-20$ per cent \citep{2011Ap&SS.332....1R,2014MNRAS.437.3863K}, and the dynamic range can be 1--5 orders of magnitude between outburst and quiescence. At least some young stellar objects (YSOs) also show large-scale variability, but its prevalence is much harder to constrain, since they evolve on much longer time-scales than compact binaries. The most dramatic accretion outbursts are FU Ori-type outbursts, where the luminosity increases by $\sim 1000$ \citep{1996ARA&A..34..207H}, with an outburst duration of at least decades and recurrence time of several thousand years. Strong luminosity variations of about 1--2 orders of magnitude on shorter (years) time-scales are also sometimes seen (the `EXor' class; \citealt{2014prpl.conf..387A}, suggesting that the mean accretion rate can vary considerably at different points in the TTauri phase. Variations in accretion rate may help explain observations in magnetospherically accreting systems that do not easily fit in to the standard spin equilibrium picture. YSOs with discs, for example, show clear indications of magnetic field regulated rotation, spinning well below their break-up values. However, attempts to confirm disc locking have been mixed, or seemed to contradict simple model predictions \citep{2012ApJ...756...68C}. Another example: a recent survey of the spin-rates in Be X-ray binaries found that the neutron star frequently rotates much more slowly than would be expected for a moderate ($10^{12}$~G) magnetic field star in spin equilibrium, suggesting that much larger fields ($10^{14}$--$10^{15}$~G) are present \citep{2014MNRAS.437.3863K}. The large number of such binaries makes this unlikely from a population point of view, and it also seems to contradict magnetic field estimates from cyclotron lines in analogous Galactic systems with similar spin rates and luminosities \citep{2014MNRAS.437.3664H}. This paper investigates how large-amplitude, short-timescale accretion rate variations affect the spin evolution of the star, and how this evolution changes for different models for stellar angular momentum loss. As described in more detail below, it is not clear whether most angular momentum is lost through stellar outflows (winds from the star, or at the disc--magnetic field interface) or whether angular momentum is mainly lost to the accretion disc. As I demonstrate below, different angular momentum loss mechanisms lead to different predictions for spin evolution as a function of accretion rate, so that comparing the long-term spin evolution of each model with different accretion rate profiles may offer new observational tests to distinguish between them. | This paper compares the long term spin evolution of magnetized stars using different models for angular momentum regulation, explicitly considering the effects of time-variable accretion. Here I briefly discuss the consequences of the results presented in Section~\ref{sec:results} for three (very different) types of magnetically-accreting, outbursting stars: TTauri stars, AMXPs and Be/X-ray binaries. I focus on two questions in particular: \begin{enumerate} \item{Can observations be used to distinguish between the trapped disc pictures and other models for spin regulation?} \item{How does considering a variable accretion rate alter predictions of the observable properties of strongly magnetized accreting stars?} \end{enumerate} \subsection{Accreting millisecond X-ray pulsars and LMXBs} \label{sec:AMXP} Accreting neutron stars with low magnetic fields ($\lesssim 10^8$ G) with low-mass companions (low-mass X-ray binaries, or LMXBs) are thought to be the progenitors of radio millisecond pulsars (e.g. \citealt{1982Natur.300..728A}), spun up to millisecond spin periods via accretion over hundreds of millions of years. AMXPs are a subset of this group that show coherent pulsations and accretion outbursts, with peak luminosities reaching $\sim 20$ per cent $L_{\rm Edd}$ (although most remain much fainter). A second, partially overlapping subset of LMXBs have spin periods measured through quasi-periodic oscillations (`burst oscillations'), which are produced by localized, accretion-induced nuclear burning on the star's surface modulated by the star's rotation. These are an additional useful sample since the accretion rates in burst oscillation sources can be significantly larger than AMXPs. The spin periods inferred from burst oscillations are shorter on average than in than AMXPs (although the sample size remains small; \citealt{2014A&A...566A..64P}). No periodicity has been detected in all remaining LMXBs, despite some very deep searches \citep{2015ApJ...806..261M}, which could mean that the magnetic field in these sources is not strong enough to channel the accretion flow, at least during the brightest phases of the outburst. What limits the spin frequency of the millisecond pulsars? Despite having relatively weak fields ($\sim 10^{8}$G; inferred from dipole spin-down) and long accretion times (the donor star lifetime is often $>1$ Gyr), the fastest radio millisecond pulsar has a spin period of 1.4ms \citep{2006Sci...311.1901H}, much longer than the theoretical mass-shedding limit of $\sim 0.7$ms. Two possible mechanisms have been proposed -- gravitational wave emission from a spin-induced quadrupole moment or $r$-modes \citep{1998ApJ...501L..89B, 1999ApJ...516..307A}, or the spin-down effects from the magnetic field/disc interactions (e.g. \citealt{2012ApJ...746....9P}), but it has proven difficult to definitively distinguish between them. The spin distribution of radio millisecond pulsars peaks at a significantly longer spin period than that of AMXPs. \cite{2012Sci...335..561T} has recently suggested that the difference in spin between the two populations could be significantly affected by the evolution of the mass-transferring companion. In this picture, AMXPs undergo a strong spin down during the `Roche lobe decoupling phase' as companion stops filling its Roche lobe so that the mass transfer rate to the pulsar decreases. \cite{2012Sci...335..561T} estimated roughly 50 per cent of the pulsar's angular momentum can be lost during this phase, during which the average accretion rate drops by $\sim 3$ orders of magnitude. The present work challenges the assertion that a decrease in $\dot{M}$ will efficiently spin the star down particularly if the accretion/ejection torque picture is the most relevant one. The results of Section~\ref{sec:results} demonstrate the uncertainty in estimating \peq\ and \teq: between uncertainties in the star-disc interactions (e.g. the parameter $\xi$), the angular momentum loss mechanism, and the presence of accretion outbursts, \peq\ and \teq\ can both easily be uncertain by 10$\times$, even when the physical parameters of the system ($\dot{M}$, $P_*$, $B_*$, $\dot{P}$) are well constrained. \teq\ lengthens with declining $\dot{M}$, so that as $\dot{M}$ decreases it takes progressively longer for the star to reach a new spin equilibrium. The results in this paper show that once outbursts are considered, \teq\ increases up to $10\times$. Observations of AMXPs suggest that {\em none} of them are in spin equilibrium with their time-averaged accretion rates (considering both quiescence and outburst). On the other hand, systems with well constrained spin derivatives show much less spin up during outburst than might be expected from their luminosity \cite{2012arXiv1206.2727P}, which could indicate spin equilibrium. \cite{2008MNRAS.389..839W} finds the average luminosity (including quiescence) in AMXPs varies between $6\times10^{-5}$ and $0.02L_{\rm Edd}$ which implies (assuming radiative efficiency and the average NS parameters adopted in this paper) \peq\ $\sim 1$--$11$ms. AMXPs have observed spin periods between $1.7-5.5$ms, with no obvious trend as a function of mean luminosity (although some luminosities are uncertain by up to $10\times$ from distance and bolometric uncertainties). In particular, the recently discovered `transitional pulsars' \citep{2009Sci...324.1411A, 2013Natur.501..517P, 2014MNRAS.441.1825B}, which switch between states of active accretion and radio pulsations, all have relatively fast spin periods (1.7-3.9ms), despite extremely low accretion rates during outbursts for two of the three systems. For one of these sources, PSR J1023+0038, recent analysis of the X-ray pulsations has found spin down during outburst is moderately larger than dipolar spindown \cite{2016ApJ...830..122J} measured when the accretion disc is absent in the radio-loud phase. Nonetheless, RMSPs are observed to spin (on average) significantly slower than AMXPs, which would be possible even if a trapped disc remains present to spin down the star even at very low $\dot{M}$. The spin-down in this case could happen gradually over the entire long term decay phase of $\dot{M}$, rather than mainly being focused at early times in the `Roche-lobe decoupling phase', as suggested by \cite{2012Sci...335..561T}. If the final large decline in $\dot{M}$ is not able to significantly spin down most pulsars in their late accretion phase, the question of what sets their maximum spin rate again becomes more urgent. In this paper, the `canonical' \peq\ for an AMXP is about 0.4ms at $\dot{M}_{\rm Edd}$, but all simulations with outbursting accretion show slower rotation rates, typically by $\sim1.5$--$3\times$ but up to $10\times$ in some cases. On the other hand, \teq\ at $\dot{M}_{\rm Edd}$ is around 50Myr (and increases when outbursts are considered). This is much shorter than the lifetimes of these systems, and (based on the observed sample of LMXBs) is unlikely to dominate the lifetime accretion rate of the star. As long as the star has a $\sim10^{8}$G field, a lifetime average $\dot{M} \sim 0.1-0.01\dot{M}_{\rm Edd}$ can limit the final spin period to within observed values without invoking an additional spin-down source like gravitational waves. (This is before considering modifications to the spin-up rate, e.g. \citealt{2005MNRAS.361.1153A}, which may limit angular momentum transfer at high $\dot{M}$). \subsection{Be/X-ray Binaries} \label{sec:Be} In strongly magnetized accreting neutron stars ($B\sim10^{12}$G), dipole radiation is unimportant for spin regulation compared with spin change from accretion, and the spin rate of the star is determined by the interaction between the magnetic field and the accretion flow. The observed spin distribution ($P_*\sim 1$--$1000$s) of these systems is much larger than in AMXPs, and many systems are observed to spin up or down considerably. However, many accreting high-field neutron stars have high mass ($M>3M_\odot$) companions and are believed to mainly accrete from a wind rather than a disc (e.g. \citealt{1997ApJS..113..367B}), which is thought to give a much larger spread in $P_*$ and $\dot{P}_*$ than results from disc accretion. A possible exception to this are Be/X-ray binaries, in which the neutron star undergoes accretion outbursts when it passes through the decretion disc of a companion Be star. Based on angular momentum conservation arguments, \cite{2014MNRAS.437.3863K} argue that as the pulsar passes through the Be star's disc most of the gas entering the pulsar's sphere of influence will have too much angular momentum to fall on to the star directly, implying that an accretion disc should form around the neutron star. The \textit{XMM--Newton} survey of the Small Magellanic Cloud (SMC) has tracked pulsars in Be X-ray binaries in the SMC over the past 14 yr, providing a unique data set to test spin evolution models \citep{2010ASPC..422..224C}. \cite{2014MNRAS.437.3664H} and \cite{2014MNRAS.437.3863K} argue that the small observed spin period derivatives suggest spin equilibrium (or else extremely low magnetic fields), and, if spin equilibrium is assumed, a surprisingly large fraction of Be X-ray binaries in the SMC should have magnetar-strength magnetic fields ($\sim 10^{14}$G). This is in contrast to systems in our own Galaxy with similar spin rates and luminosities, which have magnetic field estimates from cyclotron resonance emission lines on the order $B\sim 10^{12}$G. The conclusions of this paper suggest a somewhat different interpretation of the observations discussed by \cite{2014MNRAS.437.3863K}, which reduce (although do not completely eliminate) the need for a very large magnetic field in most pulsars. Be X-ray binaries are generally transient, so that their average luminosity is much lower (typically several orders of magnitude) than their luminosity in outburst. To estimate the magnetic field, \cite{2014MNRAS.437.3863K} assume that the star is in spin equilibrium {\em with the outburst accretion rate} (see equation~\ref{eq:Peq}). This can be reasonable assumption if the accretion/ejection model applies, since in quiescence the torque on the star is strongly reduced. However, if a trapped disc remains present during quiescence, the star continues to spin down, and it is more accurate to consider the {\em average} $\dot{M}$ rather than the outburst $\dot{M}$. (In fact, \textit{Fermi} observations of some Be-X-ray binary systems indeed show that they spin down between outbursts, see e.g. \citealt{2015arXiv150204461S}.) To see how the results of this paper could affect estimates of $B_*$ in these systems, I calculate \peq\ (equation (\ref{eq:Peq})) assuming that spin equilibrium has been reached, using \Mavg\ rather than \Mout\ (as was assumed by \citealt{2014MNRAS.437.3863K}). A rough estimate of \Mavg\ for the stars in \citep{2014MNRAS.437.3863K} is given by: \begin{equation} \langle\dot{M}\rangle \simeq \dot{M}_{\rm out}F_{\rm out}, \end{equation} where $\dot{M}_{\rm out} \simeq 0.01$--$0.2\dot{M}_{\rm Edd}$ (the inferred accretion rate from the outburst luminosity), and $F_{\rm out} \simeq N_{\rm det}/N_{\rm obs}$ is the fraction of time spent in outburst (the ratio between the number of detections to observations). \cite{2014MNRAS.437.3863K} report 1-2 weekly observations (I use 84 observations/yr) over a timespan ranging from 0.15 to 14 years, which corresponds to $F_{\rm out} \simeq 0.004$--$1$ ($\langle F_{\rm out}\rangle \sim 0.06$) and $\langle\dot{M}\rangle \simeq 7.5\times10^{-5}$-- $0.2~\dot{M}_{\rm Edd}$. This assumes that the quiescent luminosity of these sources is at least 100$\times$ lower than in outburst, which seems roughly consistent with observations (Coe, private communication). Using equation~(\ref{eq:Peq}), the estimated \Mavg, and the reported period for each pulsar from \cite{2014MNRAS.437.3863K}, I estimate a revised magnetic strength, using either the accretion/ejection or trapped disc model. For simplicity I choose the `canonical' accretion/ejection and trapped disc models from Section~\ref{sec:spin_equilibrium}, scaled to Be X-ray binary parameters. The resulting $P_{\rm eq}$ is 0.9$P_{\rm eq,0}$\footnote{i.e., $P_{\rm eq}$ from equation~(\ref{eq:Peq})} for a trapped disc, and $\sim 0.3P_{\rm eq,0}$ for the accretion/ejection model. The resulting estimated magnetic fields are shown in Fig.~\ref{fig:BvsP}. As is clear from the figure, using a time-averaged accretion rate rather than the outburst one gives systematically lower estimates for $B$ regardless of the torque model, but if the systems are able to efficiently spin down during quiescence (by transferring angular momentum into a disc), there is no need for the majority of systems to harbour magnetar-strength fields. Since the time-scales for reaching spin equilibrium in Be X-ray binaries are much shorter than for either TTauri stars or XMSPs, this result provides the strongest evidence for trapped discs around strongly magnetic stars. \begin{figure} {\includegraphics[width=90mm]{Figures/Be_BvsP.eps}} \caption{Estimated magnetic field as a function of the spin period of Be X-ray binaries in the the SMC. The green triangles show the values calculated by \protect\cite{2014MNRAS.437.3863K} (which are roughly equivalent to using the reported outburst accretion rate and measured spin periods in equation~(\ref{eq:Peq})). The cyan circles show the same data set using the accretion/ejection model and the time-averaged accretion rate, while the red squares show the same results for the trapped disc model. The black dashed line shows the quantum critical field, $B_{\rm crit} = 4.4\times10^{13}$G, where the cyclotron energy is comparable to the electron rest-mass energy, which is commonly used to define a `magnetar'. Using the time-averaged accretion rate to estimate $B$ and assuming a trapped disc persists in quiescence obviates the need for magnetar-strength magnetic fields. \label{fig:BvsP}} \end{figure} \subsection{Young Stellar Objects} \label{sec:YSO} TTauri stars also show strong evidence for spin regulation from interaction with an accretion disc \citep{2007prpl.conf..479B}, and most TTauri stars with discs spin well below their breakup rate, despite the fact that they contract as they evolve. The different mechanisms for angular momentum regulation discussed in this paper are thus relevant for these stars as well. TTauri stars are also often variable, showing variability on different time-scales. If the variability is caused by large accretion rate variations on to the stellar surface, then this should also affect the spin equilibrium rate of the star, as discussed throughout this paper. TTauri stars are more similar to AMXPs than high-field neutron stars, with a much smaller magnetosphere that is probably completely crushed at high $\dot{M}$. Since variability time-scales are much longer in TTauri stars than neutron stars, it is not straightforward to determine whether all TTauri stars are variable. Recent work looking at variability has found that the most common variability -- fluctuations on short time-scales (days to weeks) is most likely due to variations on the stellar surface that become apparent as the star rotates \citep{2014MNRAS.440.3444C}. However, larger scale variability (which is observed in a subset of TTauri stars) {\em is} attributed to accretion rate fluctuations. Variations of $\sim 10$--100 with time-scales of a few years are seen in a subclass of TTauri stars known as `EXors', after the prototype, EX Lupi \citep{2007AJ....133.2679H}. Even more dramatically, FU Ori-type stars undergo luminosity increases of $\sim10^{3}$ times, and can persist for 50--100+ years \citep{1996ARA&A..34..207H}. This paper is particularly relevant for these last two subtypes, since very large accretion rates should correspond to faster equilibrium spin rates. There is growing evidence that EXors are a distinct class (or alternately, evolutionary phase) of TTauri stars, so this phase may not generally last long enough to be relevant for long term spin rates. In contrast, the long quiescent time-scales conjectured for FU Ori stars ($10^3$--$10^4$~yrs) mean that most or all TTauri stars could pass through an extended FU Ori phase, which should then be reflected in the final spin rate. Comparing the estimated \teq\ for TTauri stars (see table \ref{tab:refcoords}) with the predicted FU Ori outburst cycles shows another important distinction between TTauri stars and magnetic accreting compact objects: the duration of an outburst cycle is a much larger fraction (up to 10\%) of the nominal equilibrium timescale (which as discussed could be much longer). As a result assuming spin equilibrium may not be valid. Are the results of this paper consistent with observations of the spin rates of TTauri stars? Assuming that most stars go through enough FU Ori outbursts to reach spin equilibrium, the answer is sensitive to how the spin rate of the star is regulated. As seen in Section \ref{sec:results}, when a simple `accretion/ejection' picture is assumed, the star tends to spin up to close to its outburst spin rate, rather than the long term averaged one. For FU Ori stars, assuming a duty cycle of between 0.1 and 1 per cent, \Mout $\sim 10^{-4}~\rm{M_{\odot}~yr^{-1}}$ versus \Mavg\ $\sim10^{-7}$--$10^{-6}~\rm{M_{\odot}~yr^{-1}}$. For a typical TTauri star, the accretion rate during outburst will be high enough to completely crush the magnetosphere, so that the disc accretes through a boundary layer directly on to the star. In standard accretion theory, the star should then spin up to close to its breakup frequency (although see discussion below). The high outburst accretion rate will also presumably inhibit a magnetically driven wind from the stellar surface, which will limit how efficiently a wind can regulate the star's spin, and likely not be able to prevent the star from spinning up. Na\"ively, one would then expect that TTauri stars in the FU Ori outburst stage should be spinning significantly faster than \peq\ estimated from observations, which is most likely $\dot{M}$ in `quiescence'. This does not immediately seem to be the case, although there may still be enough uncertainty in $B_*$ and the torque models that distinguishing between the two scenarios could be difficult. In contrast, a trapped disc spins down the star in the quiescent state, and over time will bring the star into spin equilibrium with its long term accretion rate. For a duty cycle of about 1 per cent, the accretion rate is still fairly high ($10^{-6}\rm{M_\odot~yr^{-1}}$) and corresponds to a faster spin than is observed (0.5--1 d). If the duty cycle is shorter, the mean accretion rate can be close to the quiescent one ($10^{-7}\rm{M_\odot~yr^{-1}}$), corresponding to a spin period of a few days, which is roughly consistent with observed spin periods. These conclusions are also challenged by observational evidence that suggests the magnetosphere \citep{2014MNRAS.437.3202J} and inner disc of young stars \citep{2007prpl.conf..507N} are located well within \rc. If these radius measurements are accurate it is somewhat surprising even within the `standard' steady-state accretion model, since it would suggest these stars are likely spinning up rapidly. It may indeed suggest enhanced spin down torque at relatively high accretion rates \citep{2013A&A...550A..99Z}. If most TTauri stars are FU Ors in quiescence, the problem is even larger: one would expect that the FU Ori events spin up the star even more, requiring even stronger spin down at lower accretion rates. There are several other possibilities for reconciling the high FU Ori accretion rates with relatively long spin periods. One is that the FU Ori phase of repeated outbursts may only occur for a subset of TTauri stars, or that this accretion phase does not last long enough to bring the star into spin equilibrium. This question can only be resolved observationally. A second possibility is that accretion through a boundary layer does not easily spin the star up to breakup. This has been suggested in boundary layer calculations by \cite{1996ApJ...467..749P} and more recently by new numerical and analytical work \citep{2013ApJ...770...67B}. In the latter papers, the authors find that angular momentum and energy in the boundary layers are mainly transported via acoustic waves rather than an `anomalous viscosity' as is typically assumed for both accretion discs and boundary layers. \cite{2013ApJ...770...67B} instead find angular momentum transport via waves can result in some outward transport (i.e. back into the disc), as well as into the deep layers of the star. Both these effects can limit how efficiently the star will spin up, although by how much is not yet quantified. However, without an additional very efficient and rapid source of angular momentum loss, the results here studying spin change in outbursts (both the expected final spin periods and the spin evolution time), combined with results suggesting most discs are truncated well within \rc, suggests that FU Ori phenomena are more likely a rare or brief evolutionary state, and most observed TTauri stars are not in the quiescent state of an FU Ori phase. Finally, the conclusions from this section are somewhat preliminary, since the models of spin evolution adopted in this paper do not consider the radial contraction of the protostar during its lifetime, which will make the star spin faster and hence require even more angular momentum loss. While this is straightforward to include, it is outside the scope of the current paper. \subsection{Conclusions} The results of this paper suggest that the long term spin evolution of magnetic stars can be significantly affected by large-scale changes in the mass accretion rate. In general, I find that by considering accretion outbursts, stars take significantly longer to reach their `equilibrium' spin period and that this spin period in general can be significantly different (generally shorter, but not always) than would be predicted from simple analytic arguments. The \peq\ and \teq\ are sensitive to the disc--field interactions, the outburst duration, and the transport mechanism that removes angular momentum from the star. In particular, the commonly envisioned scenario, in which gas either accretes on to the star or is expelled through a centrifugally launched wind, requires that the average accretion rate stay fairly steady in order to keep the star in near its predicted \peq. This is because the spin down mechanism is only efficient at relatively high $\dot{M}$ (when the inner disc remains close to \rc). Interestingly, this conclusion holds even for the more recent variants of this model, in which there is both accretion and ejection across a large range of $\dot{M}$ (section \ref{sec:transition}). Such a steady $\dot{M}$ is inconsistent with the most widely accepted `ionization instability' model for accretion outbursts, in which the accretion rate through the disc varies by several orders of magnitude between outburst and quiescence \citep{2001NewAR..45..449L}. If a stellar wind (launched from the stellar surface but driven in part by accretion power) can be launched, spin-down can remain efficient as long as the mass outflow rate is high enough ($\sim$10 per cent $\dot{M}$). There is some evidence supporting this idea for TTauri stars (e.g. \cite{2008ApJ...681..391M} and other works by those authors), but the idea remains somewhat schematic and controversial \citep{2011ApJ...727L..22Z}, and the outflow rate from the star itself is difficult to constrain observationally. The `trapped disc' model also has significant uncertainties, in particular the details of the coupling between the disc and the star, and the width of the coupled region (which sets the spin down efficiency), but has the distinction of being able to spin down the star very efficiently even at low $\dot{M}$. This could be very important in understanding the slow spin rates of Be X-ray binaries and possibly the long term spin rates of millisecond pulsars. In AMXPs, the large difference between outburst and quiescence means that accretion continues even when the cycle-averaged accretion rate is in the `propeller' regime. This affects the conclusions of \cite{2012Sci...335..561T}, in particular, the assertion that AMXPs can efficiently spin down via a propeller during a `Roche lobe decoupling phase' (where the mean accretion rate drops rapidly). Observations indicate that AMXPs in general are {\em not} in spin equilibrium with \Mavg. This could support the conclusion that AMXPs are not the progenitor systems for the entire class of radio millisecond pulsars \citep{2012arXiv1206.2727P} and therefore that their faster average spin periods do not indicate a general spin evolution from one population to the other; alternately it could suggest that a trapped disc remains around the star even as \Mavg\ drops and continues to spin the star down. Recent observations of transitional millisecond pulsar systems \citep{2016ApJ...830..122J}, however, suggest that the net spin down from an accretion flow is comparable to that from dipole radiation. In Be/X-ray binaries, considering the effects of outbursts changes the estimates of magnetic field (calculated assuming spin equilibrium) significantly. This conclusion applies even if an accretion/ejection model is considered, but it is especially true if a trapped disc remains present during quiescence. Considering these effects, the estimated magnetic field strengths for Be/X-ray binary systems (considering the large sample from the SMC, \citealt{2010ASPC..422..224C}) is significantly lower than estimated by \cite{2014MNRAS.437.3863K}, and in particular does not require in magnetar-strength magnetic fields except for the slowest spinning stars. It is not currently clear to what extent all protostars undergo repeated, large-scale outbursts, although at least a subset show large-scale variability. In these systems the magnetosphere is likely crushed by accretion during the outburst, so that the star should accrete via a boundary layer. The outcome of this scenario is not completely clear, but na\"ively one would expect that the final spin rate of the star would be dominated by what happens during outbursts \citep{1996ApJ...467..749P}. Observations of the innermost regions of TTauri stars suggest that the inner disc and closed magnetosphere are generally well within \rc, indicating that these stars are more likely spinning {\em up} than spinning down after a large FU Ori-level outburst. This fact, and the fact that observed spin rates are generally much slower than breakup could then imply that either the star accretes without spinning up efficiently during outburst, or that FU Ori-type outbursts are not a universal or long-lasting phase of star formation. The results of this paper are preliminary though, since they do not include the contraction (and necessary spin up) of the star as it evolves, nor spin regulation via boundary layer accretion. {\bf Note:} After this paper appeared on the arXiv, a similar work, focusing on transient accretion in AMXPs and considering only an `accretion/ejection' model was also published \citep{2017ApJ...835....4B}. The authors conclude based on their analysis that gravitational waves may be required to prevent MSPs from spinning to submillisecond periods during outburst. They broadly reach the same conclusion as for the `accretion/ejection' case considered here, namely, that stars should spin faster than predicted by the average accretion rate because a propeller outflow is generally inefficient, but do not find the same limit on spin period at the highest accretion rate (from limited spin-up efficiency because the source reaches $\dot{M}_{\rm Edd}$). Further investigation into what happens at high accretion rates is ongoing, and this will include a more detailed comparison with the results of that paper. | 16 | 9 | 1609.08654 |
1609 | 1609.01296_arXiv.txt | We compare 5 sub-grid models for supernova (SN) feedback in adaptive mesh refinement (AMR) simulations of isolated dwarf and L-star disc galaxies with $20-40$ pc resolution. The models are thermal dump, stochastic thermal, ``mechanical'' (injecting energy or momentum depending on the resolution), kinetic, and delayed cooling feedback. We focus on the ability of each model to suppress star formation and generate outflows. Our highest-resolution runs marginally resolve the adiabatic phase of the feedback events, which correspond to 40 SN explosions, and the first three models yield nearly identical results, possibly indicating that kinetic and delayed cooling feedback converge to wrong results. At lower resolution all models differ, with thermal dump feedback becoming inefficient. Thermal dump, stochastic, and mechanical feedback generate multiphase outflows with mass loading factors $\beta \ll 1$, which is much lower than observed. For the case of stochastic feedback we compare to published SPH simulations, and find much lower outflow rates. Kinetic feedback yields fast, hot outflows with $\beta\sim 1$, but only if the wind is in effect hydrodynamically decoupled from the disc by using a large bubble radius. Delayed cooling generates cold, dense and slow winds with $\beta> 1$, but large amounts of gas occupy regions of temperature-density space with short cooling times. We conclude that either our resolution is too low to warrant physically motivated models for SN feedback, that feedback mechanisms other than SNe are important, or that other aspects of galaxy evolution, such as star formation, require better treatment. | \label{Intro.sec} In our $\Lambda$CDM Universe, most of the mass is made up of dark matter. On large scales baryons trace the dark matter and its gravitational potential. Baryonic gas falls into galaxies at the centres of dark matter haloes, where it cools radiatively and collapses to form stars. By naive gravitational arguments, star formation should be a fast affair, consuming the gas over local free-fall times. However, from observations we know that it is a slow and inefficient process, taking $\sim20-100$ free-fall times, depending on the scale under consideration \citep[e.g.][]{Zuckerman1974, Krumholz2007, EvansNealJ2009}. Also, while observers have a notoriously hard time confirming the existence of gas flowing into galaxies \citep{Crighton2013}, they instead routinely detect oppositely directed \emph{outflows} at velocities of hundreds of km/s \citep[see review by][]{Veilleux2005}. To understand the non-linear problem of galaxy formation and evolution, theorists use cosmological simulations of dark matter, describing the flow and collapse of baryonic star-forming gas either with directly coupled hydrodynamics or semi-analytic models. Strong feedback in galaxies is a vital ingredient in any model of galaxy evolution, be it hydrodynamical or semi-analytic, that comes even close to reproducing basic observables, such as the star formation history of the Universe, the stellar mass function of galaxies, the Kennicutt-Schmidt relation, rotational velocities, and outflows \citep[e.g.][]{Vogelsberger2013, Dubois2014, Hopkins2014, Schaye2015, Wang2015, Somerville2015}. In order to capture the inefficient formation of stars, the first generation of galaxy evolution models included core collapse (or type II) supernova (SN) feedback, where massive stars ($\ga 8 \ \Msun$) end their short lives with explosions which inject mass, metals, and energy into the inter-stellar medium (ISM). In early hydrodynamical simulations, the time-integrated type II SN energy of a stellar population, $10^{51}$ ergs per SN event, was dumped thermally into the gas neighbouring the stellar population \citep{Katz1992}. However, such thermal dump feedback had little impact on star formation, resulting in an over-abundance of massive and compact galaxies. This so-called over-cooling problem is partly numerical in nature, and a result of low resolution both in time and space. As discussed by \cite{DallaVecchia2012}, the energy is injected into too large a gas mass, typically resulting in much lower temperatures than those at work in sub-pc scale SN remnants. The relatively high cooling rates at the typical initial temperatures attained in the remnant, of $10^5-10^6$ K, allow a large fraction of the injected energy to be radiated away before the gas reacts hydrodynamically, resulting in suppressed SN blasts and hence weak feedback. Gas cooling is, however, also a real and physical phenomenon, and while it is over-estimated in under-resolved simulations, a large fraction of the energy in SN remnants may in fact be radiated away instead of being converted into large-scale bulk motion \citep{Thornton1998}. A number of sub-resolution SN feedback models have been developed over the last two decades for cosmological simulations, with the primary motivation of reproducing large-scale observables, such as the galaxy mass function, by means of efficient feedback. The four main classes of these empirically motivated SN feedback models are i) kinetic feedback \citep{Navarro1993}, where a fraction of the SN energy is injected directly as momentum, often in combination with temporarily disabling hydrodynamical forces \citep{Springel2003}, ii) delayed cooling \citep[e.g.][]{Gerritsen1997, Stinson2006}, where radiative cooling is turned off for some time in the SN remnant, iii) stochastic feedback \citep{DallaVecchia2012}, where the SN energy is re-distributed in time and space into fewer but more energetic explosions, and iv) multiphase resolution elements that side-step unnatural `average' gas states at the resolution limit \citep{Springel2003, Keller2014}. In principle, a physically oriented approach to implementing SN feedback with sub-grid models is desirable. The goal is then to inject the SN blast as it would emerge on the smallest resolved scale, by making use of analytic models and/or high-resolution simulations that capture the adiabatic phase, radiative cooling, the momentum driven phase, and the interactions between different SN remnants. However, these base descriptions usually include simplified assumptions about the medium surrounding the SN remnant, and fail to capture the complex inhomogeneities that exist on unresolved scales and can have a large impact on cooling rates. In addition, even if the SN energy is injected more or less correctly at resolved scales, it will generally fail to evolve realistically thereafter because the multi-phase ISM of simulated galaxies is still at best marginally resolved. Hence there remains a large uncertainty in how efficiently the SN blast couples to the ISM. This translates into considerable freedom, which requires SN feedback models to be calibrated to reproduce a set of observations \citep[see discussion in][]{Schaye2015}. The most recent generation of cosmological simulations has been relatively successful in reproducing a variety of observations, in large part thanks to the development of subgrid models for efficient feedback and the ability to calibrate their parameters, as well as the inclusion of efficient active-galactic nucleus (AGN) feedback in high-mass galaxies. However, higher-resolution simulation works \citep[e.g.][]{Hopkins2012b, Agertz2013} suggest that SNe alone may not provide the strong feedback needed to produce the inefficient star formation we observe in the Universe. Attention has thus been turning towards complementary forms of stellar feedback, which provide additional support to the action of SNe. Possible additional feedback mechanisms include stellar winds \citep[e.g.][]{Dwarkadas2007, Rogers2013, Fierlinger2016}, radiation pressure (e.g. \citealt{Haehnelt1995, Thompson2005, Murray2010}, but see \citealt{Rosdahl2015}), and cosmic rays \citep[e.g.][]{Booth2013, Hanasz2013, Salem2014, Girichidis2016}. None the less, SN explosions remain a powerful source of energy and momentum in the ISM and a vital ingredient in galaxy evolution. For the foreseeable future a sub-resolution description of them will remain necessary in cosmological simulations and even in most feasible studies of isolated galaxies. The true efficiency of SN feedback is still not well known, and hence we do not know to what degree we need to improve our SN feedback sub-resolution models versus appealing to the aforementioned complementary physics. Rather than introducing a new or improved sub-resolution SN feedback model, the goal of this paper is to study existing models, using controlled and relatively inexpensive numerical experiments of isolated galaxy discs modelled with gravity and hydrodynamics in the Eulerian (i.e. grid-based) code \ramses{} \citep{Teyssier2002}. We use those simulations to assess each model's effectiveness in suppressing star formation and generating galactic winds, the main observational constraints we have on feedback in galaxies. We study five subgrid prescriptions for core-collapse SN feedback in isolated galaxy discs. We explore the `maximum' and `minimum' effects we can get from SN feedback using these models, and consider how they vary with galaxy mass, resolution, and feedback parameters where applicable. The simplest of those models is the `classic' \emph{thermal dump}, where the SN energy is simply injected into the local volume containing the stellar population. Three additional models we consider have been implemented and used previously in \ramses{}. These are, in chronological order, \emph{kinetic feedback}, described in \cite{Dubois2008} and used in the Horizon-AGN cosmological simulations \citep{Dubois2014}, \emph{delayed cooling}, described in \cite{Teyssier2013}, and \emph{mechanical feedback}, described in \cite{Kimm2014} and \cite{Kimm2015}. In addition, for this work we have implemented \emph{stochastic feedback} in \ramses{}, adapted from a previous implementation in the smoothed particle hydrodynamics (SPH) code \gadget{}, described in \citet[][henceforth \DS{}]{DallaVecchia2012}. \begin{table*} \centering \caption {Simulation initial conditions and parameters for the two disc galaxies modelled in this paper. The listed parameters are, from left to right: Galaxy acronym used throughout the paper, $\vcirc$: circular velocity at the virial radius, $\Rvir$: halo virial radius (defined as the radius within which the DM density is $200$ times the critical density at redshift zero), $\Lbox$: simulation box length, $\Mhalo$: DM halo mass, $\Mdisc$: disc galaxy mass in baryons (stars+gas), $\fgas$: disc gas fraction, $\Mbulge$: stellar bulge mass, $\Npart$: Number of DM/stellar particles, $\mstar$: mass of stellar particles formed during the simulations, $\Dxmax$: coarsest cell resolution, $\Dxmin$: finest cell resolution, $Z_{\rm disc}$: disc metallicity.} \label{sims.tbl} \begin{tabular}{l|rrrrrrrrrrrr} \toprule Galaxy & $\vcirc$ & $\Rvir$ & $\Lbox$ & $\Mhalo$ & $\Mdisc$ & $\fgas$ & $\Mbulge$ & $\Npart$ & $\mstar$ & $\Dxmax$& $\Dxmin$ & $Z_{\rm disc}$ \\ acronym & [$\kms$] & [kpc] & [kpc] & [$\Msun$] & [$\Msun$]& & [$\Msun$] & & [$\Msun$]& [kpc] & [pc] & [$\Zsun$] \\ \midrule \sbc & $65$ & $89$ & $300$ &$10^{11}$ &$3.5 \times 10^9$& $0.5$ &$3.5 \times 10^8$& $10^6$ & $2.0 \times 10^3$ & $2.3$ & $18$ & 0.1\\ \mw & $140$ & $192$ & $600$ & $10^{12}$ &$3.5 \times 10^{10}$ & $0.3$ & $3.5 \times 10^9$ & $10^6$ & $1.6 \times 10^4$ & $4.7$ & $36$ & $1.0$ \\ \bottomrule \end{tabular} \end{table*} The layout of this paper is as follows. First, we describe the setup of our isolated galaxy disc simulations in \Sec{simulations.sec}. We then describe the SN feedback models in \Sec{models.sec}. In \Sec{comparison.sec} we compare results for each of these models using their fiducial parameters in galaxy discs of two different masses, focusing on the suppression of star formation and the generation of outflows. In \Sec{res_conv.sec} we compare how these results converge with numerical resolution, both in terms of physical scale, i.e. minimum gas cell size, and also in terms of stellar particle mass. In Sections \ref{fb_stoch.sec} - \ref{fb_kin.sec} we take a closer look at the stochastic, delayed cooling, and kinetic feedback models respectively, and study how varying the free parameters in each model affects star formation, outflows and gas morphology. The reader can skip those sections or pick out those of interest, without straying from the thread of the paper. We discuss our results and implications in \Sec{Discussion.sec}, and, finally, we conclude in \Sec{Conclusions.sec}. | \label{Conclusions.sec} We used simulations of isolated galaxy discs with the \ramses{} code to assess sub-resolution models for SN feedback in AMR simulations, in particular their efficiency in suppressing star formation and generating outflows. We focused our analysis on a dwarf galaxy, ten times less massive than the MW, using a spatial resolution of $18$ pc and a stellar (DM) mass resolution of $2 \times 10^3 \ \Msun$ ($10^5 \ \Msun$), but also included a more limited analysis of a MW mass galaxy (using a resolution of $36$ pc, $1.6 \times 10^4 \ \Msun$ stellar particles and $10^6 \ \Msun$ DM particles). We studied five SN feedback models: i) thermal dump of SN energy into the host cell of the star particle, ii) stochastic thermal feedback, where the SN energy is re-distributed into fewer but more energetic explosions, iii) kinetic feedback, where momentum is deposited directly into a bubble around the star particle, iv) delayed cooling, where cooling is suppressed temporarily in the expanding SN remnant, and v) mechanical feedback, which injects energy or momentum depending on the resolution. Three of those models can be calibrated with adjustable parameters, which are the minimum local heating temperature for stochastic feedback, the bubble size and local mass loading for kinetic feedback, and the cooling suppression time for delayed cooling). The mechanical feedback model has no free parameters (once the SN energy has been decided) and the injected momentum is based on analytic derivations and high-resolution simulations of cooling losses in expanding SN blasts. We compared the results produced using these models with their fiducial settings, and for those models with adjustable parameters we studied the effects of parameter variations. Our main results are as follows. For our low-mass, high-resolution galaxy, thermal dump, stochastic, and mechanical feedback produce nearly identical results (Figs. \ref{SFR.fig} and \ref{OFtime.fig}). We showed that at our current resolution and star formation densities, stochastic feedback is actually not that stochastic, and mechanical feedback is still mostly in the adiabatic phase. Hence those feedback models are converged in that setup, and thermal dump feedback adequately resolves the energy injection (by multiple SNe in a single event). For our more massive galaxy, stochastic and mechanical feedback become significantly stronger than thermal dump feedback, but are still weak compared to delayed cooling and kinetic feedback (\Fig{SFR_MW.fig}). Strong outflows are not easily generated in our AMR simulations. Mass loading factors of unity or above require extreme measures, such as turning off cooling for a prolonged time, or kinetic feedback that is in effect hydrodynamically decoupled due to the bubble radius exceeding the disc height (\Fig{OFtime.fig}). The outflows produced by delayed cooling and kinetic feedback are very distinct, the former being cold, dense, and slow, while the latter are hot, diffuse, fast, and featureless (Figs. \ref{SQO_Fid.fig}, \ref{mapsOF_fid.fig}, and \ref{PH_Fid.fig}). The other models produce slow and remarkably similar outflows at intermediate densities and temperatures. Save for thermal dump feedback, all models do well in terms of resolution convergence when considering SFRs, while, with the exception of kinetic feedback, they produce significantly higher outflow rates at lower resolution (\Fig{SFR_res.fig}). Although a direct comparison is difficult, stochastic feedback appears to produce much weaker outflows than in the similar disc runs with the original SPH version of the model of \DS{}. This discrepancy is perhaps a result of subtle setup differences between our discs and those of \DS{}, but we cannot rule out a more fundamental AMR versus SPH difference. Stochastic feedback does become efficient at generating massive outflows in our AMR discs if we use very high values for the stochastic heating temperature (up to $10^9$ K), but this comes at the cost of strong stochasticity due to low SN probabilities (Figs. \ref{SFR_stoch.fig} and \ref{maps_stoch.fig}). The major handle on the generation of outflows appears to be how well the SN feedback model circumvents gas cooling, directly or indirectly. Delayed cooling is the only model which succeeds at generating outflows with mass loading factors exceeding unity (\Fig{OFtime.fig}), at reproducing the observed main sequence SFRs (Figs. \ref{SFR.fig} and \ref{SFR_MW.fig}), and the Kennicutt-Schmidt relation (with appropriate calibration; \Fig{KS_Fid.fig}). The other models fail to produce SN feedback strong enough to reproduce these observations. This is discouraging, as delayed cooling retains too much energy for too long, which in reality is partly lost to radiative cooling, while the other feedback models are arguably more physically motivated. Moreover, for the low-mass galaxy we argued that thermal dump, stochastic, and mechanical feedback converge because we resolve the adiabatic phase of the feedback events. This implies that in this case delayed cooling and kinetic feedback yield incorrect answers. In particular, delayed cooling results in gas occupying regions of temperature-density space where the cooling time is very short, which compromises predictions for observational diagnostics. Possible reasons for the disconnect between observations and our results are: i) a lack of additional feedback physics, such as radiation feedback or cosmic rays, ii) an incomplete setup, i.e. an insufficiently realistic description of galaxies, iii) other aspects of the subgrid physics, such as star formation, are unrealistic \citep[e.g.][]{Agertz2015, Semenov2016}, iv) overcooling on galactic scales is still an issue at our resolution, even if different feedback models converge to the same results, and a significantly higher resolution is required. The current analysis will serve as a foundation for future studies of feedback in galaxies, where we will use a sub-set of these models to study the interplay of SN feedback with different sub-grid methods for star formation and with feedback in the form of stellar radiation. | 16 | 9 | 1609.01296 |
1609 | 1609.01069_arXiv.txt | {Recent HESS observations of the $\sim200$~pc scale diffuse gamma-ray emission from the central molecular zone (CMZ) suggest the presence of a PeV cosmic-ray accelerator (PeVatron) located in the inner $10$~pc region of the Galactic Center. Interestingly, the gamma-ray spectrum of the point-like source (HESS~J1745-290) in the Galactic Center shows a cutoff at $\sim 10$~TeV, implying a cutoff around $100$~TeV in the cosmic-ray proton spectrum. Here we propose that the gamma-ray emission from the inner and the outer regions may be explained self-consistently by run-away protons from a single, yet fading accelerator. In this model, gamma rays from the CMZ region are produced by protons injected in the past, while gamma rays from the inner region are produced by protons injected more recently. We suggest that the blast wave formed in a tidal disruption event (TDE) caused by the supermassive black hole (Sgr A*) could serve as such a fading accelerator. With typical parameters of the TDE blast wave, gamma-ray spectra of both the CMZ region and HESS~J1745-290 can be reproduced simultaneously. Meanwhile, we find that the cosmic-ray energy density profile in the CMZ region may also be reproduced in the fading accelerator model when appropriate combinations of the particle injection history and the diffusion coefficient of cosmic rays are adopted. } | Recently, the HESS collaboration reported the deep gamma-ray observations with arcminute angular resolution of the central molecular zone (CMZ) surrounding the center of our Galaxy \citep{HESS_GC16}, extending out to $r\sim 250$~pc and $\sim 150$~pc at positive and negative Galactic longitudes respectively. The brightness distribution of very high energy (VHE) gamma rays shows a strong correlation with the locations of massive gas-rich complexes. This discovery, combined with the fact that in the leptonic scenario severe radiative losses would be expected for multi-TeV electrons in the Galactic Center, points to a hadronic origin of the diffuse VHE emission. The inferred radial profile of $>10$~TeV cosmic-ray (CR) energy density is consistent with $1/r$ dependence in the entire CMZ region, and it was argued for a centrally located source in the inner $\sim 10$~pc region with a quasi-constant injection operating over at least $1000$~years, given the diffusive propagation of cosmic rays \citep{HESS_GC16}. The best-fit gamma-ray spectrum shows a single power-law form with a photon index of $\sim 2.3$ extending to energies larger than tens of TeV without a break or a cutoff, which indicates the existence of a PeV proton accelerator, or the so-called "PeVatron". Interestingly, the location of the PeVatron appears to coincide with the central point-like gamma-ray source HESS~J1745-290, which shows a harder spectrum with a photon index of $\sim 2.1$ and, however, a clear cutoff at $\sim 10$~TeV \citep{HESS_GC09}. The GC has long been suggested as a cosmic-ray proton accelerator \citep[e.g.][]{Ptusin81, Said81, Aharonian05b, Liu06, Fujita16} and a hadronic origin was also suggested for the gamma-ray emission of HESS~J1745-290 \citep{Aharonian05b, Chernyakova11}, while the leptonic origin may work well too \citep{Aharonian05a, Hinton07}. There is no obvious connection between HESS~J1745-290 and the gamma rays from the outer CMZ region. However, as protons accelerated at the center source will produce gamma rays via the proton--proton collision with the gas along the path to the outer CMZ region, they may potentially also explain the gamma-ray emission from HESS~J1745-290 \citep{Aharonian05b, HESS_GC16}. In this scenario, some additional mechanisms must be introduced to reconcile the 10\,TeV cut--off in the spectrum of HESS~J1745-290 with the extension of the proton spectrum to PeV energies in the outer CMZ region. One possible solution is invoking the absorption due to a dense infrared photon field near the GC \citep{HESS_GC16}. However, based on the current infrared data of the GC, the attenuation of the multi TeV photon flux originating from the GC is probably not strong enough to cause a cutoff \citep{Moskalenko06, Zhang06, HESS_GC09, Celli16}, though an extremely clumpy distribution of the infrared photons might overcome this difficulty \citep{Guo16}. In this work, we propose a fading proton accelerator located at the inner $10$~pc region as the origin of the gamma-ray emission from both HESS~J1745-290 and the CMZ. In this model, the source was more powerful in the past and injected protons beyond PeV energies. These protons have propagated to the outer CMZ region, where, at the present time, they produce the gamma-ray emission with the spectrum extending to high energies. On the other hand, as the source power gradually fades and the maximum energy of the injected protons declines, the gamma-ray emission from the inner region, which arises from recently injected protons, develops a cutoff in the spectrum, which decreases towards the Galactic Center. We will show that this model explains simultaneously the gamma-ray spectra of the CMZ region and the inner point-like source. As the particle injection rate of a fading accelerator decreases with time as well, it is apparently inconsistent with the suggestion by \citet{HESS_GC16} that the CR energy density profile indicates a constant injection rate assuming a spatially independent diffusion coefficient. So we need to check whether the measured CR density profile can be reproduced in the fading accelerator model. The rest part of this paper is organized as follows. We fit the gamma-ray spectral data of both the inner and outer region simultaneously in a specific scenario of fading accelerator, i.e., a tidal disruption event by the supermassive black hole in the Galactic Center drives a blast wave which accelerates cosmic rays (\S 2). In \S 3, we study the energy density profile of cosmic rays in the general fading accelerator model, paying particular attention to the influence of the particle injection history and the diffusion coefficient. The discussion and conclusion are given in \S 4. | In this work, we proposed a fading accelerator located in the inner $10$~pc of the GC as the source for the gamma-ray emission of both the point-like source HESS~J1745-290 and the CMZ region. Given the decreasing maximum acceleration energy in the injected proton spectrum, gamma rays of the inner region presents a cutoff around $10$~TeV as they are produced via $pp-$collision by more recently injected protons that have a lower maximum energy, while the gamma rays of the outer CMZ region arise from earlier injected protons that have a larger maximum energy and hence produce a spectrum extending to energies beyond tens of TeV. The fading accelerator could be due to some past activity of the supermassive black hole, such as a blast wave driven by a TDE {due to the supermassive black hole in the GC}. Following the Sedov solution of the blast wave, we obtained the evolution of the particle injection rates and maximum acceleration energy in such a scenario. A simultaneous fit to the spectra of both HESS~J1745-290 and the CMZ region is obtained within $1\sigma$ deviation from the HESS data. We find that the measured CR energy density profile can also be reproduced with the same parameters. Furthermore, we performed a general study of the effects of the CR injection history and the diffusion coefficient on the resultant energy density profile and found that the fading accelerator can reproduce the measured data for different injection times. It may be worth noting the possibility that gamma rays in the inner region may have a different origin from that in the outer CMZ region. {In this case, the fading accelerator model can still account for the gamma-ray emission in the CMZ region and the fit to the data could be improved as there is no constraints from the inner source anymore}. We may adopt a larger kinetic energy and/or a higher acceleration efficiency to get a higher $E_{\rm max}$, so that the calculated spectrum in the CMZ region can extend up to tens of TeV without softening. In the meanwhile, we need to adopt a lower hydrogen gas density for the inner region in order not to overshoot the observed gamma-ray flux of HESS~J1745-290. As we already mentioned in the last section, to obtain a good fit to the CR energy density profile, the favored diffusion coefficient in the CMZ region is about one order of magnitude larger than that in the Galactic disk. Such a large diffusion coefficient would lead to a purely rectilinear propagation regime of CRs inside the inner $\ll D/c \sim 30$\,pc, which also implies that CRs are still not fully isotropized even at tens of pc \citep{Prosekin15}. The gamma-ray fluxes from such regions are enhanced (reduced) if the most CRs are moving towards (away from) our line of sight. This is because the emitted gamma--rays will concentrate on the moving direction of their ultra-relativistic parent protons (i.e., the beaming effect). As a result, we may underestimate the gamma-ray flux from the HESS~J1745-290 while overestimate the gamma-ray flux from the CMZ region, as we have assumed CRs are isotropized in our calculation. However, we note that the theoretical flux of HESS~J1745-290 is barely affected by this effect. This is because, given a much higher gas density in the inner 15\,pc region than the outer region, the flux of HESS~J1745-290 should be dominated by the emission from the inner 15\,pc region, which has spherical symmetry in our model. So the enhanced flux is exactly balanced by the shrunk observable region. The extent of the overestimation on the flux of the CMZ region depends on how far the real distribution of CRs deviate from the isotropic one. Since the pitch angle distribution of CRs becomes broader as they propagate, we may expect a typical angular size of a bundle of CRs after propagation a distance of $r$ as \citep{Prosekin15} \begin{equation} \theta=\sqrt{\frac{2r}{3D/c}}. \end{equation} Substituting the obtained parameters into the above equation, we find that the angular size of 100\,TeV protons is about { $30^\circ$ at 20\,pc and $50^\circ$ at 70\,pc}, implying a considerable fraction of CRs that initially propagate away from us are already deflected to our direction. Besides, CRs are already isotropized at larger radius. So we may expect that the overestimation of the flux from the CMZ region is not severe and can be modulated by tuning the gas density in the CMZ region. To simplify the calculation, several assumptions have been made in our calculations, such as a uniform gas density, a spatially independent and isotropic diffusion coefficient, a temporally independent particle acceleration and injection efficiency, and etc. We note that the large diffusion coefficient may be avoided if more complicated factors are taken into account. For example, if the diffusion coefficient is spatially dependent, such as increasing outwardly, a smaller diffusion coefficient might also work. This is due to that a faster diffusion (i.e., larger $D$) leads to a smaller density, so an outwardly increasing diffusion coefficient could yield a steeper profile of CR energy density compared with {that in the case of} a spatially independent diffusion coefficient, and hence fit the data with a smaller diffusion coefficient at the GC. Besides, if we consider the capability of { the confinement of CRs by the blast wave} decreases with time (i.e., $f_{\rm CR}$ increases with time), the injection rate would decrease less rapidly with time, and then a smaller diffusion coefficient could also work { well} in fitting the CR density profile. More realistic modeling using complicated numerical calculations/simulations on the relevant processes would be useful for a more careful study. We leave this study to a future publication. In the present model, cosmic rays are injected by a fading accelerator, probably arising from { some types of explosion events}, such as a { TDE}. Provided that the event rate of the explosion is $10^{-5}-10^{-4}\,{\rm yr^{-1}}$, such a fading accelerator would occur repeatedly on the timescale of $>10^{5}$yr. In this sense, the particle injection rate appears to be quasi--constant on such long timescales. {Indeed, there may be} no strict constant injection in nature. Every source {may be} variable at different time scales. We note that the measured gamma-ray emission in the CMZ region probes the CR injection on a time scale of $10^3-10^4$\,yr or longer, while the gamma-ray from the inner 10\,pc mainly reflects the CR injection in recent $10-100$\,yr. Thus, one alternative scenario is that the source (quasi-)constantly injected CRs (i.e., $\alpha=0$) with a spectrum extending up to $>$PeV during the past ten thousand years or longer while the source is variable on a short timescale of $10-100\,$yr. Coincidentally, it is just in a less active period during the recent ten years, resulting in a low particle acceleration efficiency and, subsequently, the cutoff in the measured spectrum of the inner region. A further study on the CR acceleration and injection is required to reveal more details in this scenario, which, however, is beyond the scope of this work. | 16 | 9 | 1609.01069 |
1609 | 1609.02156_arXiv.txt | From a sample of spectra of 439 white dwarfs (WDs) from the ESO-VLT Supernova-Ia Progenitor surveY (SPY), we measure the maximal changes in radial-velocity (\drvm) between epochs (generally two epochs, separated by up to 470\,d), and model the observed \drvm\ statistics via Monte-Carlo simulations, to constrain the population characteristics of double WDs (DWDs). The DWD fraction among WDs is \fb$=0.100 \pm 0.020$ (1$\sigma$, random) $+ 0.02$~(systematic), in the separation range $\lesssim 4$\,AU within which the data are sensitive to binarity. Assuming the distribution of binary separation, $a$, is a power-law, $dN/da\propto a^\alpha$, at the end of the last common-envelope phase and the start of solely gravitational-wave-driven binary evolution, the constraint by the data is $\alpha=-1.3 \pm 0.2$ ($1\sigma$) $\pm 0.2$~(systematic). If these parameters extend to small separations, the implied Galactic WD merger rate per unit stellar mass is $R_{\rm merge}=\left(1-80\right)\times 10^{-13}$\,yr$^{-1}\,M_\odot^{-1}$ ($2\sigma$), with a likelihood-weighted mean of $R_{\rm merge}=(7 \pm 2) \times 10^{-13}$\,yr$^{-1}\,M_\odot^{-1}$ ($1\sigma$). The Milky Way's specific Type-Ia supernova (SN Ia) rate is likely $R_{\rm Ia}\approx 1.1\times 10^{-13}$\,yr$^{-1}\,M_\odot^{-1}$ and therefore, in terms of rates, a possibly small fraction of all merging DWDs (e.g. those with massive-enough primary WDs) could suffice to produce most or all SNe Ia. | \label{sec:Intro} A large fraction of all stars, and a majority of intermediate-mass and massive stars, are in multiple systems. Multiplicity is an outcome of star formation and early stellar evolution, and thus serves as a probe of those poorly understood processes. Furthermore, binary, triple, and higher-order systems provide the settings for a rich variety of astrophysical phenomena, including interacting, accreting, and merging binaries, various types of supernovae, and gravitational wave sources. However, the demographics of stellar multiplicity are still poorly known, i.e. the distribution of multiplicity index (single, binary, triple...), separation, component mass ratio, and eccentricity, all as a function of stellar mass, age, metallicity, and Galactic environment \citep[e.g.,][]{Duquennoy_1991,Raghavan_2010}. These demographics must be physically linked, at some level, to those of sub-stellar companions -- brown dwarfs (BDs) and planets -- for which our knowledge is even sketchier, and also to those of stellar remnants -- white dwarfs (WDs), neutron stars, and black holes. Binarity in WDs is particularly interesting. WDs are the end state of 95\% of all stars, and they are the current state of the majority of all stars ever formed with mass above $1.2$\,\msun. As such, binary WDs provide a fossil probe of the initial binary populations and of their subsequent binary evolution. Systems consisting of close double WDs (DWDs) are potential progenitors of Type-Ia supernovae \citep[SNe Ia; e.g.][]{Maoz_2014}, AM Canum Venaticorum systems (a WD accreting from another degenerate or semi-degenerate companion star), and R Corona Borealis stars \citep[highly magnetic WDs postulated to result from WD mergers; e.g.][]{Longland_2011}. DWDs will be the main foreground of space-based gravitational-wave detectors such as {\it LISA}, both as resolved sources at higher gravitational-wave frequencies, and as an unresolved continuum at lower frequencies. Identifying the individual nearby DWD systems and measuring the binary parameter distribution as a whole for DWDs is therefore important for the budding field of gravitational-wave astronomy. Like other double-compact-remnant binaries (including neutron stars and black holes), DWDs are physically simple and ``clean'' systems in which the evolution of each WD is decoupled from the other WD and driven mainly by cooling via thermal emission from the surface, while the binary evolution is dictated solely by gravitational wave emission (except in the very final merger phases). Systematic searches for DWDs began in the 1980s \citep[see][and references therein]{Napiwotzki_2004}. There are now over 90 individual close DWD systems for which orbital parameters have been derived \citep[see, e.g.,][and references therein]{Nelemans_2005, Marsh_2011, Debes_2015, Hallakoun_2016, Brown_2016}. Excluding systems with extremely low-mass (ELM) WDs of $\sim 0.2$\,\msun, which are found to be always in binaries, there are about 30 DWD systems with orbital parameters. The statistics of the short-orbit DWD population as a whole were examined by \citet{Maxted_1999}, who studied a sample of 46 WDs and estimated a binary fraction between 1.7\% and 19\%. More recently, \citet[][hereafter M12]{Maoz_2012} developed a statistical method to characterize the DWD population in a sample of WDs, by measuring the distribution of \drvm, the maximum radial velocity (RV) shift between several epochs (two or more) of the same WD. \citetalias{Maoz_2012} showed that even with just two epochs per WD and with noisy RV measurements, the \drvm\ distribution can set meaningful constraints on the binary fraction of the population, \fb, and on the distribution of binary separations, $dN/da$. The \drvm\ method applied to such data permits also an estimate of the merger rate of the DWD population. Statistical inference about the DWD population is thus possible without follow-up observations and full binary parameter solutions for candidates. \citet[][hereafter BM12]{Badenes_2012} measured few-epoch RVs in the spectra of $\sim 4000$ WDs from the Sloan Digital Sky Survey (SDSS) and applied the method to the observed \drvm\ distribution. The data constrained \fb\ at separations $a<0.05$\,AU to a $1\sigma$ range of $3-20\%$. Assuming a power-law separation distribution, $dN/da\propto a^\alpha$, at the time of DWD formation (from hence the binary separation evolves solely via gravitational-wave emission), \citetalias{Maoz_2012} and \citetalias{Badenes_2012} showed that $\alpha$ is constrained to the range $-2$ to $+1$, with strong covariance between \fb\ and $\alpha$ (low \fb\ together with a steep negative power-law slope $\alpha$, or a higher \fb\ together with a shallower $\alpha$, can both populate the small separation range with DWD systems and produce the high-\drvm\ tail in the observed distribution). As every combination of \fb\ and $\alpha$ translates to a WD merger rate, \citetalias{Badenes_2012} showed that the WD merger rate per unit stellar mass in the Milky Way is constrained to $R_{\rm merge}=1.4^{+3.4}_{-1}\times 10^{-13}{\rm yr}^{-1}\,M_\odot^{-1}$, a range that straddles the Galactic SN Ia rate per unit stellar mass, $R_{\rm Ia}\approx 1.1\times 10^{-13}{\rm yr}^{-1}\,M_\odot^{-1}$. (The Milky Way's specific SN Ia rate can be reliably estimated from its approximate mass and from the fact that it is an Sbc galaxy; see \citetalias{Badenes_2012}). The SDSS sample of WDs analysed by \citetalias{Maoz_2012} and \citetalias{Badenes_2012}, while large, suffered from a low RV precision of $\sim 80$\kms, a result mainly of the low resolution of the SDSS spectra and of the fact that most WDs have only a few, highly Stark-broadened hydrogen Balmer absorption lines in their spectra. As a consequence, only the systems in the sample with observed \drvm$\gtrsim 250$\kms (some 15 in number) drove the statistical conclusions. This lower limit in the significantly detectable \drvm\ translated, for typical WD masses, to an upper limit in the binary separation that is probed by the study, of $a\sim 0.05$\,AU. Furthermore, the small number of systems driving the signal results in large statistical errors and in the strong degeneracy between the model parameters. In the present paper, we apply the method of \citetalias{Maoz_2012} and \citetalias{Badenes_2012} to another sample of multi-epoch WD spectra, from the European Southern Observatory (ESO), 8\,m Very Large Telescope (VLT), Supernova-Ia Progenitor surveY \citep[SPY;][]{Napiwotzki_2001}. SPY was a few-epoch spectroscopic survey of $\sim$800 bright ($V\sim 16$\,mag) WDs, with the objective of using RV differences between epochs to identify close DWD systems that will merge within a Hubble time, thus being potential SN Ia progenitors. Published results from SPY relevant to DWDs include \citet{Napiwotzki_2002} (discovery of a DWD with a mass close to the Chandrasekhar limit); \citet{Karl_2003a,Karl_2003b} (follow-up analysis of several DWDs); \citet{Nelemans_2005} (follow-up analysis of five DWDs from SPY); and \citet{Koester_2009} (catalogue and spectroscopic analysis of hydrogen-dominated WDs, including a list of DWDs). A statistical analysis of the SPY dataset as a whole, and its implications for the binary WD population, has not been published to date. Out of the full SPY dataset, we select about 500 WDs suitable for our present analysis. Although the sample size is an order of magnitude smaller than the SDSS WD sample of \citetalias{Badenes_2012}, the high spectral resolution and signal-to-noise ratio (S/N) possible with the VLT permit resolving the narrow non-local-thermodynamic equilibrium (NLTE) core of the H$\alpha$ line (and sometimes H$\beta$) that exists in the spectra of DA-type WDs (i.e. WDs with only hydrogen lines in their optical spectra, which constitute the majority of WDs). This provides a typical RV resolution of $1-2$\kms\ per epoch, a factor $\sim 50$ times better than for the SDSS sample. This RV resolution, combined with the distribution of time separations between epochs in SPY, means that the SPY sample is sensitive to DWDs out to separations $a\sim 4$\,AU (see Section~\ref{sec:monte-carlo-simul}, below). While, in principle, the lowest-separation/highest-RV systems are also detectable in SPY, the small sample size of SPY makes it unlikely to ``catch'' those systems, and in fact the largest \drvm\ that we measure is $240$\kms. The SPY sample thus nicely complements the SDSS sample, in so much as it probes the WD population's binarity in the $a=0.05-4$\,AU interval range, compared to the $a=0.001-0.05$\,AU range probed by SDSS (the lower limit in SDSS arising from the exposure length of $\sim 15$\,min, which prevents the detection of RV variations in systems with orbital periods comparable to this time.) The logarithmic interval in separation probed, in principle, by SPY, $a=0.001-4$\,AU, is 2.1 times larger than the $a=0.001-0.05$\,AU logarithmic interval of SDSS. Therefore, for example, for a separation distribution that has equal numbers of binaries per logarithmic interval, one would expect to find a binarity fraction about twice as high in SPY as in SDSS. From analysis of the SPY sample, below, we find values of \fb, $\alpha$, and the merger rate of the WD population, that are consistent with the findings of \citetalias{Badenes_2012} for the SDSS sample, but are now more tightly constrained. The allowed values of $\alpha$ and $R_{\rm merge}$ strengthen the case for the ``double-degenerate'' progenitor scenario of SNe Ia. | We have measured and analysed the distribution of maximum radial velocity differences between observing epochs, \drvm, for a sample of 439 DA-type WDs from the SPY program, and have modelled the \drvm\ distribution to set constraints on the properties of the DWD population. Assuming that every generation of the DWD population, when it emerges from its last common-envelope phase, has an initial separation distribution that can be represented by a power law over the $a<4$\,AU separation range probed by these data, then the fraction of all WDs that have companion WDs in this separation range is \fb=$0.10 \pm 0.020 (1\sigma) +0.02$ (systematic), and the power-law index of the separation distribution is $\alpha=-1.3 \pm 0.30 (1\sigma)\pm 0.2$ (systematic). Combined with current estimates of the local WD space density and the local stellar mass density, these parameters imply a gravitational-wave-loss-driven specific Milky Way WD merger rate of $1\times 10^{-13}$ to $8\times 10^{-12}\,{\rm yr}^{-1}$\,\msun$^{-1}$ ($2\sigma$ range). This is between 1 to 70 times the estimated Milky-Way SN~Ia rate per unit mass. If some fraction (perhaps as small as a few per cent) of DWD mergers can produce a normal SN~Ia explosion, our results imply that there is no shortage of the progenitor DWD population for this explosion scenario. Our results indicate about a 3\% DWD binary fraction per decade in separation among WDs. \citet{Klein_2017} have recently estimated the binary fraction among DWD stellar progenitors, where both progenitors have $>1$\,\msun, and found a 4\% fraction per decade in period, corresponding to 6\% per decade in separation, with some overlap between the separation ranges probed by their analysis and ours. In a Galactic population of WDs in binaries, for roughly half of the systems with WD companions having initial mass $>1$\,\msun, the companions will not yet have evolved into a WD, and therefore the companions will completely dominate the light, preventing the inclusion of such systems in WD samples. The 3\% fraction of DWDs per decade in separation that we see in SPY is therefore nicely consistent with the 6\% per decade found by \citet{Klein_2017} for DWD progenitors. The high-\drvm\ tail of the distribution identifies 27 very likely DWD systems (with \drvm>15\kms; in two cases the WD companions may be a BD and a cool M star) and a further 16 possible DWD systems (with 10\kms< \drvm < 15\kms). The \drvm\ distribution does not constrain the component masses of the DWDs, and thus these DWDs merit follow-up observations to confirm their nature and to derive their orbital parameters and component masses. From photospheric modelling, the masses of the photometric primaries in the DWDs tend to be of somewhat lower mass than typical single WDs, with a broad distribution centred around $\sim 0.5$\,\msun. The minority of these DWDs that have additional data in the literature generally have, for the unseen photo-secondary WDs, masses similar to, or somewhat larger than, the photo-primary. The ``double-degenerate'' scenario for SNe Ia, invoking DWD mergers as the progenitors of SNe~Ia, has been traditionally criticised on two main grounds \citep[see][]{Maoz_2014}. The numbers and hence the merger rate of the progenitor DWD populations was thought to be too small to match the SN~Ia rate, particularly if a total merged mass above the Chandrasekhar mass is required for an explosion, as often assumed in this scenario. The second long-standing problem has been theoretical -- the tidal disruption of the secondary-mass WD by the primary, and its gradual accretion onto the primary WD through a disk or a spherical configuration was thought to lead to either a stable, just more-massive, merged WD or, in the case of an above-Chandrasekhar final mass, to an ``accretion-induced collapse'' to a neutron star and an electron-capture supernova explosion. (However, the accretion-induced collapse outcome has emerged from one-dimensional calculations, which could change in 3D treatments, that are yet to be performed.) Our results for the DWD population are germane to both of these objections to the double-degenerate SN~Ia scenario. It is possible that follow-up observations of the SPY DWDs will reveal, in analogy to what has been found for ELM WDs, that their WD companions have a broad mass distribution, with a significant fraction at masses above 0.9 or 1\,\msun. If so, and given that our results allow for up to a factor-70 surplus in the total rate of DWD mergers, there could conceivably be a sufficient number of mergers of CO+CO WDs with above-Chandrasekhar merged masses. Alternatively, recent violent-merger models \citep[e.g.][]{Pakmor_2012} find that an above-Chandrasekhar total mass is not required for reproducing a normal SN~Ia explosion, but rather only a primary mass above 0.9 or 1\,\msun. The secondary, which could even be a low-mass He WD, serves only as a ``hammer'' that sets off the detonation in the primary. If this were true, then an even-larger fraction of the total DWD merger rate could lead to a SN~Ia explosion. A remaining problem for violent mergers is the asymmetry of the explosion predicted by current models, and manifested in the expected strong (but unobserved in practice) polarisation of the light from SNe~Ia \citep{Bulla_2016}. On the observational side, progress is achievable from follow-up observations and detailed characterisation of the individual DWD systems. Upcoming large and complete WD and DWD samples from the {\it Gaia} Mission will bring into much better focus the DWD population and its merger rate. In parallel, continued improvements in the theoretical study of WD mergers should further clarify if and under what conditions mergers can lead to SN~Ia explosions. | 16 | 9 | 1609.02156 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.