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
|
|---|---|---|---|---|---|---|---|
0710 | 0710.4922_arXiv.txt | Using the high-resolution spectrometer SPI on board the \emph{International Gamma-Ray Astrophysics Laboratory (INTEGRAL)}, we search for a spectral line produced by a dark matter (DM) particle with a mass in the range $40 keV < M_{DM} < 14 MeV$, decaying in the DM halo of the Milky Way. To distinguish the DM decay line from numerous instrumental lines found in the SPI background spectrum, we study the dependence of the intensity of the line signal on the offset of the SPI pointing from the direction toward the Galactic Centre. After a critical analysis of the uncertainties of the DM density profile in the inner Galaxy, we find that the intensity of the DM decay line should decrease by at least a factor of 3 when the offset from the Galactic Centre increases from $0^\circ$ to $180^\circ$. We find that such a pronounced variation of the line flux across the sky is not observed for any line, detected with a significance higher than $3\sigma$ in the SPI background spectrum. Possible DM decay origin is not ruled out only for the unidentified spectral lines, having low ($\sim 3\sigma$) significance or coinciding in position with the instrumental ones. In the energy interval from 20 keV to 7 MeV, we derive restrictions on the DM decay line flux, implied by the (non-)detection of the DM decay line. For a particular DM candidate, the sterile neutrino of mass $M_{DM}$, we derive a bound on the mixing angle. | \subsubsection*{Dark matter in the Universe} There is a vast body of evidence, suggesting that the large fraction of matter in the Universe exists in the form of the \emph{Dark matter} \emph{(DM)}. However, while the total density of the DM is measured with a very high precision ($\Omega_{\rm DM}h^2 = 0.105^{+0.007}_{-0.009}$,~\citealt{WMAP3}), little is known about its properties apart from this. The possibility that the DM is composed of the Standard Model (SM) particles has been ruled out for a long time already. Indeed, the DM cannot be made out of baryons, as producing such an amount of baryonic matter would require drastic modifications of the scenario of the Big Bang nucleosynthesis (BBN), which otherwise successfully describes the abundance of light elements~(see for example ~\citealt{Dar:95}). Recent microlensing experiments rule out the possibility that another type of baryonic DM -- massive compact halo objects (MACHOs) -- constitute dominant fraction of mass in the halo~\citep{MACHO:00,EROS:00,OGLE:98}. The only non-baryonic DM candidate in the SM candidates -- (left-handed) neutrino -- is ruled out from the large scale structure (LSS) considerations~\citep[see e.g.][]{Bond:80,Hannestad:03,Crotty:04}. What are the properties of a successful DM candidate? First of all, this particle should be massive. Many extensions of the SM present the DM candidates with the masses ranging from $\sim 10^{-10}\ev$~(massive gravitons, \citealt{Dubovsky:04}) and $\sim 10^{-6}$~eV (axions) to hundreds of GeV (WIMPs) and even to $10^{13}$~GeV \citep[WIMPZILLA,][]{Kuzmin:98,Kuzmin:99,Chung:98}. For a review of particle physics DM candidates see e.g.~\cite{Bergstrom:00,Bertone:04,Carr:06}. Secondly, there should exist mechanisms of DM production with the correct abundances. The production mechanism in particular determines the velocity distribution of particles in the early Universe. This velocity distribution can, in principle, be probed experimentally. Namely, if during the structure formation epoch the DM particles have velocities, comparable to the speed of sound in the baryon-photon plasma, they ``erase'' density fluctuations at scales, smaller than the distance, they have traveled (called the \emph{free-streaming length}). To differentiate various models in accordance with this property, the DM candidates with the negligible velocity dispersion (and, correspondingly, free-streaming) are called \emph{cold} DM (CDM), while those with the free-streaming of the order of $\sim 1\mpc$ are considered to be \emph{warm} (WDM).\footnote{The left-handed neutrino would represent \emph{hot} DM in this terminology, i.e. the DM with the free-streaming length $\gg 1$~Mpc.} It is possible to constrain the free-streaming length of a particular DM candidate by probing the structure of the Universe at galaxy-size scales. This can be done through the analysis of the Lyman-$\alpha$ forest data~\citep{Hui:97}. Lyman-$\alpha$ analysis puts an upper bound on the free-streaming of the DM particles~\citep{Hansen:01,Viel:05,Seljak:06,Viel:06,Viel:07}. It should be noted however that currently existing interpretation of the Lyman-$\alpha$ data is model-dependent, as, apart from a number of astrophysical assumptions (see~\citealt{Hui:97}) and complicated hydrodynamic simulations, it relies on {\it a priori} assumptions about the velocity distribution of the DM particles. A way to differentiate between CDM and WDM models would be to compare the numerical simulations of the DM distribution in the Milky Way-type galaxies with the actual observations. However, the resolution of the N-body simulations is not yet sufficient to answer the questions about e.g. the DM density profiles in dwarf satellite galaxies. Moreover, most of the simulations include only collisionless DM particles, and do not model the baryons and their feedback on the galaxy structure formation. These problems are not solved even for the CDM simulations, and WDM simulations have additional serious difficulties. From an observational point of view, it has been argued for some time already that there is a discrepancy between CDM simulations and observations (see e.g.~\citealt{Moore:94,Moore:99,Klypin:99,Bode:00,Avila-Reese:01,Goerdt:06}) It has been claimed recently that a number of recent observations of dwarf satellite galaxies of the Milky way and Andromeda galaxy seem to indicate the existence of the smallest scale at which the DM exists~\citep{Gilmore:06,Gilmore:07a,Gilmore:07b,Koposov:07}. However, this statement and the interpretation of the observations are still subject to debate~\citep{Klimentowski:06,Penarrubia:07,Strigari:07,Simon:07}. Therefore it is too early to say what kind of DM models is favoured by comparing simulations and observations. Usually it is also necessary for the DM candidate to be stable. For the most popular DM candidate -- weakly interacting massive particles (WIMPs), this is related to the fact that the particles of $\sim$ electroweak mass, having weak strength interaction with SM matter (required to produce the correct amount of DM), would decay too fast and would not be ``dark''. If, however, the DM particle interacts with the SM more weakly than WIMPs, it could well have a finite (although cosmologically long) life time. There exist several unstable (decaying) DM candidates e.g. gravitino~\citep{Borgani:96,Baltz:01,Roszkowski:04,Cerdeno:05,Cembranos:06,Lola:07}. In this paper we will concentrate mainly on one candidate, the sterile neutrino (although our results will be applicable for any type of decaying DM). Constraints on the decaying DM were analyzed in~\cite{DeRujula:80,Berezhiani:87,Doroshkevich:89,Berezhiani:90a,Berezhiani:90b,Bertone:07,Zhang:07} (see also the book by~\citealt{Khlopov:97}). \subsubsection*{Sterile neutrino DM} \label{sec:sterile-neutrino-dm} It was noticed long ago that the right-handed (or as it is often called \emph{sterile}) neutrino with the mass in the keV range would represent a viable DM candidate~\citep{Dodelson:93}. Such a neutrino would interact with the rest of the matter only via the quadratic mixing with left-handed (\emph{active}) neutrinos and therefore (although not stable) could have cosmologically long life-time. At the same time, it could be produced in the early Universe with the correct abundances~\citep{Dodelson:93,Shi:98,Shaposhnikov:06}. One of the decay channels of the unstable sterile neutrinos includes emission of photons of the energy equal to half of the sterile neutrino rest energy. This potentially provides a possibility to observe the decays of DM sterile neutrinos via detection of a characteristic spectral line in the % spectra of astrophysical objects with large DM concentration. Recently this DM candidate has attracted much attention (see e.g.~\citet{Shaposhnikov:07a} and references therein). It was found that a very modest and natural extension of the SM by 3 right-handed neutrinos (making the SM more symmetric as all SM fermions, including neutrino, would have now their left and right handed counterparts) provided a viable extension of the theory, capable of solving several ``beyond the SM'' problems. First of all, such an extension makes neutrinos massive and thus perhaps provides the simplest and the most natural explanation of the phenomenon of ``neutrino oscillations''~(see e.g.~\cite{Fogli:05,Strumia:06,Giunti:06} for reviews). The smallness of neutrino masses in this model (called \numsm in~\citealt{Asaka:05b}) is achieved by the usual see-saw mechanism with Majorana masses of right-handed neutrinos being below electroweak scale.\footnote{The fact that the \numsm does not introduce any new scale above the electroweak one, makes this theory especially appealing from the point of view of its experimental verification/falsification.} Secondly, if two heavier sterile neutrinos ($N_2$ and $N_3$) are almost degenerate in mass and have their masses between $\mathcal{O}(100)\mev$ and $\mathcal{O}(20)\gev$, the \numsm provides the mechanism of generating the baryon asymmetry of the Universe. Thirdly, the lightest sterile neutrino $N_1$ can have arbitrary mass and arbitrarily weak coupling with the (active) neutrino sector. At the same time, it can be produced in the early Universe in the correct amounts. It represents therefore the DM particle in the \numsm. Thus, altogether the \numsm represents (arguably) the simplest extension of the SM, capable of explaining three important questions: origin and smallness of neutrino masses, baryon asymmetry in the Universe and the existence of the DM. \subsubsection*{Existing restrictions on sterile neutrino DM parameters.} What are the current restrictions on parameters (mass and \emph{mixing}) of sterile neutrino DM? First of all sterile neutrino mass should satisfy the universal Tremaine-Gunn lower bound:\footnote{In its simplest form the Tremaine-Gunn bound comes from the fact that for the fermions there is a maximal density in the phase space~\citep{Tremaine:79,Dalcanton:00} and therefore the observed phase-space density in various DM dominated systems should be less that this (mass dependent) bound.} $\mdm\gtrsim 300-500\ev$.\footnote{A stronger lower bound from Ly-$\alpha$~\citep{Seljak:06,Viel:06,Viel:07} can be obtained in the case of the particular production mechanisms -- the Dodelson-Widrow scenario~\citep{Dodelson:93}. For other possible production mechanisms~\citep[e.g.][]{Shi:98,Shaposhnikov:06} the Ly-$\alpha$ constraints should be reanalyzed.} Next, as the sterile neutrino possesses the (two-body) radiative decay channel: $N_1 \to \nu + \gamma$, the emitted photon would carry the energy $E_\gamma = \mdm/2$. A large flux of such photons is expected from the large concentrations of the DM sterile neutrinos, like galaxies or galaxy clusters. Recently an extensive search of the DM decay line in the region of masses $M_\dm \lesssim 20\kev$ was conducted, using the data of \emph{Chandra}~\citep{Riemer:06,Boyarsky:06e,Abazajian:06b} and \emph{XMM-Newton}~\citep{Boyarsky:05,Boyarsky:06b,Boyarsky:06c,Watson:06,Boyarsky:06d}. The region of soft X-ray (down to energies $0.2$~keV) was explored by~\citet{Boyarsky:06f} with the use of the wide field of view spectrometer~\citep{McCammon:02}. The non-observation of the DM decay line in X-ray, combined with the first principles calculation of DM production in the early Universe~\citep{Asaka:06c}, implies that the \citet{Dodelson:93} (DW) scenario can work only if the sterile neutrino mass is below 4~keV~\citep{Boyarsky:07a}. If one takes into account recent \emph{lower} bound on the mass of sterile neutrino DM in the DW scenario $\mdm \ge 5.6\kev$~\citep{Viel:07}, it seems that the possibility that all the DM is produced via DW scenario is ruled out~\citep{Boyarsky:07a}. The possibility that only fraction of the DM is produced via DW mechanism remains open~\citep{Palazzo:07}. There are other viable mechanisms of DM production, including e.g. resonant oscillation production in the presence of lepton asymmetries~\citep{Shi:98}. Sterile neutrino DM can be produced by the decay of light inflaton \citep{Shaposhnikov:06} or in a similar model with the different choice of parameters~\citep{Kusenko:06a,Petraki:07}. These mechanisms are currently not constrained and remain valid for DM particles with the masses in the keV range and above. The search for the DM decay line signal produced by sterile neutrinos with masses above $\sim 20$~keV is complicated by the absence of the focusing optics telescopes (similar to {\it Chandra} or {\it XMM-Newton}) in the hard X-ray and $\gamma$-ray domain of the spectrum. For example, the existing restrictions in the $20-100$~keV mass range \citep{Boyarsky:05,Boyarsky:06c} are derived from the observations of diffuse X-ray background, with the help of non-imaging instruments, HEAO-I~\citep{Gruber:99}. The current status of astrophysical observations in summarized in~\cite{Ruchayskiy:07}. In this paper we use the spectrometer SPI on board of INTEGRAL satellite to place restrictions on parameters of decaying DM in the mass range $40\kev - 14\mev$. This range of masses is interesting, for example, the sterile neutrinos, produced in the early Universe in the presence of large lepton asymmetries~\citep{Shi:98} or through the inflaton decay~\citep{Shaposhnikov:06}. It is also relevant for the case of gravitino DM~\citep{Pagels:82,Bond:82}. When the preparation of this paper was at its final stage,~\citet[hereafter \textbf{Y07}]{Yuksel:07} published their work, which used the results of~\citet[hereafter \textbf{T06}]{Teegarden:06} to place restrictions on the parameters of sterile neutrino DM in the range $40-700\keV$. We discuss it in more details in Section~\ref{sec:discussion}. \subsubsection*{SPI spectrometer} The absence of the focusing optics significantly reduces the sensitivity of the telescopes operating in the hard X-ray/soft \gr\ energy band. Most of the instruments operating in this energy band use collimators and/or coded masks to distinguish signals from the sources on the sky from the instrumental background. Contrary to the focusing optics telescopes, both the source and background signals are collected from the entire detector, which significantly increases the irreducible background. The focusing optics enables to significantly reduce the background only in the studies of point sources. If the source under investigation occupies a large fraction of the sky (e.g. the entire Milky Way galaxy), the performance of the focusing and non-focusing instruments with the same detector collection area are, in fact, comparable. \begin{figure} \centering \includegraphics[width=\linewidth]{SENSITIVITY_SPI} % \caption{Comparison of sensitivity towards the search of the narrow DM decay line for different instruments with the wide FoV. Diagonal straight lines show the improvement of sensitivity (by a factor, marked on the line) as compared with the HEAO-I A4 low energy detector (LED), taken as a reference.} \label{fig:spi_vs_heao} \end{figure} In the case of an extended source, emitting a narrow spectral line, an efficient way of reduction of instrumental background is via the improvement of the spectral resolution of the instrument (in the case of a broad continuum background spectrum, the number of background counts at the energy of the line is proportional to the spectral resolution $\Delta E$). The best possible sensitivity is achieved when the spectral resolution reaches the intrinsic width of the spectral line (see Fig.\ref{fig:spi_vs_heao} for the case of wide FoV instruments and~\cite{Boyarsky:06f} for the case of narrow FoV instruments). In the case of the line produced by the DM decaying in the Milky Way halo, the line width is determined by the Doppler broadening by the random motion of the DM particles. The velocity dispersion of the DM motion in the halo is about the rotation velocity of the Galactic disk, $v\sim 200$~km/s. This means that Doppler broadening of the DM decay line is about \begin{equation} \label{eq:7} \frac{\Delta E}{E}\sim \frac{v}{c}\simeq 10^{-3}\;. \end{equation} Thus, the optimal spectral resolution of an instrument searching for the DM decay line produced by the Milky Way DM halo should be $\Delta E\simeq 10^{-3}E$. Such optimal spectral resolution is almost achieved with the spectrometer SPI on board of \intgr\ satellite, which has the maximal spectral resolving power of $E/\Delta E\simeq 500$ and works in the energy range 20~keV -- 8~MeV~\citep{spi}. SPI is a ``coded mask'' type instrument with an array of 19 hexagonal shaped Ge detectors (of which only 17 are operating at the moment). \begin{figure} \begin{center} \includegraphics[width=\linewidth]{FoV} \end{center} \caption{The geometry of the SPI FoV.} \label{fig:FoV} \end{figure} The SPI telescope consists of a coded mask inscribed into a circle of the radius $R_{\rm mask}=39$~cm, placed at the height $H=171$~cm above the detector plane and of the detector, which has the shape of a hexagon inscribed into a circle of the radius $R_{\rm det}\simeq 15.3$~cm (see Fig. \ref{fig:FoV}). The portion of the sky visible from each point of the SPI detector (the so-called \emph{fully coded field of view}, FCFOV) has therefore angular diameter \begin{equation} \label{eq:12} \Theta_\fcfov=2\arctan\left[\frac{R_{\rm mask}-R_{\rm det}}H\right] \approx 16^\circ\;, \end{equation} while the portion of the sky visible by at least some of the detectors (the \emph{partially coded field of view}, PCFOV) is \begin{equation} \label{eq:14} \Theta_\pcfov=2\arctan\left[\frac{R_{\rm mask}+R_{\rm det}}H\right] \approx 35^\circ\;. \end{equation} The solid angle spanned by the cone with this opening angle is $\Omega_\pcfov=2\pi\Bigl(1-\cos(\Theta_\pcfov/2)\Bigr)\simeq 0.29$ (see Fig.~\ref{fig:FoV}). Wide field of view makes the SPI telescope suitable for the study of the very extended sources, like the Milky Way DM halo. \begin{figure} \begin{center} \includegraphics[width=\linewidth]{Aeff} \end{center} \caption{The effective area of the SPI detector for an on-axis source, as a function of the photon energy. The plot is produced by collective the on-axis effective areas of the 17 SPI detectors from the instrumental characteristics files.} \label{fig:Aeffon} \end{figure} | \label{sec:discussion} The purpose of this work was to understand how to search for the DM decay line with the SPI spectrometer and to check that none of the strong lines, present in the SPI background, was confused with the DM decay line. Our analysis shows that all the strong lines were, indeed, of instrumental origin and provides the upper bound on the flux of ``weak'' ($3{-}4\sigma$ above the background) lines, which leads to the corresponding restrictions (see Sec.~\ref{sec:exclusions}). To further improve the results, one needs to work with the weak lines (or lines, coinciding in position with instrumental ones). To do this one needs more sophisticated procedures of subtraction of the instrumental background (e.g. imaging). One of the most interesting cases of the coinciding instrumental and celestial line is the positronium annihilation line at 511 keV. An excess of positron annihilation emission on top of the strong instrumental line (related to positrons annihilating inside the detector) was noticed long ago~\citep[for an incomplete set of references see e.g.][]{Prantzos:93,Milne:99,Cheng:97,Purcell:97,Knodlseder:05,Weidenspointner:06,Weidenspointner:07}. There exist many attempts of explanation of this excess. In particular, it was attributed to the annihilating or decaying DM~\citep[see e.g.][]{Boehm:03,Hooper:03,Boehm:06,Frere:06,Picciotto:04,Rasera:05}. The sterile neutrino DM with the mass $m_s > 1\mev$ possesses decay channel $N_s\to e^+e^-\nu$, with positrons annihilating either in flight or at rest, by forming the positronium atom~\citep[see e.g.][]{Beacom:06,Sizun:06}. Thus, it is possible that the decay of sterile neutrino DM contributes to such a line. The detailed analysis of this case will be reported separately. It should be also mentioned, that the region of masses between $20\kev \lesssim m_\dm \lesssim 40\kev$ remains inaccessible for the existing X-ray missions. The strongest restrictions in this region were produced, using the data of HEAO-1 mission \citep{Boyarsky:06c}. When the work on this paper was at its final stage, the work of Y07 was published. Y07 obtained the restrictions on parameters of sterile neutrino in the range 40 keV -- 700 keV. To facilitate the comparison, we plot the restrictions of Y07 on Fig.~\ref{fig:on_limit}, (divided by the factor of 2 to translate them into the restrictions for the Majorana, rather than Dirac sterile neutrino DM, see footnote~\ref{fn:1}, p~\pageref{fn:1}). As the data, used in our work, has about 5 times longer exposure than the \intgr\ first years data, on which the results of Y07 are based, we could have expected results stronger by a factor $\approx 2$ in our case. However, the Fig.~\ref{fig:on_limit} shows the opposite. The reason for this is as follows. For the SPI, the sensitivity towards the line search from a particular source depends on the shape of the source. In particular, the results of TW06, on which the work of Y07 was based, were obtained under the assumption of a particular diffuse source ($10^\circ$ Gaussian). As any realistic DM profile is much flatter than the $10^\circ$ Gaussian, the results of TW06 cannot be applied directly for the case of the DM line search. They should be rescaled to account for the diffuse nature of the DM source~(c.f. Section~\ref{sec:exclusions}). Apart from this, the estimated DM signal from the inner part of the Galaxy is about 2 times stronger in Y07 than in our work. As the DM signal in the direction of the GC is the most uncertain, we have adopted the conservative flat profile everywhere inside the solar radius, to minimize this uncertainty. \subsection* | 7 | 10 | 0710.4922 |
0710 | 0710.1500_arXiv.txt | Observations of the ELAIS-N1 field taken at 610~MHz with the Giant Metrewave Radio Telescope are presented. Nineteen pointings were observed, covering a total area of $\sim9$~deg$^{2}$ with a resolution of 6~$\times$~5~arcsec$^{2}$, PA $+45\degr$. Four of the pointings were deep observations with an rms of $\sim40$~$\mu$Jy before primary beam correction, with the remaining fifteen pointings having an rms of $\sim70~\mu$Jy. The techniques used for data reduction and production of a mosaicked image of the region are described, and the final mosaic is presented, along with a catalogue of 2500 sources detected above 6$\sigma$. This work complements the large amount of optical and infrared data already available on the region. We calculate 610-MHz source counts down to 270~$\mu$Jy, and find further evidence for the turnover in differential number counts below 1~mJy, previously seen at both 610~MHz and 1.4~GHz. | The {\it Spitzer} Wide-area Infrared Extragalactic \citep[SWIRE;][]{Lonsdale03} survey has the largest sky coverage of the legacy surveys being performed by the {\it Spitzer Space Telescope} \citep{Werner04}. A total area of $\sim$49~deg$^{2}$ of sky has been observed with the Infrared Array Camera \citep[IRAC;][]{Fazio04} and Multiband Imaging Photometer for {\it Spitzer} \citep[MIPS;][]{Rieke04} instruments at 3.6, 4.5, 5.8, 8, 24, 70 and 160~$\mu$m. The survey is broken down into six fields, three in the northern sky -- ELAIS-N1, ELAIS-N2 and the Lockman Hole -- and three in the south -- ELAIS-S1, {\it Chandra} Deep Field South and the {\it XMM}-Large Scale Structure (XMM-LSS) field. All six regions were selected to be away from the Galactic disk, in order to minimize background cirrus emission. There is a large amount of multi-wavelength information available on all six SWIRE fields. The three ELAIS fields were observed as part of the European Large-Area {\it ISO} Survey \citep{Oliver00}, which also included another northern (-N3) and southern (-S2) field. The {\it Infrared Space Observatory} ({\it ISO}) observed these regions at 6.7, 15, 90 and 175~$\mu$m, and a large number of followup observations were carried out in the optical, infrared and radio bands. A band-merged catalogue, containing the {\it ISO} data, along with $U$, $g'$, $r'$, $i'$, $Z$, $J$, $H$ and $K$-band detections, and radio observations at 1.4~GHz has been produced -- for more details, see \citet{RowanRobinson04}, and references therein. Observations of the ELAIS-N1 region were taken with {\it Spitzer} in 2004 January, covering $\sim9$~deg$^{2}$ with the IRAC and MIPS instruments. The source catalogues have been produced, and are available online \citep{Surace04}, containing over 280,000 sources. The UK Infrared Deep Sky Survey \citep[UKIDSS;][]{Lawrence07} intends to cover the ELAIS-N1 region in its Deep Extragalactic Survey plan, observing the full field in the $J$, $H$ and $K$-bands to a depth of $K$ = 21~mag. This will be a great improvement over the currently available surveys, which have a sensitivity limit of $\sim$18~mag in the $K$ band. Data Release 2 \citep{Warren07} of UKIDSS contains early shallow data on the ELAIS-N1 region. Further surveys have been carried out in the $R$-band \citep{Fadda04}, in H$\alpha$ \citep{Pascual01}, and with the {\it Chandra} X-ray telescope \citep{Manners03,Franceshini05}. There have been several redshift surveys of the region \citep{Trichas06,Berta07}, and the ELAIS-N1 region was also partially covered by the Sloan Digital Sky Survey \citep[SDSS;][]{AdelmanMcCarthy07}. While there have been a great number of observations of the ELAIS-N1 region at optical and infrared wavelengths, there is comparatively little radio information available. The existing VLA 1.4~GHz survey of the three northern ELAIS fields \citep{Ciliegi99}, which has been included into the band-merged catalogue of \citet{RowanRobinson04}, reaches a 5$\sigma$ limit of 0.135~mJy over 0.12~deg$^{2}$ but only a 1.15~mJy limit over its full coverage area of 4.22~deg$^{2}$. The NVSS \citep{Condon98} and FIRST \citep{Becker95} surveys both cover the ELAIS-N1 region, but only to relatively shallow $5\sigma$ limits of 2.25 and 0.75~mJy respectively. A recent study of polarised compact sources \citep{Taylor07} at 1420~MHz is underway, using the Dominion Radio Astrophysical Observatory Synthesis Telescope (DRAO ST) centered on $16^{\rm h}11^{\rm m}00^{\rm s}$, $+55\degr00'00''$ and covering 7.4~deg$^{2}$. The first 30~per~cent of observations have been completed, with maps in Stokes I, Q and U being produced with a maximum sensitivity of 78~$\mu$Jy~beam$^{-1}$, although with a resolution of $\sim$1~arcmin$^{2}$. In order to extend the information on this region, a much larger deep radio survey is required. In this paper, we present observations of the ELAIS-N1 survey field taken at 610~MHz with the Giant Metrewave Radio Telescope \citep[GMRT;][]{Ananthakrishnan05}, covering $\sim9$~deg$^{2}$ of sky with a resolution of $6\times5$~arcsec$^{2}$, PA~$+45\degr$, centred on $16^{\rm h}11^{\rm m}00^{\rm s}$, $+55\degr00'00''$ (J2000 coordinates, which are used throughout this paper). This survey, in combination with the deep {\it Spitzer} data, will be used to study the infrared/radio correlation for star-forming systems \citep[e.g.][]{Appleton04}, and the link between the triggering of star formation and AGN activity, as well as the properties of the faint radio population at 610~MHz. In Section~\ref{sec:observations} we describe the observations and data reduction techniques used in the creation of the survey. Section~\ref{sec:results} presents the mosaic and a source catalogue containing 2500 sources above 6$\sigma$, along with a sample of extended sources. In Section~\ref{sec:sourcecounts} we construct the 610~MHz differential source counts, and compare them to previous works. | \label{sec:results} \begin{figure} \includegraphics[width=8cm]{EN1NOISE.PS} \caption{The rms noise of the final mosaic, calculated using Source Extractor. The grey-scale ranges between 40 and 350~$\mu$Jy, and the contours are at 60 and 120~$\mu$Jy respectively.} \label{fig:EN1noise} \end{figure} Source Extractor \citep[SExtractor;][]{Bertin96} was used to calculate the rms noise $\sigma$ across the mosaic. A grid of $16\times16$ pixels was used in order to track changes in the local noise level, which varies significantly near the brightest sources. Fig.~\ref{fig:EN1noise} illustrates the local noise, with the grey-scale varying between 40~$\mu$Jy (the noise level in the centre of the deep pointings) and 350~$\mu$Jy (the noise level for the shallow pointings, at the distance where the GMRT primary beam gain was 20~per~cent of its central value). The 60 and 120~$\mu$m contours are plotted, which cover the majority of the deep and shallow survey regions respectively. A sample region of the ELAIS-N1 survey is shown in Fig.~\ref{fig:EN1sample} to illustrate the quality of the image. Most of the sources in our survey are unresolved, although there are several with extended structures. We present a sample of these in Fig.~\ref{fig:extended}. \begin{figure*} \includegraphics[width=16cm]{EN1area.eps} \caption{A $70\times70$~arcmin$^{2}$ region of the 610~MHz image, to illustrate the quality of the survey. The region is located within the deeper area of the survey, and most sources are unresolved. The grey-scale ranges between $-0.2$ and 1~mJy~beam$^{-1}$, and the noise is relatively uniform and between 40 and 60~$\mu$Jy, apart from small regions near bright sources where the noise increases.} \label{fig:EN1sample} \end{figure*} \begin{figure*} \centerline{\subfigure[GMRTEN1~J161137.8$+$555955] {\includegraphics[width=5cm]{IMG12.PS}} \subfigure[GMRTEN1~J160640.9$+$560136] {\includegraphics[width=5cm]{IMG13.PS}} \subfigure[GMRTEN1~J161530.7$+$545231] {\includegraphics[width=5cm]{IMG02.PS}}} \centerline{\subfigure[GMRTEN1~J160929.9$+$552444 and J160931.09$+$552503] {\includegraphics[width=5cm]{IMG04.PS}} \subfigure[GMRTEN1~J161858.8$+$545227, J161900.3$+$545305 and J161903.3$+$545240] {\includegraphics[width=5cm]{IMG05.PS}} \subfigure[GMRTEN1~J161148.4$+$550049, J161151.5$+$550053 and J161154.4$+$550057] {\includegraphics[width=5cm]{IMG07.PS}}} \centerline{\subfigure[GMRTEN1~J160808.7$+$544723, J160809.5$+$544659, J160810.3$+$544633 and J160810.7$+$544645] {\includegraphics[width=5cm]{IMG08.PS}} \subfigure[GMRTEN1~J161027.4$+$541246] {\includegraphics[width=5cm]{IMG09.PS}} \subfigure[GMRTEN1~J161757.1$+$545110] {\includegraphics[width=5cm]{IMG03.PS}}} \centerline{\subfigure[GMRTEN1~J161140.5$+$554703 and J161142.8$+$554727] {\includegraphics[width=5cm]{IMG14.PS}} \subfigure[GMRTEN1~J160634.8$+$543456] {\includegraphics[width=5cm]{IMG15.PS}} \subfigure[GMRTEN1~J160334.6$+$542900 and J160333.1$+$542914] {\includegraphics[width=5cm]{IMG16.PS}}} \caption{A selection of extended objects in the ELAIS-N1 610~MHz GMRT survey -- contours are plotted at $\pm200~\mu$Jy $\times$ 1, $\sqrt{2}$, 2, $2\sqrt{2}$, 4$\ldots$, with the exception of object (e), where the contours start at $\pm$500~$\mu$Jy. Negative contours are represented by dashed lines. The resolution of the beam is shown in the bottom left of each image, and the designations of each source component are given below.} \label{fig:extended} \end{figure*} Our GMRT data suffer from dynamic range problems near the brightest sources, and the final mosaic has increased noise and residual sidelobes in these regions. We had fewer problems with our survey of the xFLS region, due to the longer time spent on each pointing and the correspondingly better {\it uv} coverage. There were also fewer bright sources in the xFLS field, so a much smaller region was affected by residual sidelobes. Fig.~\ref{fig:BrightSource} shows an area around one of the bright sources, to illustrate the problems caused by the residual sidelobes. While the local noise calculated by SExtractor increases due to these residuals, some of them still have an apparent signal-to-noise level that is greater than 6. We therefore opted for a two-stage selection criteria for our final catalogue. \begin{figure} \includegraphics[width=8cm]{BrightSource.ps} \caption{Source GMRTEN1~J161212.4$+$552308, and the errors surrounding it. The grey-scale ranges between $-0.2$ and 1~mJy~beam$^{-1}$, and the source has a peak brightness of 389~mJy. The region affected by an overdensity of sources is shown by the black circle, of radius 10~arcmin -- see text for more details.} \label{fig:BrightSource} \end{figure} \subsection{Source fitting} An initial catalogue of 4767 sources was created using SExtractor. The mosaic was cut off at the point where the primary beam correction dropped to 20~per~cent of its central value (a radius of 0\fdg53 from the outer pointings), however only sources inside the 30~per~cent region (0\fdg47) were included in the catalogue to avoid the mosaic edges from affecting the estimation of local noise. The requirements for a source to be included were that it had at least 5 connected pixels with brightness greater than 3$\sigma$, and a peak brightness greater than 6$\sigma$. The image pixel size meant that the beam was reasonably oversampled, so the source peak was taken to be the value of the brightest pixel within a source. The integrated flux density was calculated using the FLUX\_AUTO option within SExtractor. This creates an elliptical aperture around each object \citep[as described in][]{Kron80}, and integrates the flux contained within the ellipse. Comparisons between the flux density obtained through this method and through the method developed in \citet{Garn07} -- pixels above a given threshold were summed, then empirically corrected for the elliptical beam shape -- give good agreement between the two techniques. Sources with Kron flux density above 1~mJy showed no statistical difference between the two flux density measurements, with an uncertainty of 3~per~cent. Sources below 1~mJy had a Kron flux density that was systematically larger than the \citet{Garn07} method by 6~per~cent, with an uncertainty of 4~per~cent. We chose to use the Kron method in order to avoid the empirical correction factor. In order to estimate the area affected by artefacts near the bright sources, we calculated the number of (potentially spurious) sources in a series of concentric rings centred on a bright source, and converted this to an effective source density as a function of distance. While the noise does vary across the map, the variation is smooth and relatively slow away from the bright sources. Using the more distant rings we calculated a mean source sky density $\mu$, and an estimate of the error in this value, $\sigma_{\mu}$, which will be unaffected by the presence of the spurious sources, and will vary slightly between each bright central source due to the changing properties of the image. Fig.~\ref{fig:spurioussources} shows the source density (in arbitrary units) away from the source in Fig.~\ref{fig:BrightSource}. The clear over-density of sources can be seen within 10~arcmin of the source. If there is a significant peak in the density (greater than $\mu + 6\sigma_{\mu}$), then we define the size of the affected region by finding the first radius at which the source density drops below $\mu + 3\sigma_{\mu}$ -- for this source, the affected radius is 10~arcmin. The affected region has been added to Fig.~\ref{fig:BrightSource}. Within this region, only sources with a peak brightness greater than 12$\sigma$ are included in our catalogue -- this value was determined empirically. The source density plot for a 10~mJy source is shown in Fig.~\ref{fig:FaintSource}, on the same scale as Fig.~\ref{fig:spurioussources} -- while there may still be a slight overdensity near the centre, the increased noise level in this region, along with the $6\sigma$ cut-off reduces the risk of selecting a spurious source near sources with weak over-densities. \begin{figure} \includegraphics[width=8cm]{Bright.eps} \caption{Density of sources, in concentric rings around the bright source shown in Fig.~\ref{fig:BrightSource}. The overdensity of sources near to the bright (389~mJy) central object can be clearly seen. The mean source density, far from the central source, is given by the large dashed line while the cutoff density defining the affected region is given by short dashes -- see text for more details.} \label{fig:spurioussources} \end{figure} \begin{figure} \includegraphics[width=8cm]{Faint.eps} \caption{Density of sources, in concentric rings around a fainter (10~mJy) object in the ELAIS-N1 survey field. There is still a slight overdensity near the central source, but the signal-to-noise cutoff of $6\sigma$ ensures that spurious sources are not included in the catalogue.} \label{fig:FaintSource} \end{figure} This analysis was repeated for all sources with a peak greater than 10~mJy in order to filter spurious sources. The final catalogue contains 2500 sources -- we have erred on the side of caution in order to produce a catalogue with little contamination from spurious sources. The size of the affected region is correlated with the peak brightness of a source (with Pearson product-moment correlation coefficient of 0.53), and the number of spurious sources is also correlated with the peak brightness, with correlation coefficient 0.73. The precise size of the affected region depends on the {\it uv} coverage for the relevant pointing, the time spent on observations and the local noise levels. Table $\ref{tab:catalogue}$ presents a sample of 60 entries in the catalogue, which is sorted by right ascension. The full table is available via {\tt http://www.mrao.cam.ac.uk/surveys/}. Column~1 gives the IAU designation of the source, in the form GMRTEN1~Jhhmmss.s$+$ddmmss, where J represents J2000.0 coordinates, hhmmss.s represents right ascension in hours, minutes and truncated tenths of seconds, and ddmmss represents the declination in degrees, arcminutes and truncated arcseconds. Columns~2 and 3 give the right ascension and declination of the source, calculated by first moments of the relevant pixel brightnesses to give a centroid position. Column~4 gives the brightness of the peak pixel in each source, in mJy~beam$^{-1}$, and column~5 gives the local rms noise in $\mu$Jy~beam$^{-1}$. Columns~6 and 7 give the integrated flux density and error, calculated from the local noise level and source size. Columns 8 and 9 give the $X$, $Y$ pixel coordinates of the source centroid from the mosaic image. Column 10 is the Source Extractor deblended object flag -- 1 where a nearby bright source may be affecting the calculated flux, 2 where a source has been deblended into two or more components from a single initial island of flux, and 3 when both of the above criteria apply. There are 232 sources present in our catalogue with non-zero deblend flags; it is necessary to examine the images to distinguish between the case where one extended object has been represented by two or more entries, and where two astronomically distinct objects are present. \begin{table*} \label{tab:catalogue} \caption{A sample of 60 entries from the 610-MHz ELAIS-N1 catalogue, sorted by right ascension. The full version of this table is available as Supplementary Material through the online version of this article, and via {\tt http://www.mrao.cam.ac.uk/surveys/}.} \begin{tabular}{cccccccccc} \hline Name & RA & Dec.\ & Peak & Local~Noise & Int.\ Flux Density & Error & $X$ & $Y$ & Flags\\ & J2000.0 & J2000.0 & mJy~beam$^{-1}$ & $\mu$Jy~beam$^{-1}$ & mJy & mJy & & & \\ (1) & (2) & (3) & (4) & (5) & (6) & (7) & (8) & (9) & (10) \\ \hline GMRTEN1~J160319.2$+$542543 & 16:03:19.22 & $+$54:25:43.6 & 0.496 & 75 & 0.415 & 0.067 & 6599 & 2946 & 0\\ GMRTEN1~J160319.2$+$553149 & 16:03:19.24 & $+$55:31:49.1 & 0.469 & 66 & 0.383 & 0.054 & 6527 & 5589 & 0\\ GMRTEN1~J160319.3$+$554950 & 16:03:19.38 & $+$55:49:50.2 & 0.514 & 85 & 0.265 & 0.054 & 6506 & 6309 & 0\\ GMRTEN1~J160320.8$+$550645 & 16:03:20.89 & $+$55:06:45.9 & 2.617 & 97 & 3.425 & 0.175 & 6545 & 4587 & 0\\ GMRTEN1~J160321.1$+$553654 & 16:03:21.13 & $+$55:36:54.2 & 0.418 & 69 & 0.309 & 0.076 & 6510 & 5792 & 0\\ GMRTEN1~J160322.6$+$554320 & 16:03:22.67 & $+$55:43:20.8 & 0.560 & 75 & 0.444 & 0.068 & 6495 & 6049 & 0\\ GMRTEN1~J160323.1$+$543737 & 16:03:23.13 & $+$54:37:37.3 & 1.920 & 74 & 2.256 & 0.119 & 6564 & 3421 & 0\\ GMRTEN1~J160324.0$+$550618 & 16:03:24.05 & $+$55:06:18.2 & 0.635 & 84 & 0.495 & 0.079 & 6527 & 4568 & 0\\ GMRTEN1~J160326.0$+$540817 & 16:03:26.00 & $+$54:08:17.1 & 0.595 & 94 & 0.486 & 0.081 & 6579 & 2248 & 0\\ GMRTEN1~J160327.7$+$552647 & 16:03:27.78 & $+$55:26:47.3 & 3.240 & 84 & 3.867 & 0.149 & 6484 & 5387 & 0\\ GMRTEN1~J160327.9$+$543326 & 16:03:27.99 & $+$54:33:26.0 & 0.452 & 75 & 0.338 & 0.064 & 6540 & 3253 & 0\\ GMRTEN1~J160329.5$+$540705 & 16:03:29.55 & $+$54:07:05.7 & 0.869 & 131 & 0.681 & 0.117 & 6559 & 2200 & 0\\ GMRTEN1~J160330.8$+$542454 & 16:03:30.83 & $+$54:24:54.3 & 0.483 & 78 & 0.837 & 0.097 & 6533 & 2912 & 0\\ GMRTEN1~J160331.1$+$554327 & 16:03:31.14 & $+$55:43:27.2 & 0.473 & 76 & 0.555 & 0.071 & 6447 & 6052 & 0\\ GMRTEN1~J160331.3$+$545000 & 16:03:31.39 & $+$54:50:00.7 & 0.408 & 67 & 0.322 & 0.055 & 6503 & 3916 & 0\\ GMRTEN1~J160332.3$+$553000 & 16:03:32.38 & $+$55:30:00.2 & 0.508 & 69 & 0.536 & 0.076 & 6454 & 5514 & 0\\ GMRTEN1~J160332.6$+$554622 & 16:03:32.66 & $+$55:46:22.6 & 3.212 & 76 & 4.152 & 0.153 & 6435 & 6169 & 0\\ GMRTEN1~J160332.9$+$541746 & 16:03:32.90 & $+$54:17:46.8 & 0.474 & 73 & 0.381 & 0.060 & 6528 & 2627 & 0\\ GMRTEN1~J160333.1$+$542914 & 16:03:33.11 & $+$54:29:14.2 & 2.377 & 73 & 9.697 & 0.226 & 6515 & 3085 & 3\\ GMRTEN1~J160333.6$+$552623 & 16:03:33.69 & $+$55:26:23.7 & 0.773 & 78 & 0.878 & 0.101 & 6451 & 5370 & 0\\ GMRTEN1~J160333.7$+$540540 & 16:03:33.70 & $+$54:05:40.6 & 7.541 & 294 & 9.799 & 0.552 & 6536 & 2142 & 0\\ GMRTEN1~J160334.6$+$542900 & 16:03:34.68 & $+$54:29:00.2 & 2.326 & 80 & 4.235 & 0.167 & 6506 & 3075 & 3\\ GMRTEN1~J160335.1$+$551534 & 16:03:35.19 & $+$55:15:34.1 & 0.462 & 74 & 0.333 & 0.066 & 6454 & 4937 & 0\\ GMRTEN1~J160335.2$+$555419 & 16:03:35.24 & $+$55:54:19.5 & 0.762 & 105 & 0.984 & 0.108 & 6412 & 6486 & 0\\ GMRTEN1~J160335.2$+$540515 & 16:03:35.25 & $+$54:05:15.3 & 38.494 & 383 & 50.351 & 0.961 & 6528 & 2125 & 0\\ GMRTEN1~J160336.5$+$555147 & 16:03:36.55 & $+$55:51:47.5 & 0.500 & 79 & 0.662 & 0.082 & 6408 & 6385 & 0\\ GMRTEN1~J160336.9$+$544120 & 16:03:36.90 & $+$54:41:20.7 & 0.532 & 68 & 0.853 & 0.092 & 6480 & 3568 & 0\\ GMRTEN1~J160337.6$+$545944 & 16:03:37.64 & $+$54:59:44.2 & 0.547 & 80 & 0.530 & 0.077 & 6457 & 4303 & 0\\ GMRTEN1~J160338.7$+$554348 & 16:03:38.77 & $+$55:43:48.1 & 0.771 & 80 & 1.743 & 0.135 & 6404 & 6065 & 0\\ GMRTEN1~J160339.0$+$545943 & 16:03:39.05 & $+$54:59:43.8 & 0.500 & 80 & 0.415 & 0.065 & 6449 & 4303 & 0\\ GMRTEN1~J160339.3$+$551352 & 16:03:39.37 & $+$55:13:52.0 & 0.516 & 78 & 0.414 & 0.063 & 6432 & 4868 & 0\\ GMRTEN1~J160339.6$+$542953 & 16:03:39.68 & $+$54:29:53.2 & 0.701 & 66 & 0.647 & 0.078 & 6476 & 3110 & 0\\ GMRTEN1~J160339.7$+$550600 & 16:03:39.75 & $+$55:06:00.4 & 0.584 & 80 & 0.418 & 0.069 & 6438 & 4554 & 0\\ GMRTEN1~J160340.1$+$545127 & 16:03:40.10 & $+$54:51:27.4 & 0.711 & 79 & 0.639 & 0.079 & 6451 & 3972 & 0\\ GMRTEN1~J160340.1$+$550544 & 16:03:40.15 & $+$55:05:44.9 & 0.495 & 81 & 0.392 & 0.069 & 6436 & 4543 & 0\\ GMRTEN1~J160340.8$+$554325 & 16:03:40.83 & $+$55:43:25.8 & 1.447 & 81 & 4.633 & 0.210 & 6392 & 6050 & 0\\ GMRTEN1~J160341.2$+$552611 & 16:03:41.27 & $+$55:26:11.8 & 0.475 & 68 & 0.460 & 0.065 & 6408 & 5361 & 0\\ GMRTEN1~J160341.5$+$552205 & 16:03:41.59 & $+$55:22:05.4 & 0.638 & 69 & 0.614 & 0.080 & 6411 & 5197 & 0\\ GMRTEN1~J160341.6$+$553203 & 16:03:41.66 & $+$55:32:03.4 & 0.462 & 69 & 0.365 & 0.062 & 6400 & 5595 & 0\\ GMRTEN1~J160343.1$+$540324 & 16:03:43.13 & $+$54:03:24.6 & 0.908 & 126 & 0.556 & 0.092 & 6483 & 2050 & 0\\ GMRTEN1~J160344.8$+$540320 & 16:03:44.84 & $+$54:03:20.4 & 1.168 & 118 & 0.856 & 0.110 & 6473 & 2047 & 0\\ GMRTEN1~J160345.8$+$553021 & 16:03:45.84 & $+$55:30:21.4 & 0.381 & 60 & 0.370 & 0.065 & 6378 & 5527 & 0\\ GMRTEN1~J160345.8$+$554238 & 16:03:45.88 & $+$55:42:38.3 & 0.653 & 76 & 2.566 & 0.156 & 6365 & 6018 & 0\\ GMRTEN1~J160346.5$+$552855 & 16:03:46.56 & $+$55:28:55.8 & 0.577 & 73 & 0.510 & 0.076 & 6375 & 5469 & 0\\ GMRTEN1~J160346.6$+$550826 & 16:03:46.62 & $+$55:08:26.0 & 0.525 & 70 & 0.416 & 0.065 & 6396 & 4650 & 0\\ GMRTEN1~J160348.3$+$542626 & 16:03:48.37 & $+$54:26:26.4 & 2.328 & 80 & 2.660 & 0.133 & 6429 & 2970 & 0\\ GMRTEN1~J160348.6$+$550124 & 16:03:48.61 & $+$55:01:24.3 & 0.477 & 78 & 0.257 & 0.050 & 6392 & 4368 & 0\\ GMRTEN1~J160349.2$+$554243 & 16:03:49.21 & $+$55:42:43.9 & 3.349 & 85 & 4.224 & 0.163 & 6346 & 6021 & 0\\ GMRTEN1~J160350.0$+$545634 & 16:03:50.06 & $+$54:56:34.6 & 0.435 & 67 & 0.371 & 0.058 & 6389 & 4175 & 0\\ GMRTEN1~J160350.4$+$550717 & 16:03:50.43 & $+$55:07:17.2 & 0.597 & 78 & 0.626 & 0.083 & 6376 & 4603 & 0\\ GMRTEN1~J160350.5$+$541302 & 16:03:50.51 & $+$54:13:02.7 & 0.770 & 80 & 0.535 & 0.068 & 6430 & 2435 & 0\\ GMRTEN1~J160351.1$+$555644 & 16:03:51.13 & $+$55:56:44.4 & 3.996 & 120 & 7.391 & 0.287 & 6321 & 6581 & 0\\ GMRTEN1~J160351.2$+$552346 & 16:03:51.24 & $+$55:23:46.1 & 0.519 & 80 & 0.640 & 0.072 & 6354 & 5262 & 0\\ GMRTEN1~J160351.3$+$540609 & 16:03:51.38 & $+$54:06:09.9 & 0.951 & 110 & 0.843 & 0.110 & 6432 & 2159 & 0\\ GMRTEN1~J160352.6$+$550012 & 16:03:52.62 & $+$55:00:12.8 & 0.510 & 71 & 0.592 & 0.076 & 6370 & 4320 & 0\\ GMRTEN1~J160352.8$+$542943 & 16:03:52.80 & $+$54:29:43.5 & 2.258 & 82 & 2.422 & 0.124 & 6400 & 3101 & 0\\ GMRTEN1~J160353.2$+$550309 & 16:03:53.29 & $+$55:03:09.9 & 0.570 & 74 & 0.500 & 0.072 & 6363 & 4438 & 0\\ GMRTEN1~J160355.5$+$553844 & 16:03:55.52 & $+$55:38:44.9 & 0.463 & 73 & 0.242 & 0.050 & 6314 & 5861 & 0\\ GMRTEN1~J160356.2$+$550126 & 16:03:56.20 & $+$55:01:26.7 & 0.472 & 77 & 0.380 & 0.063 & 6348 & 4369 & 0\\ GMRTEN1~J160356.2$+$550415 & 16:03:56.25 & $+$55:04:15.3 & 0.506 & 83 & 0.451 & 0.068 & 6345 & 4481 & 0\\ \hline \end{tabular} \end{table*} \subsection{Comparison with other surveys} In order to test the positional accuracy of our catalogue, we paired it with the 393 objects in the VLA 1.4~GHz source catalogue of \citet{Ciliegi99}, using a pairing radius of 6~arcsec. The VLA survey covers only $\sim25$~per~cent of our 610~MHz observations. Fig.~\ref{fig:deltaRADEC} shows the position offset of the 263 matched sources compared with their VLA counterparts -- the distribution of offsets is approximately Gaussian, with mean offset in Right Ascension of $-0.9$~arcsec, and $-0.1$~arcsec in Declination. The standard deviations of the distribution are 0.9 and 0.7~arcsec respectively. \begin{figure} \includegraphics[width=8cm]{deltaRADEC.ps} \caption{Source positions in the GMRT catalogue relative to the positions found in the VLA catalogue of \citet{Ciliegi99}, for unique matches within 6~arcsec. Offsets are distributed in a Gaussian fashion in RA and DEC, and the ellipse corresponding to 1$\sigma$ for the distribution is shown.} \label{fig:deltaRADEC} \end{figure} We repeated this analysis with data from the FIRST survey \citep{Becker95}. The whole of our 610~MHz survey region is covered by FIRST, and even with the reduced sensitivity of FIRST ($\sim$150~$\mu$Jy noise), 504 pairs are found within 6~arcsec. We again find Gaussian-distributed position errors, with RA offset of $-0.1$~arcsec, standard deviation 0.4~arcsec and DEC offset of $-0.1$~arcsec, standard deviation 0.6~arcsec. We have not corrected the positions given in our catalogue, since it agrees closely with FIRST. The spectral index distribution of the matched sources from both surveys is shown in Fig.~\ref{fig:alpha}, using the integrated flux density measurements. The distribution peaks around $\alpha=0.7$, where $\alpha$ is defined such that the flux density $S$ scales with frequency $\nu$ as $S = S_{0}\nu^{-\alpha}$. \begin{figure} \includegraphics[width=8cm]{Alpha.eps} \caption{Radio spectral index $\alpha$ between 610~MHz and 1.4~GHz, for sources in the VLA ELAIS-N1 catalogue of Ciliegi et al.\ (solid lines) and in the FIRST catalogue (dashed lines).} \label{fig:alpha} \end{figure} Fig.~\ref{fig:AlphaFlux} shows the spectral index distribution for all sources with matches in the GMRT and FIRST catalogues (black diagonal crosses), and sources in the GMRT and \citet{Ciliegi99} catalogues (red upright crosses). There are significant biases in Fig.~\ref{fig:AlphaFlux}, due to the varying sensitivity levels of the three surveys. In the region covered by \citet{Ciliegi99}, the 610~MHz completeness level is 360~$\mu$Jy, shown by the black dotted line. The limiting spectral indices for sources at the sensitivity levels of the 1.4~GHz surveys are shown (FIRST -- solid black line, Ciliegi -- dashed red line). In order to look for variations in the source population, we calculate the mean and median spectral indices for sources with detections in the Ciliegi catalogue, with 610~MHz flux density between 500~$\mu$Jy and 1~mJy -- the point at which the turnover in source counts becomes visible (see Section~\ref{sec:sourcecounts}) -- and above 1~mJy. The mean values of $\alpha$ are $0.22\pm0.09$ and $0.45\pm0.04$ respectively, and the median values are $0.36\pm0.12$ and $0.56\pm0.04$. There are 48 and 168 sources in the two flux density bins. The bias against steep-spectrum sources at low flux densities (which is visible in Fig.~\ref{fig:AlphaFlux}) means that, for a source with 610~MHz flux density of 500~$\mu$Jy, the largest value of $\alpha$ that would be detectable is $\sim$0.8 and so this apparent flattening at fainter flux densities may simply be due to sample bias. However, \citet{Bondi07} also find significantly flatter spectral indices for fainter radio sources, again comparing 610~MHz and 1.4~GHz data, and attribute this to the emergence of a population of low-luminosity AGNs -- see Section~\ref{sec:sourcecounts} for more details. \begin{figure} \includegraphics[width=8cm]{AlphaFlux.eps} \caption{The variation in spectral index $\alpha$ with 610~MHz flux density. Black diagonal crosses represent sources in the GMRT and FIRST catalogues, with the solid black line showing the limiting spectral index that could be detected, given the respective sensitivity levels. Red upright crosses represent sources in the corresponding GMRT and Ciliegi et al.\ catalogues, with the dashed red line showing the limit on $\alpha$. The 610~MHz flux density limit is shown by the dotted black line.} \label{fig:AlphaFlux} \end{figure} | 7 | 10 | 0710.1500 |
0710 | 0710.0506_arXiv.txt | Using a combination of deep MID-IR observations obtained by IRAC, MIPS and IRS on board Spitzer we investigate the MID-IR properties of Lyman Break Galaxies (LBGs) at z$\sim$3, establish a better understanding of their nature and attempt a complete characterisation of the population. With deep mid-infrared and optical observations of $\sim$1000 LBGs covered by IRAC/MIPS and from the ground respectively, we extend the spectral energy distributions (SEDs) of the LBGs to mid-infrared. Spitzer data reveal for the first time that the mid-infrared properties of the population are inhomogeneous ranging from those with marginal IRAC detections to those with bright rest-frame near-infrared colors and those detected at 24$\mu$m MIPS band revealing the newly discovered population of the Infrared Luminous Lyman Break Galaxies (ILLBGs). To investigate this diversity, we examine the photometric properties of the population and we use stellar population synthesis models to probe the stellar content of these galaxies. We find that a fraction of LBGs have very red colors and large estimated stellar masses $M_{\ast}$$>$5$\times$$10^{10}$$M_{\odot}$. We discuss the link between these LBGs and submm-luminous galaxies and we report the detection of rest frame 6.2 and 7.7 $\mu$m emission features arising from Polycyclic Aromatic Hydrocarbons (PAH) in the Spitzer/IRS spectrum of an infrared-luminous Lyman break galaxy at z=3.01. | Observation and study of high-redshift galaxies is essential to constrain the history of galaxy evolution and give us a systematic and quantitative picture of galaxies in the early universe, an epoch of rigorous star and galaxy formation. Large samples of high-z galaxies that have recently become available, play a key role to that direction and have revealed a zoo of different galaxy populations at z. There are various techniques for detecting high-z galaxies involving observations in wavelengths that span from optical to far-IR. Among the various methods the Lyman break dropout technique (\cite{Steidel93}), sensitive to the presence of the 912{\AA} break, is designed to select z$\sim$ 3 galaxies. LBGs constitute at the moment the largest galaxy population at z$\sim$3 (\cite{Steidel03}). With observations spanning from X-rays (eg. \cite{Nan02}) to near-infrared (\cite{Shapley03}) there has been a considerable progress into understanding the nature of population, but to fully characterize their properties (such as stellar mass, dust content, link to other z$\sim$3 populations) observations of longer wavelengths are required. With the advent of Spitzer Space Telescope (\cite{Werner04}) we have access to longer wavelengths. IRAC bands (3.6, 4.5, 5.8, 8.0$\mu$m) are crucial as they trace the rest-frame near infrared luminosities for galaxies at 0.5$<$z$<$5, (where the bulk of the stellar mass of a galaxy radiates) while MIPS (24, 70, 160 $\mu$m) and IRS (5.3--40$\mu$m) provide an insight into the interstellar medium of the population as they are sensitive to PAH features and dust re-radiation. In this study we use IRAC and MIPS data covering $\sim$1000 and 244 LBGs respectively, lying on the fields Q1422+2309 (Q1422), DSF2237a,b (DSF), Q2233+1341 (Q2233), SSA22a,b (SSA22), B20902+34 (B0902), QSOHS1700+6416 (Q1700), Extended Groth Strip (EGS) and Hubble Deep Field North (HDFN). Those LBGs have previously beeen identified from their optical colours by \cite{Steidel03}. In section 2 we search for mid-infrared counterparts of the LBGs, extend their SEDs to mid-infrared and investigate their mid-infrared colours as well as their physical properties such as stellar mass and dust content. In section 3 we examine the possible link between the IRAC/MIPS bright LBGs and the SMGs while in Section 4 using data obtained by IRS, we report the detection of PAH features arising from the mid-inrared spectrum of an ILLBG at z=3.01. In Section 5 we summurize the results of this Spitzer view of LBGs. | The advent of Spitzer has dramatically improved our understanding of the LBGs. Using data obtained by IRAC/MIPS/IRS on board Spitzer we have reached the following conclusions for the population of the LBGS: \begin{itemize} \item IRAC colors have revealed the diversity of LBGs ranging from those with marginal detection in IRAC bands and R-[3.6]$<$1.5 colors, to those that have bright in IRAC bands and exhibit R-[3.6]$>$1.5 colors. \item LBGs detected at 8$\mu$m have redder R-[3.6] colors and on average are more massive, suffer more obscuration and have relatively older stellar populations when compared to the rest of the population. \item A fraction of about $\sim$5\% of the LBGs do have dust as evidenced by MIPS 24$mu$m detections, and are classified as ILLBGs. Those LBGs share many properties in common with the SMGs and preliminary results show that they can be detected at submm bands. It can therefore be suggested that a link between these two populations must exist. \item Strong PAH features arising from the mid-infrared spectrum of an ILLBG at z=3.01 indicates the existence of dust in the interstellar medium of LBGs and suggest that the emission is dominated by star formation rather than an AGN. \end{itemize} \begin{figure} \centering \includegraphics[width=10cm,height=4cm]{magdis_fig2.eps} \caption {\small{IRS spectrum of EGS20 J1418+5236. The dashed line is the M82 SED shifted to $z=3.01$. The spectrum has remarkably similar 6.2 and 7.7 $\mu$m PAH emission-feature strength and shape to those of M82. We cross-correlated the IRS spectrum with that of M82 to derive a redshift of $z =3.01$$\pm$$0.016$.}} \end{figure} | 7 | 10 | 0710.0506 |
0710 | 0710.5609_arXiv.txt | {} {We present the results of a spectroscopic analysis on three young embedded sources (HH26 IRS, HH34 IRS and HH46 IRS) belonging to different star-forming regions and displaying well developed jet structures. The aim is to investigate the source accretion and ejection properties and their connection.} {We used VLT-ISAAC near-IR medium resolution ($R\sim9000$) spectra ($H$ and $K$ bands) to derive, in a self-consistent way, parameters like the star luminosity, the accretion luminosity and the mass accretion rate. Mass ejection rates have also been estimated from the analysis of different emission features.} {The spectra present several emission lines but no photospheric features in absorption, indicating a large veiling in both $H$ and $K$ bands. In addition to features commonly observed in jet driving sources ([\ion{Fe}{ii}], H$_2$, \ion{H}{i}, CO), we detect a number of emission lines due to permitted atomic transitions, such as \ion{Na}{i} and \ion{Ti}{i} that are only 2-5 times weaker than the Br$\gamma$ line. Some of these features remain unidentified. Emission from \ion{Na}{i} 2.2$\mu$m doublet is observed along with CO(2-0) band-head emission, indicating a common origin in an inner gaseous disc heated by accretion. We find that accretion provides about 50\% and 80\% of the bolometric luminosity in HH26 IRS and HH34 IRS, as expected for accreting young objects. Mass accretion and loss rates spanning $10^{-6}$--$10^{-8}$ M$_{\sun}$\,yr$^{-1}$ have been measured. The derived $\dot{M}_\mathrm{loss}/\dot{M}_\mathrm{acc}$ is $\sim$0.01 for HH26 IRS and HH34 IRS, and $>$0.1 for HH46 IRS. These numbers are in the range of values predicted by MHD jet launching models and found in the most active classical T Tauri stars.} {Comparison with other spectroscopic studies performed on Class Is seems to indicate that Class Is actually having accretion-dominated luminosities are a limited number. Although the analysed sample is small, we can tentatively define some criteria to characterise such sources: they have $K$-band veiling larger than 2 and in the majority of the cases present IR features of CO and \ion{Na}{i} in emission, although these do not directly correlate with the accretion luminosity. Class Is with massive jets have high $L_{\mathrm{acc}}/L_{\mathrm{bol}}$ ratios but not all the identified accretion-dominated objects present a jet. As suggested by the SEDs of our three objects, the accretion-dominated objects could be in an evolutionary transition phase between Class 0 and I. Studies of the kind presented here but on larger samples of possible candidates should be performed in order to test and refine these criteria.} | The process of mass accretion accompanying the formation of solar type stars is always associated with mass ejection in form of collimated jets, that extend from few AU up to parsecs distance from the exciting source. According to an established class of models \citep{koenigl00,casse00}, accretion and ejection are indeed intimately related through the presence of a magnetised accretion disc: the jets carry away the excess angular momentum, so that part of the disc material can move toward the star. The efficiency of this process is measured by the ratio between the mass ejection and mass accretion rates, and depends on the jet acceleration mechanism at work. Measurements of such an efficiency have been so far obtained only for classical T Tauri stars, whose accretion properties are rather well studied through the observation of the excess continuum emission at optical and UV wavelengths Values of $\dot{M}_\mathrm{loss}/\dot{M}_\mathrm{acc}$ in the range 1-10\% have been found by different studies \citep[e.g.][]{hartigan95,woitas05,ferreira06}. It is nevertheless important to test accretion/ejection models in young sources at earlier stages of evolution, when accretion dominates the energetics of the system and thus the mechanism to extract angular momentum is expected to be more efficient. To this aim, an interesting sample of objects are Class I sources, i.e. the class of embedded stars characterised by a steeply rising IR spectral energy distribution (SED) between 2 and 10 $\mu$m and usually considered younger than visible T Tauri stars (the Class II sources). However, the high extinction pertaining to these objects strongly limits the measurement of their stellar and accretion properties, needed to prove that they are indeed in a phase of higher accretion with respect to Class II sources. The general assumption so far applied has been that most of the bolometric luminosity of Class I objects is due to accretion. Recently, however, thanks to the use of high dispersion sensitive instrumentation, it has become possible to define the stellar properties of small samples of Class I stars through their weak photospheric lines detected in the optical scattered light \citep{white04} and in the near-IR direct emission \citep{greene02,nisini05a,doppmann05}. Such studies have shown that the characteristics of Class I objects vary in fact significantly. In particular, the accretion luminosity may span from few percent up to 80\% of the bolometric luminosity; these findings show that not all sources defined as Class I are indeed actively accreting objects and suggest that a classification based on different criteria is indeed required. \citet{white04}, notably, derived from their analysis of the scattered light spectra of Taurus-Auriga sources, that the average $\dot{M}_\mathrm{loss}/\dot{M}_\mathrm{acc}$ for Class I of their sample is larger than for Class II and close to unity. They interpret this result as due to an observational bias induced by the effect of disc orientation in their optical spectra; if the Class I are seen prevalently edge-on, then the extended region emitting the forbidden lines from which they estimate $\dot{M}_\mathrm{loss}$ is seen more directly than the obscured stellar photosphere. It is clear that such kind of biases can be minimised by performing spectroscopic observations directly in the IR, where features originating in the photosphere, in the accretion region and in the jet can be simultaneously detected by instrumentation which is sensitive enough. In this framework, we report here the results of near-IR spectroscopic observations at medium resolution of three embedded sources (HH34 IRS, HH26 IRS and HH46 IRS) displaying well developed jet structures and having a spectral index between 2 and 10\,$\mu$m, typical of Class I objects. We have derived accretion and ejection parameters of these sources through the analysis of the different features detected on the spectra. The goal is to study how much of their energy is due to accretion and to investigate the efficiency of the ejection mechanism. We describe the sample and the observations in Sec. 2 and report the results in Sec. 3; in Sec. 4 we present the procedures applied for the analysis of the data and infer the physical properties of the objects and their jets. These results are then discussed in Sec. 5, where a comparison with similar sources analysed in previous studies will be also made. Main conclusions of our work are summarised in Sec. 6. \begin{table*}[!t] \begin{center} \caption[]{The observed targets and their main observational properties.} \begin{tabular}{l c c | c | c c | c c | c c c c} \hline \hline Source & R.A.(2000) &DEC.(2000) &$m_\mathrm{J}$& \multicolumn{2}{|c|}{$m_\mathrm{H}$} & \multicolumn{2}{c|}{$m_\mathrm{K}$} & $D$ & $\alpha^{(a)}$ & $L_\mathrm{bol}$ \\ & & &(mag) &\multicolumn{2}{|c|}{(mag)}&\multicolumn{2}{|c|}{(mag)}&(pc) & & (L$_{\sun}$) \\ & & &2MASS$^{(1)}$&2MASS$^{(1)}$&This work$^{(1)}$&2MASS$^{(b)}$&This work$^{(b)}$& & & \\ \hline HH 26 IRS & 05 46 03.9 & -00 14 52 &16.77& 14.07 &14.6 & 11.88 & 12.3 & 450 & 2.01 & 4.6--9.2 \\ HH 34 IRS & 05 35 29.9 & -06 26 58 &15.06& 13.60 &13.5 & 12.38 & 12.4 & 460 & 1.14 & 12.4--19.9 \\ HH 46 IRS & 08 25 43.9 & -51 00 36 &14.20& 12.88 &14.8 & 12.72 & 13.4 & 450 & 1.96 & $<$15.0$^{(c)}$ \\ \hline \end{tabular} \label{targets} \end{center} \vspace{0.2 cm} \small{Notes. (a) The spectral index $\alpha=d\mathrm{Log}(\lambda F_{\lambda})/d\mathrm{Log}(\lambda)$ is calculated between 2 and 10 $\mu$m. (b) The magnitude values in the $H$ and $J$ bands have been estimated from the calibrated spectra. (c) Total luminosity: the source is a binary (see text for details). References. (1) from the 2MASS catalogue \citep{2mass}.} \end{table*} | We have investigated the accretion and ejection properties of three embedded sources (HH26 IRS, HH34 IRS, HH46 IRS) showing prominent jet-like structures. To this aim we have analysed their medium resolution (R $\sim$ 9000) near IR spectra acquired with VLT-ISAAC. The main results we obtained can be summarised as follows: \begin{itemize} \item The bolometric luminosity and SEDs of the three sources have been revised on the basis of Spitzer and recent sub-mm observations, and theoretical radiative transfer models available in the literature. It turns out that the sources are probably very young, in a transition phase between Class 0 and I. \item The spectra of the three sources show important differences in their characteristics: in fact, the number and the absolute and relative intensity of the observed emission features (associated both with the accretion region and the jet environment) vary among the sources. In particular, we point out that there is no clear relationship between the presence of the jet and the spectral accretion signatures detected. Moreover, the spectra show no sign of absorption features, indicating large amount of veiling which in turn suggests the presence of warm dusty envelopes around the sources, heated by the active ongoing accretion. \item In two of our sources (HH34 IRS and HH26 IRS) we find a Br$\gamma$/\ion{Na}{i} ratio much larger (by a factor of three or more) than that observed in other Class Is or T Tauri stars. Conversely, the ratio between the CO(2-0) 2.3$\mu$m overtone emission and the \ion{Na}{i} 2.20$\mu$m doublet, shows the opposite trend, i.e. is smaller in the objects with jets. This may indicate the presence of massive discs around the jet sources, characterised by large amounts of warm gas in neutral state. \item We consistently derive $A_K$, $L_*$ and $L_{\mathrm{acc}}$ assuming the three sources on the birthline and adopting the relationship between Br$\gamma$ luminosity and accretion luminosity derived by \citet{muzerolle98}. In the case of HH34 IRS and HH26 IRS, accretion largely contributes to the total energy, as expected for young sources in the main accretion phase. In HH46 IRS, we estimate an $L_{\mathrm{acc}}/L_{\mathrm{bol}}$ ratio of the order of 0.2, but the real value largely depends on how the bolometric luminosity is shared among the two binary components. Taking into account $K$-band veiling values $r_K$ greater than the lower limits inferred from the spectra would lead in general to higher $A_K$, $L_{\mathrm{acc}}$ and $\dot{M}_{acc}$. \item Mass accretion and loss rates span (including errors) in the range $10^{-8}$--$10^{-6}$ M$_{\sun}$\,yr$^{-1}$. The derived $\dot{M}_{loss}/\dot{M}_{acc}$ ratio is $\sim$0.01-0.03 for HH26 IRS and HH34 IRS, and $>$0.1 for HH46 IRS. These numbers are in the range of values usually predicted by models and found in the more active classical T Tauri stars. \item Comparing the results found in this work with other spectroscopic studies recently performed on Class I sources, we conclude that the number of Class Is actually having accretion-dominated luminosities (Accretion-Dominated Young Objects, ADYOs) could be limited. From the properties inferred in a small sample of objects we can tentatively define some criteria to characterise such sources: ADYOs have all $K$-band veiling larger than 2 and in the majority of the cases present (in addition to \ion{H}{i}) IR features of CO and \ion{Na}{i} in emission, although these latter do not directly correlate with the accretion luminosity. Class Is with massive jets have high $L_{\mathrm{acc}}/L_{\mathrm{bol}}$ ratios but not all the identified ADYOs present a jet. The SEDs of our small sample of three objects, suggest that accretion-dominated sources could be in an evolutionary phase in transition between Class 0 and I. Of course, studies of the kind presented here but carried out on larger samples of possible candidates should be performed in order to test and refine the tentative criteria that we have just mentioned. \end{itemize} | 7 | 10 | 0710.5609 |
0710 | 0710.0440_arXiv.txt | We examine the ability of a future X-ray observatory, with capabilities similar to those planned for the Constellation-X or X-ray Evolving Universe Spectroscopy (XEUS) missions, to constrain dark energy via measurements of the cluster X-ray gas mass fraction, $f_{\rm gas}$. We find that $f_{\rm gas}$ measurements for a sample of $\sim500$ hot ($kT\gsim5$keV), X-ray bright, dynamically relaxed clusters, to a precision of $\sim 5$ per cent, can be used to constrain dark energy with a Dark Energy Task Force (DETF; Albrecht et al. 2006) figure of merit of $15-40$, with the possibility of boosting these values by 40 per cent or more by optimizing the redshift distribution of target clusters. Such constraints are comparable to those predicted by the DETF for other leading, planned `Stage IV' dark energy experiments. A future $f_{\rm gas}$ experiment will be preceded by a large X-ray or SZ survey that will find hot, X-ray luminous clusters out to high redshifts. Short `snapshot' observations with the new X-ray observatory should then be able to identify a sample of $\sim 500$ suitably relaxed systems. The redshift, temperature and X-ray luminosity range of interest has already been partially probed by existing X-ray cluster surveys which allow reasonable estimates of the fraction of clusters that will be suitably relaxed for $f_{\rm gas}$ work to be made; these surveys also show that X-ray flux contamination from point sources is likely to be small for the majority of the targets of interest. Our analysis uses a Markov Chain Monte Carlo method which fully captures the relevant degeneracies between parameters and facilitates the incorporation of priors and systematic uncertainties in the analysis. We explore the effects of such uncertainties for scenarios ranging from optimistic to pessimistic. We conclude that the $f_{\rm gas}$ experiment offers a competitive and complementary approach to the best other large, planned dark energy experiments. In particular, the $f_{\rm gas}$ experiment will provide tight constraints on the mean matter and dark energy densities, with a peak sensitivity for dark energy work at redshifts midway between those of supernovae and baryon acoustic oscillation/weak lensing/cluster number counts experiments. In combination, these experiments should enable a precise measurement of the evolution of dark energy. | \label{introduction} In the early 1990s, measurements of the baryonic mass fraction in X-ray luminous galaxy clusters provided compelling evidence that we live in a low density Universe. Under the assumption that large clusters provide approximately fair samples of the matter content of the Universe, X-ray observations require that the mean matter density, $\Omega_{\rm m}$, is significantly less than the critical value, with a best-fit value $\Omega_{\rm m}\sim 0.2-0.3$ \citep[e.g.][]{White:91,Fabian:91, Briel:92, White:93, David:95, White:95, Evrard:97, Mohr:99, Ettori:99, Roussel:00, Grego:01, Allen:02, Allen:04, Allen:07, Ettori:03, Sanderson:03, Lin:03, LaRoque:06}. When combined with the expectation from inflation models, later confirmed by Cosmic Microwave Background (CMB) studies \citep[][and references therein]{Bennett:03, Spergel:03}, that the Universe should be close to spatially flat, X-ray results on the cluster baryon mass fraction quickly lead to the suggestion that the mass-energy density of the Universe may be dominated by a cosmological constant \citep[e.g.][]{White:93}. The first direct evidence for late-time cosmic acceleration, as would be produced by a sizeable cosmological constant, was provided in the late 1990s by \cite{Riess:98} and \cite{Perlmutter:98} based on measurements of the light curves of type Ia supernovae (SNIa). Since then, larger SNIa data sets \citep{knop:03, Riess:04, Astier:06, Riess:07, WoodVasey:07, Davis:07} and an increasingly wide array of other, complementary experiments have confirmed and improved upon this striking measurement. The combination of CMB data from the Wilkinson Microwave Anisotropy Probe (WMAP) \citep{Spergel:03, Spergel:06, Dunkley:08} with large scale structure (LSS) data from the Sloan Digital Sky Survey (SDSS) \citep{Eisenstein:05, Percival:07} and/or 2dF Galaxy Redshift Survey (2dFGRS) \citep{Cole:05} provides powerful evidence for dark energy. The cross-correlation of CMB and LSS fluctuations reveals the effects of dark energy on the Integrated Sachs-Wolfe effect \citep{Scranton:03, Fosalba:03, Rassat:07}. Measurements of the amplitude and evolution of matter fluctuations using X-ray galaxy clusters \citep{Borgani:01,Reiprich:02, Allen:03, Schuecker:03, Voevodkin:04, Henry:04, Mantz:07}, optically-selected clusters \citep{Gladders:06, Rozo:07}, Lyman-$\alpha$ forest data \citep{Viel:04, Seljak:04}, and weak lensing \citep{vanWaerbeke:05, Jarvis:05, Hoekstra:06, Benjamin:07}, also provide important, powerful confirmation of the new, standard cosmological paradigm: a universe in which the main mass and energy components are dark matter and dark energy, and where dark energy drives the current acceleration. The standard model for dark energy remains the cosmological constant, which is mathematically equivalent to vacuum energy. In principle, however, cosmic acceleration could be driven by either dark energy or a modification to the laws of gravity on cosmological scales \citep[see][for an extensive review]{Copeland:06}. Building on the early X-ray work, \cite{Allen:04, Rapetti:05}; and \cite{Allen:07} showed that measurements of the evolution of the X-ray gas mass fraction, $f_{\rm gas}$, in the largest, dynamically relaxed galaxy clusters provides a further powerful, complementary approach for studying dark energy. As with SNIa data, $f_{\rm gas}(z)$ measurements probe the redshift-distance relation; whereas the peak SNIa luminosity varies as the square of the distance, $f_{\rm gas}$ measurements vary as distance, $d^{1.5}$. \citep[The distance dependance derives from the way in which $f_{\rm gas}$ values are determined from the observed X-ray temperature and surface brightness data;][]{Allen:07} In combination with the tight constraint on $\Omega_{\rm m}$ provided by the normalization of the $f_{\rm gas}(z)$ curve, under the assumption of fair matter samples, the $f_{\rm gas}(z)$ data contain sufficient information to break the degeneracy between $\Omega_{\rm m}$ and the dark energy equation of state, $w$, in the distance equations. The additional combination of $f_{\rm gas}$ and CMB data breaks other important degeneracies between parameters in cosmological analyses \citep{Rapetti:05, Allen:07}. \cite{Allen:07} show that the current constraints on dark energy from the $f_{\rm gas}$ experiment are of comparable precision to other leading techniques, and are robust under the inclusion of conservative systematic allowances, e.g. relaxing the requirement for exact hydrostatic equilibrium and allowing for moderate redshift evolution in the cluster baryon fraction. These authors also show that intrinsic, systematic scatter remains undetected in the current $f_{\rm gas}$ data, despite a weighted mean statistical scatter in the individual distance measurements of only $\sim 5$ per cent; in contrast, SNIa studies \citep{Riess:07, Jha:07, WoodVasey:07} have established the presence of systematic scatter at the $\sim 7$ per cent in distance measurements from the best current SNIa data. The key to determining the nature of dark energy is to obtain precise measurements of its evolution with redshift, $z$, or scale factor, $a=1/(1+z)$. The Dark Energy Task Force report \cite[][hereafter DETF]{Albrecht:06} presented estimates of the constraints on dark energy parameters that should be achievable with a number of future proposed or planned dark energy experiments. In particular, the report forecasted the ability of these experiments, in combination with CMB data from the Planck satellite, to constrain a dark energy model of the form $w(a)=w_{\rm 0}+w_{\rm a}(1-a)$, and defined a figure of merit (hereafter FoM) to allow for easy comparison of the constraints. In this paper, we use the same dark energy parameterization and FoM to quantify the constraining power of future $f_{\rm gas}$ experiments, to be carried out with e.g. the Constellation-X or X-ray Evolving Universe Spectroscopy (XEUS) missions, in combination with CMB data. We show that the $f_{\rm gas}$ experiment is likely to provide comparable constraining power to the best other, contemporary space and ground-based experiments described by the DETF. When combined, future CMB, SNIa, baryon acoustic oscillation (BAO), weak lensing, cluster number count and $f_{\rm gas}$ experiments should provide precise, accurate constraints on $w(z)$ and allow significant progress in understanding the origin of cosmic acceleration. The structure of this paper is as follows: in Section~\ref{sec:de} we define the dark energy model and the FoM. In Section~\ref{sec:simdata} we describe the simulated $f_{\rm gas}$ and CMB data sets. For the $f_{\rm gas}$ data, we assume instrument characteristics appropriate for the baseline Constellation-X mission. The CMB data set approximates that expected from two years of Planck data. We also simulate a data set representative of that produced by follow-up observations of the Sunyaev-Zel'dovich effect in the clusters targeted for the $f_{\rm gas}$ work. Section~\ref{analysis} describes the Markov Chain Monte Carlo (MCMC) pipeline and details of the analysis method. Our main results are presented in Section~\ref{constraints}. Section~\ref{conclusions} summarizes our conclusions. | \label{conclusions} \begin{figure} \includegraphics[width=3.2in]{Prospects_w0waP5_new.ps} \caption{The 68 and 95 per cent confidence contours in the $w_{\rm 0}-w_{\rm a}$ plane determined from the $f_{\rm gas}$+CMB data (black, dashed contours) using the default dark energy model and $5$ per cent systematic allowances. The solid red lines show the constraints obtained from the $f_{\rm gas}$ data alone, using priors on $\Omega_{\rm b}h^2$, $\Omega_{\rm dm}h^2$, $l_{\rm a}$ and $h$, as described in the text (Section~\ref{cmbede}). The blue, dotted lines show the constraints from the $f_{\rm gas}$ alone using priors on $\Omega_{\rm b}h^2$ and $h$ and assuming flatness. The figure shows how the CMB data contribute in constraining dark energy, especially at early times.} \label{fig:cmbede} \end{figure} We have examined the ability of a future X-ray observatory, with capabilities similar to those planned for Constellation-X, to constrain dark energy via the $f_{\rm gas}$ experiment. We find that $f_{\rm gas}$ measurements for a sample of 500 hot ($kT_{2500}\gsim5$keV), X-ray bright, dynamically relaxed clusters, with a precision of $\sim 5$ per cent, can be used to constrain dark energy with a FoM of $15-40$. These constraints are comparable to those predicted by the DETF \citep{Albrecht:06} for other leading, planned (DETF Stage IV) dark energy experiments. We also find that, for the $f_{\rm gas}$ experiment, the FoM can be boosted up by at least $\sim 40$ per cent by selecting an optimal redshift distribution of suitable clusters on which to carry out the $f_{\rm gas}$ observations. Interestingly, the optimal redshift distribution of $f_{\rm gas}$ measurments appears to be shifted towards low redshifts. As discussed in the text, a future $f_{\rm gas}$ experiment will need to be preceded by a large X-ray or SZ cluster survey that will find hot, X-ray luminous clusters out to high redshifts. A survey such as that planned with the Spectrum-RG/eROSITA mission should find several thousand of such clusters. Short `snapshot' follow-up observations of the clusters with a new, large X-ray observatory should be able to identify a sample of $\sim 500$ suitable systems for $f_{\rm gas}$ work. Attaining a precision of $\sim 5$ per cent with individual $f_{\rm gas}$ measurements should be straightforward for an observatory with characteristics similar to Constellation-X, requiring exposure times of $\sim 20$ks on average. We note that the population of galaxy clusters in the redshift, temperature and X-ray luminosity range of interest has already been partially probed by the MACS survey \citep{Ebeling:01}; Chandra observations of MACS clusters are used extensively in current $f_{\rm gas}$ studies \citep{Allen:04,LaRoque:06,Allen:07}. The low-level of X-ray flux contamination from point sources observed in MACS clusters also alleviates the requirements on the instrumental PSF for dark energy work via the $f_{\rm gas}$ method. In determining the predicted dark energy constraints, we have employed the same MCMC method used to analyze current data. The MCMC method encapsulates all of the relevant degeneracies between parameters and allows one to easily and efficiently incorporate priors and allowances in the analysis. We have included an array of such systematic allowances, with tolerances ranging from optimistic to pessimistic. Our technique differs from the DETF \citep{Albrecht:06}, who use a simpler Fisher matrix approach in the prediction of dark energy constraints. Despite these differences, we have endeavored to make our calculations of the FoM (Section 2) as comparable as possible. Benchmarking our results against those of the DETF for other, future `Stage IV' dark energy experiments i.e. large, long-term missions, we find that the $f_{\rm gas}$ experiment should provide a comparable FoM to future ground-based SNIa (FoM=$8-22$), space-based SNIa (FoM=$19-27$), ground-based BAO (FoM=$5-55$), space-based BAO (FoM=$20-42$) and space-based cluster counting (FoM=$6-39$) experiments. Formally, the predicted FoM for the $f_{\rm gas}$ experiment is comparable to `pessimistic' scenarios for weak lensing experiments discussed by \cite{Albrecht:06}, although the value falls short of the most optimistic DETF weak lensing predictions. The tight constraints on $\Omega_{\rm m}$ and $\Omega_{\rm de}$ for the $f_{\rm gas}$ experiment will be of importance when used in combination with other techniques. Interestingly, the `pivot point' for the $f_{\rm gas}$ experiment lies between those of the SNIa and BAO/weak lensing/cluster number count experiments, offering excellent redshift coverage in attempts to pin down the evolution of dark energy. We conclude that the $f_{\rm gas}$ experiment offers a powerful approach for dark energy work, which should be competitive with and complementary to the best other planned dark energy experiments. | 7 | 10 | 0710.0440 |
0710 | 0710.2997_arXiv.txt | {Gas-rich dwarf galaxies are probably the closest counterparts to primeval objects we can find in the local Universe, therefore it is interesting to study their evolution in different astrophysical contexts.} {We study the effects of interstellar clouds on the dynamical and chemical evolution of gas-rich dwarf galaxies. In particular, we focus on two model galaxies similar to IZw18 and NGC1569 in comparison to models in which a smooth initial distribution of gas is assumed.} {We use a 2-D hydrodynamical code coupled with a series of routines able to trace the chemical products of SNeII, SNeIa and intermediate-mass stars. Clouds are simulated by adding overdense regions in the computational grid, whose locations are chosen randomly and whose density profiles match observed ones. We consider both cloud complexes put at the beginning of the simulation and a mechanism for continuous cloud formation. The clouds are inherently dynamically coupled to the diffuse gas, and they experience heat conduction from a hot surrounding gas.} {Due to dynamical processes and thermal evaporation, the clouds survive only a few tens of Myr. Due to the additional cooling agent, the internal energy of cloudy models is typically reduced by 20 -- 40\% compared with models of diffuse gas alone. The clouds delay the development of large-scale outflows by mass loading, therefore helping to retain a larger amount of gas inside the galaxy. However, especially in models with continuous creation of infalling clouds, their bullet effect can pierce the expanding supershell and create holes through which the superbubble can vent freshly produced metals. Moreover, assuming a pristine chemical composition for the clouds, their interaction with the superbubble dilutes the gas, reducing the metallicity. The resulting final metallicity is therefore generally lower (by $\sim$ 0.2 -- 0.4 dex) than the one attained by diffuse models. } {} | \label{intro} Gas-rich dwarf galaxies are commonly classified into low surface-brightness dwarfs, called dwarf irregulars (dIrrs), and higher surface-brightness objects, usually called blue compact dwarf (BCD) galaxies. These classes of galaxies tend to have low metallicities, blue colors and complex and chaotic gas phases. A large fraction of these galaxies shows an ongoing star formation (SF) or at least hints that this process has been quenched in the recent past. In this case, these objects are commonly referred to as {\it starburst} galaxies and their gas consumption timescales are much shorter than the Hubble time (Kennicutt \cite{ken98}), making this a transient phase of their evolution. Owing to the energy released by stellar winds and supernovae (SNe), intense episodes of SF are also associated to the development of galactic winds or at least of large-scale outflows. The broad distinction between these two phenomena is the final fate of the outwards-directed flow of gas: galactic winds generally exceed the escape velocity while outflows do not, therefore they tend to recede towards the center of the galaxy. Clear signatures of outflows are present in NGC1705 (Hensler et al. \cite{hen98}; Heckman et al. \cite{hek01}), NGC1569 (Martin, Kobulnicky \& Heckman \cite{mkh02}), NGC3079 (Cecil et al. \cite{cec01}), IZw18 (Martin \cite{m96}), NGC3628 (Irwin \& Sofue \cite{is96}) among others. Perhaps the best examples of large-scale outflows driven by SN feedback are at large redshifts (Pettini et al. \cite{pet98}; Pettini et al. \cite{pet01}). Although it is not certain, in any of the above-mentioned objects, that the metals will definitely leave the parent galaxy, indirect hints of the ubiquity of galactic winds are given by the mass-metallicity relation (Tremonti et al. \cite{tre04}; Dave\'e, Finlator \& Oppenheimer \cite{dfo06}) and effective yields (Garnett \cite{gar02}). From a theoretical point of view, the study of the evolution of gas-rich dwarf galaxies through numerical simulations has been performed by several authors in the recent past. The overall picture is that the occurrence of large-scale outflows is initially driven by the thermal pressure of a very hot, high pressurized gas and is favored by a flat distribution of the interstellar medium (ISM), which allows an easy vertical transport of material. However, since the transport of gas along the disk is very limited, outflows are not able to eject a significant fraction of the ISM, whereas the fraction of ejected metals can be very large (D'Ercole \& Brighenti \cite{db99}; MacLow \& Ferrara \cite{mf99}; Recchi, Matteucci \& D'Ercole \cite{rmd}, hereafter RMD). For NGC1569, Martin et al. (\cite{mkh02}) derived a supersolar metal content in the galactic wind from X-ray spectra but also advocated mass-loading of it with the ISM. Most of these studies, however, have focused on flows in homogeneous media, neglecting the multiphase nature of the ISM, although several attempts to perform multiphase hydrodynamical simulations have been made in the past, particularly using the so-called {\it chemodynamical} approach (Theis, Burkert \& Hensler \cite{tbh92}; Rieschick \& Hensler \cite{rh00}; Hensler, Theis \& Gallagher \cite{hen04}). The multiphase nature of the ISM, in particular its clumpiness, is observationally well established in dwarf galaxies (Cecil et al. \cite{cec01}; Cannon et al. \cite{cann05}; Leroy et al. \cite{leroy06}) and it has a solid theoretical background with the seminal work of McKee \& Ostriker (\cite{mo77}). According to this model, the ISM is composed by a cold neutral phase (representing the cores of molecular clouds), confined by a warm medium (with temperatures of the order of 10$^4$ K) and these two phases (which are in pressure equilibrium) are embedded in a hot, diluted intercloud medium (HIM), continuously produced by SN explosions and stellar winds. Sufficiently dense clouds can pierce the HIM without being swept up, so they can become embedded therein (Vieser \& Hensler \cite{vh07b}). At the interface between clouds and HIM, condensation-evaporation processes establish the final fate of the cloud and its impact on the development of a galactic wind. In two previous papers, we have studied the dynamical and chemical evolution of model galaxies similar to IZw18 (Recchi et al. \cite{rec04}, hereafter Paper I) and NGC1569 (Recchi et al. \cite{rec06}, hereafter Paper II). The main results can be briefly summarized as follows: \begin{itemize} \item most of the analyzed models develop large-scale outflows. These outflows carry out of the galaxy mostly the chemical elements freshly produced during the most recent episodes of SF, with large escape fraction of metals with delayed production (like Fe and N). \item Models with very short burst(s) of SF can cool and mix the newly formed metals in a very short timescale, whereas, when the SF is more complex, most of the metals are either directly ejected outside the galaxy through galactic winds or are confined in a too hot medium, therefore cannot contribute to the chemical enrichment of the warm ionized medium observed by emission lines from the \hii gas. \item Models with complex and long-lasting SF episodes reproduce the chemical composition and the abundance ratios of the above-mentioned galaxies much better than models with bursting SF. \end{itemize} In this paper we simulate models with structural parameters similar to IZw18 and NGC1569. We increase arbitrarily the gas density of some specific regions of the computational grid, in order to create a ``cloudy'' phase, and we address the question how and to which extent a ``cloudy gas phase'' alters the former results. The clouds possess a specific density profile and can be either added at the beginning of the simulation or continuously created during the evolution of the model. We then analyze the differences between the dynamical and chemical evolution of these models with the ones presented in Paper I and Paper II. We point out that, at variance with the above-mentioned works, in this paper we will not specifically look for the best initial setups and the best assumptions in order to reproduce chemical and dynamical features of well-known objects. We will just stress the main variations produced by a clumpy initial setup. For this reason, we will also consider models which failed in Paper I and II at reproducing the observations of IZw18 and NGC1569. The paper is organized as follows: in Sect.~\ref{cloud} we briefly recall the evolution of a cloud embedded in a hot medium; in Sect.~\ref{model} we present the model and the adopted assumptions in the simulations. Results are presented in Sect.~\ref{results_fix} (models with clouds fixed at the beginning of the simulation) and in Sect.~\ref{results_inf} (continuous creation of clouds). Finally, a discussion and some conclusions are drawn in Sect.~\ref{discussion}. | \label{discussion} In this paper we have computed the chemical and dynamical evolution of model galaxies, with structural parameters similar to IZw18 and NGC1569, but in which a complex of clouds has been added, both perturbing the initial gaseous distribution and creating clouds, at a rate which equals the SF rate, and with infall velocity of 10 km s$^{-1}$ along the polar direction. The main focus of our work has been the comparison of these models with those presented in previous publications, in which similar setups but a smooth distribution of gas was considered. We have seen that the clouds are subject to a variety of disruptive phenomena like evaporation (when embedded in a hot medium), formation of shocks, development of thermal instabilities (in particular the Kelvin-Helmholtz instability) and expansion due to the larger pressure compared to the surrounding interstellar medium. The average lifetime of the clouds is therefore relatively short, depending on the cloud size (which is not constant in our simulations) but being of the order of a few tens of Myr. In spite of their transient nature, the clouds leave a significant imprint on the dynamical and chemical evolution of dwarf galaxies. The clouds, when they evaporate inside the superbubble, produce mass loading, increase the mean density of the cavity and, therefore, enhance the radiative losses (which are proportional to the square of the density). This results in a significant decrease of the total thermal energy (of the order of $\sim$ 20 -- 40\% compared to the diffuse models, depending on the assumptions), therefore less energy to drive the development of a large-scale outflow. On the other hand, the relative motion of supershell and clouds, in particular when the clouds infall motion is considered, can structure, pierce and create holes and fingers in the expanding supershell. These holes destroy the spherical symmetry initially present and favor the rushing out of the highly pressurized gas contained in the cavity. Therefore, in spite of the reduced thermal energy budget, the creation of large-scale outflows is not suppressed but, in most of the explored cases, only slightly delayed. Complex structures and fingers are indeed relatively common features in galaxies showing large-scale outflows like NGC1800 (Hunter \cite{hun96}), NGC4214 (MacKenty et al. \cite{mack00}) or NGC1705 (Heckman et al. \cite{hek01}). The pressure inside the cavity is reduced compared to diffuse models, therefore in any case the total amount of ejected pristine gas is very small (smaller than in the models with smooth gas distribution) and, when averaging the size of the supershell in any direction, it turns out to be smaller than in diffuse models. But the piercing of the supershell can lead to an ejection efficiency of freshly produced metals as high as the one attained by diffuse models. This has, of course, important consequences on the chemical evolution of these objects. Since the differential winds are not suppressed, the diminished thermal energy of these models does not imply an increase of metals inside the galactic regions. On the other hand, the dilution effect of clouds plays a dominant role in determining the final metallicity of our model galaxies. Since the clouds have primordial chemical composition, their destruction and mixing with the surrounding medium reduces the total chemical composition without altering the abundance ratios. This produces a final metallicity $\sim$ 0.2 -- 0.4 dex smaller than the corresponding diffuse models. We have examined the effect of a different choice of the IMF slope and of the nucleosynthetic set of yields (in particular for what concerns intermediate-mass stars). Flatter-than-Salpeter IMF slopes lead to an excessive production of energy, able to unbind most of the gas before the end of the simulation. On the other hand, in models with steeper IMF the development of large-scale outflows is almost completely suppressed. Different sets of intermediate-mass stars yields affect in particular the log(N/O) ratio. Renzini \& Voli (\cite{rv81}) yields tend to overestimate the primary production of nitrogen. When compared to the results of models implementing van den Hoek \& Groenewegen (\cite{vg97}) yields, the results differ by $\sim$ 0.3 dex. Due to the assumption of a metallicity-dependent cooling function, also the dynamics is affected by the choice of the nucleosynthetic prescriptions. Our main results can be briefly summarized as follows: \begin{itemize} \item the clouds suffer thermal instabilities, formation of shocks and evaporation, therefore their lifetimes is limited to a few tens of Myr. \item In spite of that, they are able to increase the main density of the cavity, provoking a reduction of the total thermal energy by $\sim$ 20 -- 40\% compared with a diffuse model. \item The interaction clouds-supershell leads to strong structuring and piercing of the shell (in particular for models with continuous creation of infalling clouds), allowing the venting out of metals in spite of the reduced thermal energy. The development of large-scale outflows is therefore generally delayed but the ejection efficiency of metals remains unchanged. \item From a chemical point of view, the effect of the clouds is to significantly reduce the total metallicity of the galaxies, without altering the abundance ratios. \end{itemize} | 7 | 10 | 0710.2997 |
0710 | 0710.1285.txt | We present results of long-slit spectroscopy in several positions of the Orion nebula. Our goal is to study the spatial distribution of a large number of nebular quantities, including line fluxes, physical conditions and ionic abundances at a spatial resolution of about 1$''$. In particular, we have compared the O$^{++}$ abundance determined from collisionally excited and recombination lines in 671 individual 1D spectra covering different morphological zones of the nebula. We find that protoplanetary disks (proplyds) show prominent spikes of {\elect}([{\nii}]) probably produced by collisional deexcitation due to the high electron densities found in these objects. Herbig-Haro objects show also relatively high {\elect}([{\nii}]) but probably produced by local heating due to shocks. We also find that the spatial distribution of pure recombination {\oii} and {\foiii} lines is fairly similar, in contrast to that observed in planetary nebulae. The abundance discrepancy factor (ADF) of O$^{++}$ remains rather constant along the slit positions, except in some particular small areas of the nebula where this quantity reaches somewhat higher values, in particular at the location of the most conspicuous Herbig-Haro objects: HH 202, HH 203, and HH 204. There is also an apparent slight increase of the ADF in the inner 40$''$ around $\theta^1$ Ori C. We find a negative radial gradient of {\elect}([{\oiii}]) and {\elect}([{\nii}]) in the nebula based on the projected distance from $\theta^1$ Ori C. We explore the behavior of the ADF of O$^{++}$ with respect to other nebular quantities, finding that it seems to increase very slightly with the electron temperature. Finally, we estimate the value of the mean-square electron temperature fluctuation, the so-called {\ts} parameter. Our results indicate that the hypothetical thermal inhomogeneities --if they exist-- should be smaller than our spatial resolution element. | \footnotetext[1]{Based on observations made with the 4.2m William Herschel Telescope (WHT) operated on the island of La Palma by the Isaac Newton Group in the Spanish Observatorio del Roque de los Muchachos of the Instituto de Astrof\'\i sica de Canarias.} \label{intro} The analysis of the spectrum of {\hii} regions allows to determine the chemical composition of the ionized gas phase of the interstellar medium from the Solar Neighbourhood to the high-redshift Universe. Therefore, it stands as an essential tool for our knowledge of the chemical evolution of the Universe. In photoionized nebulae, the abundance of the elements heavier than He is usually determined from collisional excitation lines (hereinafter CELs), whose intensity depends exponentially on the electron temperature, {\elect}, of the gas. It was about 20 years ago when the first determinations of C$^{++}$ abundance from the intensity of the weak recombination line (hereinafter RL) of {\cii} 4267 \AA\ were available for planetary nebulae (PNe). The comparison of the abundance obtained from {\cii} 4267 \AA\ and from the CELs of this ion in the ultraviolet (UV) showed a difference that could be as large as a order of magnitude in some objects \citep[e.g.][]{french83, rolastasinska94, mathisliu99}. \citet{peimbertetal93} were the first in determinig the O$^{++}$ abundance from the very weak RLs, obtaining the same qualitative result: the abundances obtained from RLs are higher than those determined making use of CELs. This observational fact is currently known as the ``abundance discrepancy" (hereinafter AD) problem. In the last years, our group has obtained a large dataset of intermediate and high resolution spectroscopy of Galactic and extragalactic {\hii} regions using medium and large aperture telescopes \citep{estebanetal02, estebanetal05, garciarojasetal04, garciarojasetal05, garciarojasetal06, garciarojasetal06b, lopezsanchezetal06}. The general result of these works is that the O$^{++}$/H$^+$ ratio calculated from RLs is between 0.10 and 0.35 dex higher than the value obtained from CELs in the same objects. The value of the AD that we usually find in {\hii} regions is rather similar for all objects and ions and is much lower than the most extreme values found in PNe. The results for {\hii} regions obtained by our group are fairly different that those found for PNe, and seem to be consistent with the predictions of the temperature fluctuations paradigm formulated by \citet{peimbert67}, as it is argued in \citet{garciarojas06} and \citet{garciarojasesteban07}. In the presence of temperature fluctuations (parametrized by the mean square of the spatial variations of temperature, the so-called $t^{\rm 2}$ parameter) the AD can be naturally explained because the different temperature dependence of the intensity of RLs and CELs. The existence and the origin of temperature fluctuations are still controversial problems and a challenge for our understanding of ionized nebulae. Recently, \citet{tsamispequignot05} and \citet{stasinskaetal07} have proposed an hypothesis to the origin of the AD, which is based on the presence of cold high-metallicity clumps of supernova ejecta still not mixed with the ambient gas of the {\hii} regions. This cold gas would produce most of the emission of the RLs whereas the ambient gas of normal abundances would emit most of the intensity of CELs. Our group is interested in exploring on what variable or physical process the AD depends from different approaches. One of the most promising is based on the study of the behaviour of this magnitude at small spatial scales, something that has still not been explored in depth in nearby bright Galactic {\hii} regions. In this paper, we make use of deep intermediate-resolution long-slit spectroscopy of the Orion nebula to study the dependence of the AD with respect to different nebular parameters: electron temperature and density, local ionization state of the gas, presence of high velocity material, and its correlation with different morphological structures (proplyds, ionization fronts, globules, Herbig-Haro objects), in {\hii} regions. The spatial distribution of the physical conditions in the Orion nebula has been investigated by several authors. \citet{baldwinetal91} obtained the density and temperature distribution in 21 and 14 points, respectively, along a 5$'$ line west of $\theta^1$ Ori C, finding a density gradient that decreases to the outskirts of the nebula and a constant {\elect}. \citet{walteretal92} determined electron densities and temperatures and chemical abundances for 22 regions of the Orion nebula. Using also data from the literature, these authors find radial gradients of the physical conditions, but with a positive slope in the case of the temperature determined from {\foiii} lines. \citet{poggeetal92} obtained Fabry-Perot images of the inner 6$'$ of the nebula covering several bright CELs and taken with an average seeing of about 1\farcs8. Those authors present a density map obtained from the ratio of the {\fsii} doublet confirming the presence of a density gradient that reaches its highest point immediately south-southwest of the Trapezium stars, and some localized density enhancements in the Orion bar and some Herbig-Haro objects. Very recently, \citet{sanchezetal07} have obtained an integral field spectroscopy mosaic of an area of 5$'\times 6'$ of the center of the Orion nebula, with a spatial resolution of 2\farcs7. The electron density map they obtain is consistent with that obtained by \citet{poggeetal92} but richer in substructures, some of them possibly associated to Herbig-Haro objects. \citet{sanchezetal07} also obtain a electron temperature map (derived from the line ratio of {\fnii} lines) that shows clear spatial variations, which rise near the Trapezium and drop to the outer zones of the nebula. However, an important drawback of the temperature map of S\'anchez et al. is that is based on non-flux calibrated spectra and possible effects due to variations in the dust extinction distribution cannot be disregarded. \citet{odelletal03} obtained a high spatial resolution map of the electron temperature --derived from the line ratio of {\foiii} lines-- of a 160$''\times 160''$ field centered at the southwest of the Trapezium. The data were obtained from narrow-band images taken with the WFPC of the $HST$. Although they do not find a substantial radial gradient of {\elect} in the nebula, \citet{odelletal03} report the existence of small-scale temperature variations down to a few arcseconds compatible with the values of the temperature fluctuations parameter calculated from the AD determinations by \citet{estebanetal04}. \citet{rubinetal03} obtained $HST$/STIS long-slit spectroscopy at several slit positions on the Orion nebula analysing the electron temperature and density spatial profiles with resolution elements of 0\farcs5 $\times$ 0\farcs5. These last authors do not find large-scale gradients of the physical conditions along the slits but a relatively large point-to-point variation and some correlation of such variations with several small-scale structures. The spatial mapping of the AD factor has been performed in few ionized nebulae but largely for PNe. \citet{liuetal00}, \citet{garnettdinerstein01}, and \citet{krabbecopetti06} have found significant differences in the spatial profiles of the O$^{++}$/H$^+$ ratio derived making use of RLs and CELs suggesting the presence of chemical inhomogeneities or additional mechanisms for producing the {\oii} lines in these objects. \citet{tsamisetal03} have performed the only available study so far of the spatial distribution of the AD factor in an {\hii} region: 30 Doradus. However, considering the extragalactic nature of this object and the spatial sampling of 3\farcs5 used by those authors, their final spatial resolution is very low --about 1pc. In any case, \citet{tsamisetal03} find a rather constant AD factor along the zone covered with their observations, a quite different behavior than that observed in PNe. In \S\S~\ref{obsred} and~\ref{linesel} of this paper we describe the observations, the data reduction procedure and the aperture extraction and measurement of the emission lines. In \S~\ref{phiscondabund} we derive the physical conditions and the ionic abundances from both kinds of lines: CELs and RLs. In \S~\ref{spat_prof} we present and discuss the spatial profiles of the physical conditions, line fluxes, and the abundance discrepancy factor along the slit positions. In \S~\ref{rad_dist} we discuss the large-scale radial distribution of some nebular properties along the nebula. In \S~\ref{cor_ADF} we explore possible correlations between the AD and different nebular parameters. In \S~\ref{t2} we address and estimate the possible temperature fluctuations inside the nebula. Finally, in \S~\ref{conclu} we summarize our main conclusions. | \label{conclu} We have studied the spatial distribution of a large number of nebular quantities along five slit positions covering different morphological zones of the Orion nebula. The resolution element of the observations was 1\farcs2 $\times$ 1\farcs03. The studied quantities were c(H$\beta$), {\elecd}, {\elect}([{\nii}]), {\elect}([{\oiii}]), the intensity of several selected lines (H$\beta$, {\cii} 4267 \AA, {\oii} 4649 \AA, [{\oiii}] 4959 \AA, [{\feiii}] 4881 \AA, [{\nii}] 5755 and 6584 \AA, [{\oi}] 6300 \AA, and [{\sii}] 6717 + 6731 \AA), the O$^{++}$/H$^+$ ratio obtained from collisionally excited lines (CELs) and recombination lines (RLs), and the C$^{++}$/H$^+$ ratio obtained from RLs. The total number of apertures or 1D spectra extracted was 730. We have been able to determine the O$^{++}$/H$^+$ ratio from the faint RLs of this ion in a 92\% of the apertures. The spatial distribution of {\elecd} shows a large range of variation --larger than an order of mag\-ni\-tu\-de-- across the nebula, with local maxima associated with the position of protoplanetary disks (proplyds), Herbig-Haro objects, the Orion bar, and the brightest area of the nebula at the southwest of the Trapezium. The proplyds show quite prominent spikes of {\elect}([{\nii}]) and much lesser ones of {\elect}([{\oiii}]). This fact could be due to collisional deexcitation on the nebular lines of {\fnii} because of the high densities of these objects. Herbig-Haro objects also show somewhat higher values of {\elect}([{\nii}]) but, in this case, the origin could be related to extra heating of the gas due to shock excitation. The spatial distribution of the {\oii} 4949 \AA\ and {\foiii} 4959 \AA\ lines is fairly similar along all the slit positions, a very different behavior to that observed in planetary nebulae. We have found that the abundance discrepancy factor (ADF) of O$^{++}$ --the difference between the O$^{++}$ abundance determined from RLs and CELs-- remains, in general, rather constant along most of the observed areas of the nebula, showing values between 0.15 and 0.20 dex. However, there are some localized enhancements of the ADF, specially at the position of the Herbig-Haro objects HH 202, HH 203, and HH 204. The combined data of all slit positions indicate a clear decrease of {\elect}([{\nii}]) and {\elect}([{\oiii}]) with increasing distance from the main ionizing source of the nebula, $\theta^1$ Ori C. On the other hand, the radial distribution of the ADF shows a rather constant value across the nebula except at the inner 40$''$, where the ADF seems to increase very slightly toward $\theta^1$ Ori C. We have explored possible correlations between the ADF of O$^{++}$ and other nebular quantities, finding a possible very weak increase of the ADF for higher electron temperatures. There are not apparent trends between the ADF and c(H$\beta$), {\elecd}, the {\elect}([{\nii}])/{\elect}([{\oiii}]) ratio, O$^{++}$ abundance, and the O$^{++}$/O$^+$ ratio. Our spatially resolved spectroscopy allows to estimate the value of the mean-square electron temperature fluctuation in the plane of the sky, a lower limit to the traditional {\ts} parameter. We find very low values in all cases, result that is in contradiction with previous estimates from the literature. Our results indicate that the hypothetical thermal inhomogeneities --if they exist-- should be lower than our spatial resolution limit of about 1$''$. It is clear that further studies on the {\elect}, chemical abundances, and ADF distributions at sub-arcsec spatial scales are necessary in trying to disentangle (a) whether small spatial scale temperature fluctuations and/or metal-rich droplets are really present in the Orion nebula and HII regions in general, and (b) the origin of AD problem and its possible relation with {\ts} and other nebular properties. The observations needed for this task are very difficult even for ground-based large-aperture telescopes and, by now, unfeasible with the current space telescopes and their available instrumentation. We thank G. Stasi\'nska and M. Rodr\'{\i}guez for their fruitful comments and help. We are grateful to the referee, Y. Tsamis, for his careful reading of the paper and his comments. This work has been funded by the Spanish Ministerio de Ciencia y Tecnología (MCyT) under project AYA2004-07466. | 7 | 10 | 0710.1285 |
0710 | 0710.2359_arXiv.txt | {} {We evaluate the generation of magnetosonic waves in differentially rotating magnetized plasma.} {Differential rotation leads to an increase of the azimuthal field by winding up the poloidal field lines into the toroidal field lines. An amplification of weak seed perturbations is considered in this time-dependent background state.} {It is shown that seed perturbations can be amplified by several orders of magnitude in a differentially rotating flow. The only necessary condition for this amplification is the presence of a non-vanishing component of the magnetic field in the direction of the angular velocity gradient.} {} | In astrophysical bodies, differential rotation is often associated with magnetic fields of various strength and geometry. If the poloidal field has a component parallel to the gradient of angular velocity, then differential rotation can stretch toroidal field lines from the poloidal ones. In the presence of the magnetic field, differential rotation can be the reason for various magnetohydrodynamic instabilities, particularly if the field geometry is complex. Some of these instabilities occur in the incompressible limit (Velikhov 1959; Fricke 1969; Acheson 1978; Balbus \& Hawley 1991) which applies if the magnetic field is weak and the Alfv\'en velocity is smaller than the sound speed. Other instabilities become important only in sufficiently strong magnetic fields when the effect of compressibility plays a significant role (Pessah \& Psaltis 2005; Bonanno \& Urpin 2006, 2007). Note that incompressible instabilities can often be suppressed by a strong magnetic field. For example, the magnetorotational instability (MRI) does not occur if the magnetic field satisfies the condition $B^2 > -8 \pi \rho s \Omega \Omega' L^2$ where $\rho$ is the density, $L$ is the lengthscale of disturbances, $\Omega=\Omega(s)$ is the angular velocity, and $s$ is the cylindrical radius; $\Omega'= d \Omega/ds$ (see, e.g., Urpin 1996, Kitchatinov \& R\"udiger 1997). Moreover, if $\Omega$ increases with the cylindrical radius MRI cannot arise. In recent years, many simulations of differentially rotating magnetized bodies have been performed, and much of the dynamics was interpreted as being a direct consequence of the MRI (Brandenburg et al. 1995; Hawley at al. 1995; Matsumoto \& Tajima 1995; Hawley 2000). Obviously, the MRI cannot be the only instability that operates in a rotating magnetized gas. For example, stratification can lead to a number of strong non-axisymmetric instabilities (Agol et al. 2001; Narayan et al. 2002; Keppens et al. 2002). Blokland et al. (2005) consider the influence of a toroidal field on the growth rate of the MRI and find that it leads to overstability (complex eigenvalue). Van der Swaluw et al. (2005) study the interplay between different instabilities and argue that the growth rate of convection can be essentially increased due to magnetorotational effects. Note, however, that these studies treat the stability of the magnetic field with a vanishing radial component, a condition which is often not met in astrophysical bodies. In fact, the presence of a radial magnetic field can change substantially the stability properties (Bonanno \& Urpin 2006, 2007). In this paper, we consider stability of a differentially rotating gas in the presence of a non-vanishing radial magnetic field. Differential rotation causes the azimuthal field to increase with time by winding up the poloidal field lines into the toroidal ones. Therefore, a development of small perturbations occurs in the time-dependent background state. We show that stretching of the azimuthal field leads to the generation of magnetosonic waves in a flow. Magnetohydrodynamic waves and turbulence generated by this instability can play an important role in enhancing transport processes in various astrophysical bodies, such as accretion and protoplanetary Disks, galaxies, stellar radiative zones, etc. | We have shown that the winding up of toroidal field lines from the poloidal field lines is accompanied by an amplification of seed perturbations of the velocity and magnetic field. A very simplified model has been considered in this paper, but we believe that qualitatively the same results can be obtained for more general background states and perturbations. Differential rotation and compressibility of the gas lead to the generation of magnetosonic waves with the amplitude that grows with time. The physical processes responsible for this amplification are exactly the same that result in the instability considered by Bonanno \& Urpin (2006, 2007). The only difference is that, in this paper, we consider the development of perturbations on a time-dependent background state and, as a result, the growth of perturbations is not exponential. The behaviour of seed perturbations depends essentially on various parameters and can generally be rather complex. If the parameter $q=s \Omega'/\Omega$ is relatively small ($\leq 0.1$), then the perturbations of velocity and magnetic field initially grow monotonously and can reach quite high values. For example, the perturbation of the vertical velocity becomes approximately $150-200$ times greater than its initial value after only 15-30 rotation periods (see Figs.~2 and 3). Perturbations of the magnetic field reach even higher values during the initial stage. For instance, the Alfv\'en velocity corresponding to the perturbation of the radial field component, $C_{As}$, can exceed the initial velocity perturbation by a factor $\sim (1-3) \times 10^5$ after the same time, but the perturbations of the toroidal field are even stronger. As a result of such a strong initial amplification, seed perturbations can already reach a non-linear regime after 15-30 rotation periods if their initial values are sufficiently large. Further evolution of perturbations will then be entirely determined by non-linear effects. However, if the non-linear regime is not reached during this initial stage, the behaviour of perturbations becomes oscillatory with slowly growing amplitude ($\propto t^{1/2}$). At sufficiently large $t$, the frequency of oscillations grows linearly with time and is given approximately by \begin{equation} \omega \approx \Omega \left( \frac{1}{2} k H \sqrt{\beta_s} s \Omega' t \right). \end{equation} Note that perturbations of the magnetic field exhibit more regular behaviour because they can be expressed in terms of the time integrals of a rapidly oscillating velocity. In the case of a strong differential rotation ($q \sim 1$), perturbations exhibit the oscillatory behaviour from the very beginning and the initial growth of their amplitude is less significant. The generation of magnetosonic waves occurs even if the magnetic field is very strong and suppresses different MHD-instabilities which can arise in a differentially rotating flow (for example, the MRI). The presence of differential rotation and radial magnetic field is, however, crucially important for the considered process. Since both differential rotation and radial field are quite common in astrophysics, we believe that the considered mechanism can occur in various astrophysical bodies and plays an important role in enhancing transport processes in plasma. \vspace{0.5cm} \noindent {\it Acknowledgments.} This research project has been supported by a Marie Curie Transfer of Knowledge Fellowship of the European Community's Sixth Framework Programme under contract number MTKD-CT-002995. VU thanks INAF-Ossevatorio Astrofisico di Catania for hospitality. | 7 | 10 | 0710.2359 |
0710 | 0710.0878_arXiv.txt | Using a population synthesis approach, we compute the total merger rate in the local Universe for double neutron stars, double black holes, and black hole -- neutron star binaries. These compact binaries are the prime source candidates for gravitational-wave detection by LIGO and VIRGO. We account for mergers originating {\em both\/} from field populations and from dense stellar clusters, where dynamical interactions can significantly enhance the production of double compact objects. For both populations we use the same treatment of stellar evolution. Our results indicate that the merger rates of double neutron stars and black hole -- neutron star binaries are strongly dominated by field populations, while merging black hole binaries are formed much more effectively in dense stellar clusters. The overall merger rate of double compact objects depends sensitively on the (largely unknown) initial mass fraction contained in dense clusters ($f_{\rm cl}$). For $f_{\rm cl} \lesssim 0.0001$, the Advanced LIGO detection rate will be dominated by field populations of double neutron star mergers, with a small but significant number of detections $\sim 20$ yr$^{-1}$. However for a higher mass fraction in clusters, $f_{\rm cl} \gtrsim 0.001$, the detection rate will be dominated by numerous mergers of double black holes originating from dense clusters, and it will be considerably higher, $\sim 25 - 300$ yr$^{-1}$. In addition, we show that, once mergers of double black holes are detected, it is easy to differentiate between systems formed in the field and in dense clusters, since the chirp mass distributions are strikingly different. If significant field populations of double black hole mergers are detected, this will also place very strong constraints on common envelope evolution in massive binaries. Finally, we point out that there may exist a population of merging black hole binaries in intergalactic space. | Gravitational wave astronomy is entering a new era: LIGO (Abramovici et al.\ 1992) has now taken a full year of data at its design sensitivity; VIRGO (Bradaschia et al.\ 1990) is nearing completion. Other detectors like GEO or TAMA are also on track. There are well defined plans for improving the LIGO and VIRGO detectors so that, within the next few years, their sensitivity will increase by a factor of up to 30. The list of potential sources for these high-frequency detectors is long and includes supernova explosions, neutron star oscillations, and persistent radiation from rapidly rotating, nonaxisymmetric neutron stars. However, compact object binaries remain the most promising sources. It has been known for some time that their formation may take place in two very different environments: in the galactic field, where their progenitors are massive binaries that evolve in isolation, and in dense star clusters, where they can form at high rates through dynamical interactions. The properties of the populations of double compact object binaries have been previously investigated using both observational and theoretical approaches. In the typical observational approach, the known compact objects binaries, i.e., radio pulsars in double neutron star (NS-NS) systems, were analyzed in detail. Based on their observed properties and the radio selection effects the properties of the entire population can be reconstructed and the expected merger rate can be calculated (Kim et al.\ 2005). This approach, however, cannot be extended to black hole -- neutron star (BH-NS) binaries and double black hole (BH-BH) binaries since these systems have never been observed. Moreover, we know only one merging NS-NS binary in a globular cluster, and so this method has little power for the population originating in clusters. The second, theoretical, approach is based on numerical simulations of stellar evolution in field populations, combined with dynamical simulations of star clusters. For field populations some recent examples of theoretical studies include those by Belczynski, Kalogera \& Bulik (2002, hereinafter BKB02), Voss \& Tauris (2003), Pfahl et al.\ (2005), Dewi et al.\ (2006) and Belczynski et al.\ (2007a). The importance of globular cluster evolution for double compact object formation has long been suspected. The fate of BH populations in globular clusters was studied by a number of groups (e.g., Sigurdsson \& Hernquist 1993; Kulkarni, Hut \& McMillan 1993; Portegies Zwart \& McMillan 2000; Merritt et al.\ 2004). However, previous studies have employed only very simplified treatments of stellar evolution for single stars and binaries in clusters, focusing instead mostly on dynamical interactions. It was pointed out (Phinney et al.\ 1991; and more recently Grindlay, Portegies Zwart \& McMillan 2006) that the formation of NS-NS and BH-NS binaries in clusters is not very efficient and the contribution of clusters to their total merger rate is rather small ($\sim 10-30 \%$). Although the predicted formation rate of merging double NS has a strong density dependence (Ivanova et al.\ 2007), even for dense clusters like like 47~Tuc the NS contribution to total cluster merger rates should be overwhelmed by double BH mergers. Such systems are expected to be formed via dynamical interactions in cluster cores (Gultekin, Miller \& Hamilton 2004; O'Leary et al.\ 2006, 2007). O'Leary et al.\ (2006, 2007) used realistic initial conditions for their cluster simulations (which included the stellar evolution of massive stars and binaries in the first few Myr of the cluster life) and assumed immediate creation of a BH subcluster. Their calculations did not consider further stellar or binary evolution and neglected the effects of lower-mass stars on BH populations. Their results provide the most up-to-date estimates of BH-BH merger rates from globular clusters and the corresponding LIGO detection rates assuming rapid BH segregation into an isolated subcluster. Mackey et al.\ (2007) investigated the effects of BHs on the structural evolution of globular clusters. They used a realistic initial mass function for single stars and observed rapid mass segregation of the BHs into the cluster core (as expected from the Spitzer instability; see Watters et al.\ 2000 and references therein). However, this study was based on direct $N$-body simulations that did not include {\em any\/} primordial binaries, even though binaries could affect the mass segregation and subsequent dynamics very significantly. In this paper we take the next step and calculate the evolution of representative star clusters from the onset of star formation taking into account the large binary fraction for massive stars. We assume that inelastic interactions of hard binaries in the cluster core are effective enough to prevent BHs from completely separating from the rest of the cluster, and we therefore treat the BHs as always well mixed and in thermal equilibrium with other stars in the cluster core. This simplifying assumption is opposite, and complementary, to the one adopted in the models of O'Leary et al. (2006, 2007). Their results and the merger rates calculated in this work give, respectively, lower and upper bounds on the LIGO detection rates. In our new models we include both stellar dynamics and full stellar evolution for single and binary stars to predict the merger rate of double compact objects. All stellar populations are evolved and allowed to interact through an entire cluster lifetime ($\sim 13$ Gyr). Then we combine our estimates for star clusters and the merger rates calculated in O'Leary et al.\ (2006) with the most recent population synthesis field calculations (Belczynski et al.\ 2007a) to deduce the total merger rate. In Section~2 we present a description of the stellar binary evolution treatment used and we introduce our model for globular clusters. In Section~3 we present our results and in Section~4 a brief discussion. | \label{summary} We have presented the ranges of total cosmic merger rate of double compact objects in the local Universe. Our results apply to the distances that will be reached by ground based gravitational-wave detectors such as LIGO or VIRGO ($\sim 300$ Mpc for a typical NS-NS system). In calculating the rates we have considered formation of close double compact objects in field populations as well as in dense stellar clusters, i.e., globular clusters. The main result of our study shows that the predicted merger rates are too small for detection with the current instruments (i.e., initial LIGO) but are very promising for the upgraded detectors (i.e., advanced LIGO). We find that field populations dominate the formation of close NS-NS and BH-NS systems. This was already predicted by Phinney (1991) on the basis of observed binary pulsars. Recently Grindlay, Portegies Zwart \& McMillan (2006) have shown that globular clusters can contribute only up to $\sim 20 \%$ to the cosmic NS-NS merger rates. Formation of double compact object systems with NSs is inhibited by the rather large natal kicks NSs receive in supernovae (Hobbs et al.\ 2005). These lead to {\em (i)} disruption of binaries hosting NS progenitors and {\em (ii)} ejection of NSs from the cluster. The remaining systems containing a NS interact with heavier stars, and in particular BHs, and through exchange interactions are removed from binaries and do not form double compact objects. Also, the initial cluster mass that we use in our simulations ($\sim 5 \times 10^5\msun$, which corresponds to the current average globular cluster mass in our Galaxy; e.g., Meylan \& Heggie 1997, see their Fig.~10.2) is too small to produce a significant number of NS-NS or BH-NS systems. In particular, in our cluster simulations, we do not find any mergers of NS-NS nor BH-NS systems. The results for field populations of NS-NS and BH-NS mergers were discussed in detail by Belczynski et al.\ (2007a). They find (their model A) that field Galactic merger rates are $\sim 15$ Myr$^{-1}$ and $\sim 0.1$ Myr$^{-1}$ for NS-NS and BH-NS, respectively. In our work we have adopted these rates as the total NS-NS and BH-NS merger rates (field + clusters) since the contribution from clusters is rather small for these systems. As discussed by Belczynski et al. (2007a) these rates are too small for any detection with the initial LIGO, while for advanced LIGO a small but significant number of detections is predicted: $\sim 15$ yr$^{-1}$ and $\sim 1$ yr$^{-1}$ for NS-NS and BH-NS, respectively. Formation of close BH-BH systems that merge in a Hubble time is expected to be very effective in clusters. This was already noted in studies that use realistic initial conditions for the evolution of BH-BH binaries in clusters (e.g., Miller \& Hamilton 2002; Gultekin, Miller \& Hamilton 2004; O'Leary et al.\ 2006). Here, we have followed the evolution of a realistic average cluster with full stellar evolution and physical treatment of all BH-BH binaries, assuming that binary interactions are able to prevent the BHs from separating into an isolated subcluster. We have found that the production of BH-BH binary mergers under these assumptions is indeed remarkably effective in dense clusters. The most striking and counter-intuitive difference with other studies is that the predicted BH-BH merger rate in clusters is rather constant over long periods of time ($\sim$ Hubble time). In earlier studies the BH populations were usually introduced in the cluster core and evolved separately from the other stars in the cluster. That leads to a larger merger rate at early times in the cluster evolution followed by a very rapid drop in the merger rates once BHs are eliminated (through mergers and ejections). However, the assumptions introduced in our work, namely {\em (i)} stars that form BHs are initially placed throughout the cluster, and {\em (ii)} BHs do not segregate so strongly as to form a completely decoupled subcluster and are instead allowed to interact with other stars in the cluster, make our results qualitatively different. We find that BHs that form in the core, in fact, produce mergers at the early stages of cluster evolution. But later many massive stars that were in the halo and that formed BHs will steadily feed the cluster core with BHs, i.e., continued mass segregation in the cluster halo is providing BHs to the cluster core on long timescales. Additionally, exchange interactions of BHs with unevolved (e.g., main sequence stars) in the core are effective over a long periods of time, and usually lead to formation of close BH-BH systems that merge later in the core. A simplified scenario involves two single BHs sinking into the core; each catches a main sequence companion, and then two BH-MS binaries interact, forming a close BH-BH system and two single main sequence stars are released back into the cluster. Our average cluster merger rate for BH-BH systems is $\sim 3$ Gyr$^{-1}$ (see Figure~\ref{f.merg}) for our simulated cluster with mass $M_{\rm cl}=4.8 \times 10^5 \msun$. If we scaled this rate to the mass of the galactic disk ($M_{\rm MW}=3.5 \times 10^{10} \msun$), then the BH-BH merger rate would come up to $\sim 180$ Myr$^{-1}$ (i.e., this is the expected BH-BH merger rate if the entire mass of our Galactic disk were contained in dense globular clusters). We should compare this rate to the Galactic field BH-BH merger rate, $0.025$ Myr$^{-1}$ (Belczynski et al.\ 2007a). It is clear that the production of BH-BH mergers is much more efficient (by $\sim 3-4$ orders of magnitude) in clusters as compared to field evolution\footnote{ We have used the rate from the same evolutionary model used for field population (model A of Belczynski et al.\ 2007a) as it was employed in our standard cluster model.}. To combine our predicted cluster rates with the field rates we need to know the initial stellar mass fraction contained in clusters. If we look at the Galactic globular clusters we find that they contain about 0.001 of the total mass in stars found in the field ($f_{\rm cl}=0.001$). However, it is reasonable to expect that many clusters may have been completely destroyed and that the clusters we see today were initially more massive and lost significant mass through evaporation (e.g., Vesperini 1998; Joshi et al.\ 2001). Although we present our predictions for the entire range of plausible $f_{\rm cl}$ values, we consider that the most reasonable estimate is still $f_{\rm cl} \gtrsim 0.001$. While we do not have reliable mass estimates for globular clusters in elliptical galaxies, it has been shown that the specific frequency of GCs per galaxy luminosity in ellipticals is significantly (about an order of magnitude) higher than in spiral galaxies (Kim \& Fabbiano 2004). Moreover, elliptical galaxies are on average more massive than spiral galaxies. Therefore, an upper limit on the initial mass fraction contained in all clusters, although highly uncertain, can probably be set to $f_{\rm cl} \lesssim 0.01$. The total cosmic merger rate for BH-BH systems is then strongly dependent on the mass contained in globular clusters. For a small fraction ($f_{\rm cl}=0.001$) we find the merger rate per Milky Way in our model to be $\sim 0.2$ Myr$^{-1}$, while for a larger fraction ($f_{\rm cl}=0.01$) the merger rate is $\sim 2$ Myr$^{-1}$. At this point we can also estimate the detection rate for a given GW detector. We must keep in mind that the average chirp mass of field BH-BH binaries is much smaller ($M_{\rm c} \sim 7 \msun$) than for cluster BH-BH mergers ($M_{\rm c} \sim 20 \msun$). This allows us to observe cluster BH-BH mergers in a much larger volume and their relative contribution to the detection rate is greater than indicated simply by the merger rates (see \S\,~\ref{ligo}). One could also imagine forming of few very massive BHs (up to $\sim 100\msun$) in the cluster (e.g., through mergers of massive binaries; Belczynski et al.\ 2006) which could form BH-BH mergers characterized by extremely high chirp masses. Such mergers would be detectable from much greater distances, making the observed rates even higher. The predicted ranges of total detection rates, for all types of double compact objects, are presented in Figure~\ref{f.rates} as a function of cluster contribution. The most likely range of values for this parameter ($f_{\rm cl} \sim 0.001\div 0.01$) is marked with the vertical shaded area in Figure~\ref{f.rates}. For very low cluster contributions ($f_{\rm cl}=0.0001$) the detection rates correspond to mergers coming only from field populations and are adopted from the reference model of Belczynski et al.\ (2007b; their model A). With increasing cluster contribution we see a drastic increase in predicted detection rates. This increase is connected to the very effective production of BH-BH mergers in clusters as discussed above. For advanced LIGO the detection rates could be as high as $\sim 25-3000$ yr$^{-1}$ and are higher by more than an order of magnitude than the rates just for field populations. The total rates are dominated by dynamically formed BH-BH mergers in dense stellar clusters. If advanced LIGO does not observe this population of BH-BH mergers it will put strong constraints on the initial stellar mass fraction contained in dense stellar clusters. The production of BH-BH mergers in the field is inhibited by the process identified in Belczynski et al.\ (2007b): many potential BH-BH progenitors evolve through a common envelope phase while the donor is evolving through the Hertzsprung gap. Such a common envelope leads most likely to a merger and aborts potential formation of a BH-BH system (because, in the Hertzsprung gap, the star has not yet developed a clear core-envelope structure and the inspiral does not stop before complete merger of the two interacting stars). If this current understanding of the common envelope phase is correct, we do not expect detection of more than a few field BH-BH mergers per year. However, if the progenitors somehow survive this phase, we could then expect up to $\sim100$ detections of field BH-BH mergers per year by Advanced LIGO. Since the chirp mass distribution of the field and cluster populations are so different (see Figure~\ref{f.mchirp} and Figure 4 of Belczynski et al.\ 2007b) it would be easy to tell the two populations apart. This would, in turn, allow us to both {\em (i)} derive the initial mass fraction in clusters, and {\em (ii)} constrain the fate of massive binary systems going through a common envelope phase. This result highlights the importance of the chirp mass distribution as a diagnostic tool in gravitational wave astronomy (Bulik \& Belczynski, 2003; Bulik, Belczynski, \& Rudak, 2004). | 7 | 10 | 0710.0878 |
0710 | 0710.2750_arXiv.txt | {The transition from galactic to extragalactic cosmic rays is discussed. One of critical indications for transition is given by the Standard Model of Galactic cosmic rays, according to which the maximum energy of acceleration for iron nuclei is of order of $E_{\rm Fe}^{\rm max} \approx 1\times 10^{17}$~eV. At $E > E_{\rm Fe}^{\rm max}$ the spectrum is predicted to be very steep and thus the Standard Model favours the transition at energy not much higher than $E_{\rm Fe}^{\rm max}$. As observations are concerned there are two signatures of transition: change of energy spectra and elongation rate (depth of shower maximum in the atmosphere $X_{\rm max}$ as function of energy). Three models of transition are discussed: dip-based model, mixed composition model and ankle model. In the latter model the transition occurs at the observed spectral feature, ankle, which starts at $E_a \approx 1\times 10^{19}$~eV and is characterised by change of mass compostion from galactic iron to extragalactic protons. In the dip model the transition occures at the second knee observed at energy $(4 -8)\times 10^{17}$~eV and is characterised by change of mass composition from galactic iron to extragalactic protons. The mixed composition model describes transition at $E \sim 3\times 10^{18}$~eV with mass composition changing from galactic iron to extragactic mixed composition of different nuclei. These models are confronted with observational data on spectra and elongation rates from different experiments, including Auger.} \begin{document} | The Ultra High Energy Cosmic Ray (UHECR) has two most important problems. One of them is a presence of spectrum features produced by propagation of UHECR particles through Cosmic Microwave Radiation (CMB) and the second is transition from galactic to extragalactic Cosmic Rays (CR). In the case of extragalactic protons two spectral signatures caused by interaction with CMB are predicted: Greisen-Zatsepin-Kuzmin (GZK) cutoff \cite{GZK} and pair-production dip \cite{BG88}. GZK cutoff is most spectacular prediction for UHECR, which status is still uncertain in present observations, though there are the indications to its presence. The pair-production dip is the spectral feature originated due to electron-positron pair production by extragalactic protons interacting with CMB: $p+\gamma_{\rm CMB} \rightarrow p+e^++e^-$. Recently this feature has been studied in the works \cite{Stanev2000,BGGPL,BGG}. The dip has been observed with very good statistical significance $\chi^2$/d.o.f.$\sim 1$ by the Fly's Eye, Yakutsk, Akeno-AGASA and HiRes detectors, and with much worse statistical significance by Auger detector. The pair-production dip and GZK cutoff are signatures of protons. The confirmation of the shape of these features is the evidence for proton-dominated composition of primary CRs. For nuclei as primaries the shape of the dip and GZK cutoff are strongly modified. The different explanation of the dip has been proposed by Hill and Schramm \cite{HS85}. They interpreted the dip observed in 1980s in terms of two-component model. The low energy component can be either galactic or produced by Local Supercluster. The similar model has been considered in \cite{YT}. The Hill-Schramm dip is widely used now for the explanation of the observed dip. From 1970s in the UHECR spectrum there was observed a flattening, which is called {\em ankle}. Discovery of this feature at Haverah Park detector was interpreted as transition from the steep galactic component to more flat extragalactic one. The transition at ankle has been recently considered in \cite{ankle}. In the dip model the transition is completed at the beginning of the dip at $E \approx 1\times 10^{18}$~eV. The ankle in this model appears as intrinsic part of the dip. Like in ankle model, the transition occurs here also as intersection of flat extragalactic component (this flatness is especially prominent in case of diffusive propagation) with steep galactic spectrum. In the dip and ankle models the extragalactic component is assumed to be proton dominated, while the galactic component is most probably composed by iron nuclei. In the {\em intermediate model}, where transition occurs in the middle of the dip, the extragalactic CRs are assumed to have mixed composition \cite{mixed}. In this paper all three above-mentioned models of transition are discussed. The logic of our discussion is as follows: we approach first the transition from the high energy end of galactic CRs, then we discuss the properties of UHECR relevant for transition problem and finally we describe the transition from properties of these two components. | \vspace{1.3mm} The region of transition from galactic to extragalactic CRs at energy between $1\times 10^{17} - 1\times 10^{19}$~eV is the key energy range for understanding the origin of CRs. At low energy part it includes the high energy end of galactic CRs. The information on maximum energy of acceleration, chemical composition and propagation in Galaxy at these energies will clarify the total picture of origin at lower energies. The low energy part of UHECRs is important for understanding of origin of UHECRs and their propagation in extragalactic magnetic fields. The transition from galactic to extragalactic CRs is the central issue of this energy region. There are two detectors which cover partially the above-mentioned region: KASCADE-GRANDE \cite{KASCADE-G} and TALE \cite{TALE}. There are also the proposals to extend the observations of Auger to energy $E \sim 1\times 10^{17}$~eV (see e.g. \cite{LE-Auger}). The Auger detector has great potential to explore this region, building more dense part of the detector covered with fluorescent, scintillator and muon detectors. The basic information which can be obtained includes precise measurement of energy spectra and mass composition (there is little hope to detect anisotropy in this energy region, though in some models the galactic sources can be observed in protons with energy $E \lesssim 10^{18}$~eV \cite{BGH}). At present we have the sufficiently good data on spectra and mass composition at energy range $1\times 10^{18} - 4\times 10^{19}$~eV. The spectra are measured with high statistics (especially in case of the Auger detector), but problem is the accuracy of energy determination. From quite disappointing Fig.~\ref{fig:AgHiYaAu} (left panel) one concludes that scales of energy determination is quite different in all detectors. Energy calibration with help of the pair-production dip suggests that energy measured by scintillator detectors is systematically higher than that by the fluorescence detectors and it gives a reasonable recipe of increasing energies given by fluorescent method and decreasing it for the scintillation method. In this case the curves 'Yakutsk' and 'Akeno-AGASA' in Fig.~\ref{fig:AgHiYaAu} go down and 'HiRes' and 'Auger' - up. For HiRes, AGASA and Yakutsk the method of calibration with help of dip works successfully (see Fig.~\ref{fig:AgHiYa}) with energy shift within the allowed systematic errors, but for Auger it requires the shift by factor 2 greater than systematic error. The pair-production proton dip in terms of modification factor is an excellent tool to measure {\em spectrum shape} independently of absolute flux. From Fig.~\ref{fig:dips} one sees the excellent agreement of the theoretical dip with data of AGASA, HiRes and Yakutsk. By the standards of cosmic-ray physics the agreement with Auger data is also good, but $\chi^2$ for comparison with SD data is very large. This is a result of very big statistics in the surface detectors at lowest energies $E \geq 4.5\times 10^{18}$~eV. In the lowest energy bin at $E=4.5\times 10^{18}$~eV there are 4128 events and the error in determination of flux provided mostly by this statistics is $\delta J/J=0.024$. The theoretical value of modification factor at this energy is only 14\% higher than experimental value, but owing to very small $\delta J/J$,the contribution of this bin to $\chi^2$ is 99.27 ! Most probably the other sources of errors should be included in the bins with small $\delta J/J$, and a possible source of this error is the energy errors which are changing randomly inside a bin. These could be statistical errors and energy-dependent part of systematic errors. Assuming that number of events are distributed in a bin as $N(E)=K E^{-\gamma}$ one obtains $\delta J/J = \gamma (\delta E/E)_r$, where $(\delta E/E)_r$ is the random energy error inside the bin. The estimated value $\delta J/J$ is much larger than what obtained in Auger analysis for all reasonable values of $(\delta E/E)_r$ and $\gamma$. More generally, according to Markus Roth's remark, $\chi^2$ analysis is not adequate for the cases of small $\delta J/J$ and large $(\delta E/E)$. At this stage of analysis we do not consider Fig.~\ref{fig:dips} as contradiction with Auger data.\\*[4mm] Coming to the transition from galactic to extragalactic CRs, we emphasize that at present there are only two experimental methods to study it: measuring the spectrum and mass composition. The transition will be clearly seen if spectrum of iron nuclei and that of protons are measured separately (see Fig.~\ref{fig:dip-ankle}), but even without this ideal possibility the total spectrum has signatures of transition in the form of the spectral features - {\em second knee} in case of the dip model and {\em ankle} in case of the ankle model. The spectrum can be measured nowadays with high accuracy and its shape contains the information about mass composition, which is the other characteristic of the transition. The pair-production dip with its specific shape is a signature of proton-dominated composition (nuclei contribution should be not more than 10 -15 \% \cite{BGGPL}) and its observational confirmation is an argument not weaker than that due to $X_{\rm max}$ measurement (we remind that only two free parameters are involved in describing about 20 energy bins in each experiment). \\*[4.1mm] The mass composition gives another way to test the transition. The best method at present is given by measuring of elongation rate $X_{\rm max}(E)$. Unfortunately this method has many uncertainties, including those in value of fluorescent yield, absorption of UV light in the atmosphere and uncertainties in the models of interactions, needed to convert the tested mass composition into $X_{\rm max}$. The systematic errors in measuring $X_{\rm max}$ can be as large 30~g/cm$^2$ to be compared with difference about 100~g/cm$^2$ between $X_{\rm max}$ for protons and iron. The better sensitivity for distinguishing different nuclei is given by distribution over $X_{\rm max}$ \cite{ABBO}. There are three models of the transition: ankle, dip and mixed-composition model. They differ most notably by the energy of transition (ankle: $E \sim 1\times 10^{19}$~eV, dip: $E \approx 1\times 10^{18}$~eV and mixed composition model $E \approx 3\times 10^{18}$~eV), and by mass composition of extragalactic component (protons - for the ankle model, proton-dominated - for the dip model and mixed composition - for the third model). The {\em ankle model} contradicts the Standard Model of Galactic CRs (energy where galactic flux is half of that observed is two orders of magnitude higher than energy of iron knee) and severely disagrees with $X_{\rm max}$ measured in all experiments at $(1.5 - 5)\times 10^{18}$~eV. The {\em dip model} is based on well confirmed signature of proton interaction with CMB - pair-production dip. The two other models must assume that agreement of pair-production dip with data is accidental and the observed dip is produced by two components, galactic and extragalactic. The dip model assumes the iron-dominated galactic flux below $5\times 10^{17}$~eV and proton-dominated extragalactic flux above $1\times 10^{18}$~eV. This mass composition is confirmed by HiRes and HiRes-Mia data for elongation rate. It does not contradict the bulk of all data on $X_{\rm max}$, but contradicts $X_{\rm max}$ measured by Auger, especially the highest energy points. The generation spectrum in this model is $E^{-2}$ or $E^{-2.2}$ as needed by shock acceleration with a steepening to $\gamma_g=2.7$ due to distribution of sources over maximum energy of acceleration of source luminosities. The proton-dominated composition can be produced in some models of injection to the shock acceleration. The {\em mixed composition model} assumes mixed composition generation spectrum for extragalactic component with generation index 2.1 - 2.3. It has many free parameters, most notably ones describing the mass composition of the generation spectrum, and thus it can in principle explain any observed mass composition. However, this model has a robust prediction at energy $E \gtrsim 3\times 10^{19}$~eV: proton-dominated composition and the GZK feature. As far as Auger elongation rate is concerned, the mixed composition model explains well the break in elongation rate at $2\times 10^{18}$~eV and contradicts the two Auger points at $E > 2\times 10^{19}$~eV. The energy where transition to extragalactic CRs is completed in most versions of this model equals $E \approx 3\times 10^{18}$~eV. Much better quality of data on $X_{\rm max}$ is needed to distinguish the dip and mixed-composition models by $X_{\rm max}$ measurements. Probably it is possible to do using $X_{\rm max}$ distribution \cite{ABBO}.\\*[1mm] We will comment now on agreement of the transition models with the measured galactic spectrum. For all three models it is reached by the formal subtraction procedure: the galactic spectrum is found as difference between measured total spectrum and calculated extragalactic spectrum. But the galactic spectrum calculated in the Standard Model at $E \gtrsim 1\times 10^{17}$~eV is very steep and, as was demonstrated in \cite{Tanco}, for diffusive model of propagation all three models contradict the calculated galactic spectrum, the dip model to the less extent. Strictly speaking this contradiction is produced by exponential cutoff in the acceleration spectrum at $E > E_{\rm max}^{\rm acc}$. \\*[2mm] The most consistent conclusions on nature of observed UHECRs are obtained at present by HiRes detector: it has confirmed the pair-production dip and thus proton-dominant composition at $1\times 10^{18} - 4\times 10^{19}$~eV, the $X_{\rm max}$ measurements agree with proton-dominant composition at $E > 1\times 10^{18}$~eV , and $E_{1/2}$ measurement confirms that steepening of the spectrum observed at $E > 4\times 10^{19}$~eV is really the GZK cutoff. Therefore, according to these data CRs observed at $E \gtrsim 1\times 10^{18}$~eV are extragalactic protons exhibiting two signatures of interaction with CMB: pair-production dip and GZK feature. | 7 | 10 | 0710.2750 |
0710 | 0710.5036_arXiv.txt | We study the 37 brightest radio sources in the Subaru/\textit{XMM-Newton} Deep Field (SXDF). Using mid-IR (Spitzer MIPS 24 $\mu \rm m$) data we expect to trace nuclear accretion activity, even if it is obscured at optical wavelengths, unless the obscuring column is extreme. Our results suggest that above the `FRI/FRII' radio luminosity break most of the radio sources are associated with objects that have excess mid-IR emission, only some of which are broad-line objects, although there is one clear low-accretion-rate FRI. The fraction of objects with mid-IR excess drops dramatically below the FRI/FRII break, although there exists at least one high-accretion-rate QSO. Investigation of mid-IR and blue excesses shows that they are correlated as predicted by a model in which a torus of dust absorbs $\sim$30\% of the light, and the dust above and below the torus scatters $\gtsim$1\% of the light. | Powerful radio sources are believed to have central super-massive black holes (SMBH) with uniformly high accretion rates at the highest radio luminosities and typically lower accretion rates at lower radio luminosities (\cite{rs91}). Low-luminosity radio jets can, however, be associated with high-accretion-rate systems, and these so-called `radio quiet' quasars appear to have similar FRI-like radio structures to low-accretion-rate counterparts of similar radio luminosity (e.g. \cite{hbr07}). At low redshift, the most massive ($\gtsim$ $10^{8} \rm M_{\odot}$) SMBH typically have very low accretion rates with systematically higher average values at $z \gtsim$ 2, the so-called `quasar epoch' (\cite{yt02}). These observational results fit in with theoretical ideas that a `quasar mode' of feedback is prevalent in the distant universe, and that a `radio mode' feedback is dominant at low redshift (e.g. \cite{cro06}). The central region of an AGN is surrounded by a dusty torus which absorbs light and re-emits it in the infrared. Above and below the plane of the torus, dust scatters light yielding a blue excess. Such mechanisms make it difficult to observe objects viewed through the torus directly in the optical, UV and soft X-rays. The torus creates anisotropic obscuration of the central regions resulting in two different types of observed objects, type 1 that are viewed face-on and type 2 that are viewed edge-on. Here we use mid-IR observations to search for evidence of accretion in a manner which is far less dependent on orientation. The sample studied here is the 37 brightest radio sources from the VLA survey of the Subaru/\textit{XMM-Newton} Deep Field (SXDF; \cite{sim06}) with flux densities greater than 2 mJy at 1.4 GHz. Optical, X-ray and radio observations of the SXDF were made within the 1.3 square degree Subaru/\textit{XMM-Newton} Deep Field with Subaru, \textit{XMM-Newton} and the VLA respectively. Thirteen of our objects are not as yet spectroscopically confirmed so we use photometric redshifts in these cases, calculated with the HYPERz code (\cite{boz00}) and typically nine data points from $B-$band (440 nm) to 4.5 $\mu \rm m$ (Vardoulaki et al. in prep). | \begin{figure} \begin{center} \setlength{\unitlength}{1mm} \begin{picture}(140,38) \put(65,-21){\special {psfile="l1_4_D.ps" vscale=32 hscale=32 angle=90}} \put(143,-21){\special {psfile="blue_red.ps" vscale=32 hscale=32 angle=90}} \end{picture} \end{center} \vspace{0.5in} {\caption[junk]{\label{fig1} a {\it Left}: Radio Luminosity $\log_{10}(L_{1.4{\rm GHz}}/ \rm [W Hz^{-1} sr^{-1}])$ versus largest projected linear size $D$: symbols indicate optical/IR classification; filled red circles for quasars `Q'; filled blue triangles for obscured quasars `OQ'; filled black squares for possible galaxies `G?'; black squares for secure galaxies `G'; green upside-down triangles for starbursts 'SB'; orange diamonds for weak quasars `WQ'; and light blue stars for BL Lac `BL'. The horizontal lines show the RLF and FRI/FRII breaks calculated from the values in \cite{fr74} using a typical steep-spectrum spectral index of 0.8 and translated to our assumed cosmology. b {\it Right}: Blueness versus mid-IR excess. Symbols are the same as in the left figure. The black dotted line corresponds to the best-fit line in the log-linear plane where all objects, were treated as detections; the slope and intercept are 0.26 and -1.27 respectively, giving (see Eqn (2)) $\rm k_{1} = 0.05$ and $\rm k_{2} = 0.03$. The blue solid line corresponds to the best-fit line in log-linear plane where objects without detections at 24 $\mu \rm m$ were treated as limits; the slope and intercept are 0.10 and -0.41 respectively, giving $\rm k_{1} = 0.39$ and $\rm k_{2} = 0.09$. The Buckley-James method in the ASURV statistics package (\cite{lav92}) was used in these calculations. `SB' and `BL' objects were excluded from the calculations since they have SEDs dominated by different physical processes to those assumed in the model described by Eqns (1) and (2). We adopt a radio spectral index $\alpha$ = 0.8 ($S_{\nu} \propto \nu^{-\alpha}$), unless a spectral index could be calculated using 1.4 GHz data from \cite{sim06} and 325 MHz data from \cite{tasse} (see Vardoulaki {\it et al.} in prep.). We assume throughout a low-density, $\Lambda$-dominated Universe in which $H_{0}=70~ {\rm km~s^{-1}Mpc^{-1}}$, $\Omega_{\rm M}=0.3$ and $\Omega_{\Lambda}=0.7$. }} \end{figure} \addtocounter{figure}{0} We use optical/IR observations to classify a radio source as either Quasar `Q', Obscured Quasar `OQ', Galaxy? `G?', Galaxy `G', Starburst `SB', Weak Quasar `WQ' or BL Lac `BL'. We deem that nuclear accretion is `significant' in objects that obey $\log_{10}(L_{24 \mu \rm m}/ \rm [W Hz^{-1} sr^{-1}]) > 23.1$ (or $[\lambda L]_{24 \mu \rm m} > 10^{37.3}$ ${\rm W}$). This value corresponds to $[\lambda L]_{24 \mu \rm m} \ge 10^{-1.8} L_{Edd}$, a typical lower limit for quasars (\cite{mcl04}), for a black hole mass $M_{\rm BH} \ge 10^{8} M_{\odot}$, a typical lower limit for radio sources (\cite{mcl_etal04}); $L_{Edd}$ is the Eddington luminosity. We then define the following categories:\\ i) {\bf Q}: Broad lines in the optical spectrum (3/37 cases). None of these are detected at 24 $\mu \rm m$, although their limits are insufficient to rule out significant accretion.\\ ii) {\bf OQ}: Objects with a 24-$\mu \rm m$ detection (5/37 cases) and with sufficient $L_{24}$ to represent significant accretion. This class may be incomplete in that some objects in the `G?' class, as described next, have limits above this critical value.\\ iii) {\bf G?}: A galaxy that has a 24-$\mu \rm m$ limit consistent with it lying above the $\log_{10}(L_{24 \mu \rm m}/$ $\rm [W Hz^{-1} sr^{-1}]) = 23.1$ line\footnote{Because of the 24 $\mu \rm m$ flux density limit, these objects are at high redshift, and hence, because of the 1.4-GHz flux density limit, a high-$L_{1.4 \rm GHz}$ sub-set of the objects lacking Spitzer 24 $\mu \rm m$ detections.} (11/37 cases).\\ iv) {\bf G}: All other objects (15/37 cases) without significant accretion, unless they fall into three special categories defined by properties derived from spectroscopy, the SED and the optical structure: {\bf SB}: evidence from the SED of a starburst component (1/37 cases); {\bf WQ}: evidence from the SED of a quasar component but no 24 $\mu \rm m$ detection (1/37 cases); {\bf BL}: featureless red continuum and a point source at $K$ (1/37 cases).\\ Figure 1a shows the 1.4-GHz radio luminosity at $L_{\rm 1.4 GHz}$ versus the projected linear size $D$ with the symbols denoting the different optical/IR classes. We see that nearly all `Q', `OQ' and `G?' objects of our sample lie above the `FRI/FRII' luminosity break\footnote{Although the FRI/FRII classification scheme is on the basis of radio structure, there is a sharp change in radio structure at a characteristic radio luminosity \cite{fr74}.}, with the exception of the `OQ' sxdf\_0034 (the `G?' object near sxdf\_0034 lies very close to the boundary of significant accretion). In previous studies, the quasar fraction has been defined as the number of sources with quasar-like optical features (e.g. broad lines) and has a value of $\sim 0.1 \rightarrow 0.4$ over this range of $L_{1.4 \rm GHz}$ (e.g. \cite{wil00}). We introduce the `quasar-mode fraction' $f_{\rm QM}$ to describe the fraction of objects with high accretion rates to the total number of objects. Above the FRI/FRII break $f_{\rm QM} \sim 0.5 - 0.9$ (the lower value assumes the 24 $\mu \rm m$ limits are much higher than the true 24 $\mu \rm m$ values, whereas the higher value assumes the true values lie just below the limits). The one clear exception in this regime is sxdf\_0001, which has no evidence of a QSO and a clear Twin-Jet (FRI) radio structure. The quasar-mode fraction drops dramatically below the FRI/FRII break\footnote{We note that objects in our sample above the FRI/FRII break have median redshift $z_{\rm med} \sim 1.6$, whereas those below have $z_{\rm med} \sim 0.65$, so evolutionary effects may also be important.}, and whether or not one excludes some of the compact ($D <$ 100 kpc) sources as potentially part of a separate (beamed) population, then $f_{\rm QM} \ltsim$ 0.1 because nearly all objects are galaxies `G'. The counter example here are sxdf\_0034, the only `OQ' below the FRI/FRII break, and potentially an optically-obscured example of unobscured FRI QSOs already studied in this radio luminosity regime (e.g. \cite{hbr07}). Inspection of the SEDs (Vardoulaki {\it et al.} in prep) shows that some of our objects have an excess at 24 $\mu \rm m$ above that expected from extrapolation of the stellar populations. This is quantified via a measure of the mid-IR excess, $\log_{10}([\nu L]_{10 \mu \rm m rest} / [\nu L]_{1\mu \rm m rest})$. A comparison of mid-IR excess and blueness is presented in Fig. 1b where a positive correlation is obvious. The generalised Spearman correlation calculated using survival analysis statistical package ASURV (\cite{lav92}) is 0.657 with a 99\% probability for a correlation. Consider a simple model in which blueness is connected to mid-IR excess through the following equations\footnote{Equation (2) relies on the $ln(1+x) \approx x$ approximation which is only accurate around and below the knees of the functions plotted in Fig. 1b.}: \begin{equation} [\nu L]_{4000 \rm \AA \rm rest} = \rm k_{1} \times [\nu L]_{1\mu \rm m rest} + \rm k_{2} \times [\nu L]_{10 \mu \rm m rest} \Rightarrow \end{equation} \begin{equation} \log_{10}\left(\frac{[\nu L]_{4000 \rm \AA rest}}{[\nu L]_{1\mu \rm m rest}} \right) = \log_{10}(\rm e) \times \frac{\rm k_{2}}{\rm k_{1}} \times \left(\frac{[\nu L]_{10 \mu \rm m rest}}{[\nu L]_{1\mu \rm m rest}}\right) + \log_{10}(\rm k_{1}), \end{equation} where $\rm k_{1}$ encodes the contribution of the stellar population of a passively evolving galaxy formed at high redshift ($z > 5$), and $\rm k_{2}$ the mid-IR-excess parameter that we are looking to calculate for this sample of radio sources. This model assumes that light from the nucleus with intrinsic optical luminosity $L_{\rm opt}$ is i) absorbed by dust and re-emitted in the mid-IR generating luminosity $[\nu L]_{10 \mu \rm m rest}$ and ii) scattered, generating luminosity $[\nu L]_{4000 \rm \AA rest}$. Fig. 1b shows best-fit lines for two scenarios: 1) all objects were treated as detections (black dotted line), and 2) objects are treated as upper limits according to their 24 $\mu \rm m$ detection (blue solid line), where in both cases `SB' and `BL' objects were excluded (Fig. 1b). Averaging these results we deduce $\rm k_{1} \sim 0.2$ and $\rm k_{2} \sim 0.05$, which agrees well with independent evidence. The value deduced for $\rm k_{1}$ is in line with the expectations of template spectra of galaxies which formed their stars at high redshift. Optical polarisation studies (e.g. \cite{kis01}) tell us that $[\nu L]_{4000 \rm \AA rest} \gtsim$ $0.01 [\nu L]_{\rm opt}$, which is consistent with our value of $\rm k_{2}$ given that QSO SED studies suggest $[\nu L]_{10 \mu \rm m rest} \sim 0.3 [\nu L]_{\rm opt}$ (\cite{rr95}). We conclude that whenever nuclear accretion is significant in our sample of radio sources, dust in the torus absorbs 30\% of the photons and dust above and below the torus scatters $\gtsim$1\% of the photons. | 7 | 10 | 0710.5036 |
0710 | 0710.0339_arXiv.txt | The observational evidence for central black holes in globular clusters has been argued extensively, and their existence has important consequences for both the formation and evolution of the cluster. Most of the evidence comes from dynamical arguments, but the interpretation is difficult, given the short relaxation times and old ages of the clusters. One of the most robust signatures for the existence of a black hole is radio and/or X-ray emission. We observed three globular clusters, NGC6093 (M80), NGC6266 (M62), and NGC7078 (M15), with the VLA in the A and C configuration with a 3-$\sigma$ noise of 36, 36 and 25 $\mu$Jy, respectively. We find no statistically-significant evidence for radio emission from the central region for any of the three clusters. NGC6266 shows a 2-$\sigma$ detection. It is difficult to infer a mass from these upper limits due to uncertainty about the central gas density, accretion rate, and accretion model. | Although we do not understand how the nuclei of galaxies form or why they have black holes (BH) at their centers, the correlation between BH mass and bulge velocity dispersion does shed light on their formation and evolutionary histories (Gebhardt et al. 2000a, 2000b: Ferrarese and Merritt 2000). A number of different theories (e.g., Silk \& Rees 1998; Haehnelt \& Kauffmann 2000; Robertson et al. 2006) predict a BH mass bulge-velocity-dispersion relation, although they predict different slopes and intercepts for this relation. Exploration of the extreme ends of this relationship will help illuminate the underlying physical model, and in this paper we focus on the low mass end. Black holes at the low end of the relations, with masses between 100 and $10^6~\Msun$, are generally referred to as intermediate-mass black holes (IMBHs). There is significant evidence that black hole masses from $10^5-10^6~\Msun$ exist from the work of Barth, Greene \& Ho (2005) and Greene \& Ho (2006). To go to yet smaller black hole masses, an extrapolation of the correlation between black hole mass and stellar velocity dispersion suggests studying stellar systems with velocity dispersions of 10--20~\kms. These dispersions are characteristic of globular clusters. Whether the existence of black holes in globular clusters could shed light on the formation and correlations of supermassive black holes is unknown, but clearly it is a possibility. Furthermore, the existence of massive black holes in clusters will have a significant effect on the cluster evolution. Thus, quantifying the demographics of black holes in clusters may be related to how supermassive black holes grow, and will definitely yield useful information about the evolution of clusters. Theoretical work suggests that we might expect IMBHs at the centers of steller systems (Ebisuzaki et al. 2001; Portegies Zwart \& McMillian 2002; Miller \& Hamilton 2002), although it appears to be difficult to make black holes more massive than 100~$\Msun$. Gurkan et al. (2004) suggest that IMBHs may be easy to form through runaway collisions with massive stars. Discoveries of BHs in globular clusters have been claimed --- G1 in M31 (Gebhardt, Rich \& Ho 2002) and M15 (van der Marel et al. 2002; Gerssen et al. 2002). In fact, the M15 claim has been made for the past 30 years, starting with the result of Newell, da Costa \& Norris (1976) and subsequently challenged by Illingworth \& King (1977). The basic issue is being able to distinguish a rise in the central mass-to-light ratio being due to either a black hole or the expected stellar remnants (neutron stars, massive white dwarfs and solar mass black holes). The most recent M15 result has been challenged by Baumgardt et al. (2003a). The result in G1 has also been challenged by Baumgardt et al. (2003b) but Gebhardt, Rich \& Ho (2005) include additional data and analysis that support the black hole interpretation. There has been two further observations which strongly support the existence of a black hole in G1. Trudolyubov \& Priedhorsky (2004) measure X-rays from G1 using the Chandra Observatory, centered to within 2\arcsec\ of the center of G1. Subsequently, Pooley \& Rappaport (2006) suggest the X-ray emmission is from accretion onto a black hole, and Maccarone \& Koerding (2006) point out that if a black hole is present then a 30 $\mu$Jy radio source may be expected. The most significant observation comes from Ulvestad, Greene \& Ho (2007) who find a 28 $\mu$Jy (4.5$\sigma$) emission centered on G1. Other interpretations are a pulsar wind or a planetary nebula. The pulsar wind seems unlikely given the age of G1 and the point-like radio source (an old pulsar would have a large size). A planetary nebula would show optical emission lines which are not seen in the HST or Keck spectra of Gebhardt et al. (2003). Other studies of the existence of black holes in globular clusters have been less compelling. Colpi, Mapelli, \& Possenti (2003) use indirect dynamical arguments to suggest a few hundred solar mass black hole in NGC~6752. McLaughlin et al. (2006) provide an estimate of black hole in 47Tuc of $900\pm900~\Msun$. To date, there are no published upper limits of black hole masses that are significantly below that expected from an extrapolation of the correlation between black hole mass and stellar velocity dispersion. While the dynamical arguements strongly support the black hole interpretation in at least G1, the radio emission provides a clear and obvious result. Unfortunately, it is difficult to predict the radio emission from a given black hole mass. The next step is to explore other globular clusters with a similar setup and deep exposures. | Failure to detect radio radiation at 8.6 GHz from the centers of three globular clusters does not prove that no globular clusters have IMBHs at their centers. Besides not having a black hole, other interpretations include 1) accretion by the BH could be episodic and we happened to observe the BHs in an ``off-state", 2) the gas density could be much lower compared to galaxies, 3) the radiative efficiency may be lower than assumed (although the assumed efficiencies are already quite low), 4) or the accretion model may not be adequate in general. We would predict, using the relation of Merloni et al. (2003) or using standard accretion models and gas density estimates (as done in Maccarone 2004), that we should have detected radio radiation at 8.6 GHz if accretion is steady and the accretion rate times the Bondi rate is 10$^{-4}\times$ or higher. We would not have been able to detect the flux density predicted by a rate of 10$^{-5}\times$ or less. Ulvestad et al. (2007) estimate the fraction of the Bondi rate of just under 1\% for G1, but it is difficult to interpret due to the unknown radiative efficiency. For galactic black holes, the radiative efficiencies appear to vary greatly with some lower than $10^{-5}$ (Lowenstein et al. 2001), although consistent with rates of around 10\% of the Bondi rate. Models which predict 8.6 GHz flux densities from central BHs in globular clusters above about 25 $\mu \rm Jy$/beam can be tested with the VLA currently. The EVLA should produce, for continuum observations, a sensitivity improvement of about a factor of 15, making 8.6 GHz flux densities above about 2$\mu\rm Jy$/beam detectable. | 7 | 10 | 0710.0339 |
0710 | 0710.1804_arXiv.txt | Low-mass X-ray binaries, recycled pulsars, cataclysmic variables and magnetically active binaries are observed as X-ray sources in globular clusters. We discuss the classification of these systems, and find that some presumed active binaries are brighter than expected. We discuss a new statistical method to determine from observations how the formation of X-ray sources depends on the number of stellar encounters and/or on the cluster mass. We show that cluster mass is not a proxy for the encounter number, and that optical identifications are essential in proving the presence of primordial binaries among the low-luminosity X-ray sources. | The first celestial maps in X-rays, in the early 1970s, show that globular clusters harbour more X-ray sources than one would expect from their mass. As a solution to this puzzle it was suggested that these bright ($L_x\gtap10^{36}$\,erg/s) X-ray sources, binaries in which a neutron star captures mass from a companion star, are formed in close stellar encounters. A neutron star can be caught by a companion in a tidal capture, or it can take the place of a star in a pre-existing binary in an exchange encounter. Verbunt \&\ Hut (1987) showed that the probability of a cluster to harbour a bright X-ray source indeed scales with the number of stellar encounters occurring in it; whereas a scaling with mass does not explain the observations. With the \textit{Einstein} satellite a dozen less luminous ($L_x\ltap10^{35}$\,erg/s) X-ray sources were discovered in the early 1980. \textit{ROSAT} enlarged this number to some 55, and now thanks to \textit{Chandra} we know hundreds of dim X-ray sources in globular clusters. The nature and origin of these dim sources is varied. Those containing neutron stars, i.e.\ the quiescent low-mass X-ray binaries in which a neutron star accretes mass from its companion at a low rate and the recycled or millisecond radio pulsars, have all formed in processes involving close stellar encounters. The magnetically active binaries, on the other hand, are most likely primordial binaries, with stars that are kept in rapid rotation via tidal interaction. Cataclysmic variables are binaries in which a white dwarf accretes matter from a companion. In globular clusters they may arise either via stellar encounters, or from primordial binaries through ordinary binary evolution -- this is expected to depend on the mass and density of the globular cluster. In this paper we describe the classification and identification of the dim sources in Section\,2, and make some remarks on the theory of their formation in Section\,3. In Section\,4 we will discuss a new, and in our view more accurate, way to compare the numbers of these sources with theoretical predictions. \begin{figure} \centerline{ \parbox[b]{0.55\columnwidth}{\psfig{figure=verbuntf1a.eps,width=0.54\columnwidth,clip=t}} \parbox[b]{0.45\columnwidth}{\psfig{figure=verbuntf1b.ps,width=0.44\columnwidth,clip=t}} } \caption{Left: X-ray hardness-luminosity diagram for dim sources in globular clusters. I: quiescent low-mass X-ray binaries, II: cataclysmic variables III: cataclysmic variables and magnetically active binaries. From Pooley \&\ Hut (2006). Right: Colour-magnitude diagram of NGC\,6752 on the basis of HST-WFPC2 data; objects within X-ray position error circles are marked. Left of the main sequence we find cataclysmic variables, above it active binaries. Updated from Pooley et al.\ (2002a).\label{xcol}} \end{figure} | \begin{itemize} \item mass $M$ is {\em not} a proxy for collision number $\Gamma$ \item the number of dim sources scales both with collision number $\Gamma$ and with mass $M$ \item scaling with mass only is not acceptable \item correct treatment of the background is important, esp.\ for faint sources \item to prove the mass-dependence optical identifications are essential \end{itemize} | 7 | 10 | 0710.1804 |
0710 | 0710.3030_arXiv.txt | We present the results of the timing analysis of five {\it Rossi X-ray Timing Explorer} observations of the Black Hole Candidate GRS 1915+105 between 1996 September and 1997 December. The aim was to investigate the possible presence of a type-B quasi-periodic oscillation (QPO). Since in other systems this QPO is found to appear during spectral transitions from {\it Hard} to {\it Soft} states, we analyzed observations characterized by a fast and strong variability, in order to have a large number of transitions. In GRS 1915+105, transitions occur on very short time scales ($\sim$ sec): to single them out we averaged Power Density Spectra following the regular path covered by the source on a 3D Hardness-Hardness-Intensity Diagram. We identified both the type-C and the type-B quasi-periodic oscillations (QPOs): this is the first detection of a type-B QPO in GRS 1915+105. As the spectral transitions have been associated to the emission and collimation of relativistic radio-jets, their presence in the prototypical galactic jet source strengthens this connection. | \label{par:intro} Systematic variations in the energy spectra and intensity of transient Black-Hole Candidates (BHC) have been recently identified in terms of the pattern described in an X-ray Hardness-Intensity diagram (HID, see Homan et al. 2001, Homan et al. 2005b, Belloni et al. 2005). Four main bright states (in addition to the quiescent state) have been found to correspond to different branches/areas of a square-like HID pattern. In this framework much importance is given to the intermediate states (called Hard Intermediate State, HIMS, and Soft Intermediate State, SIMS) and to the transitions between them, identified from the behaviour in several bands of the electromagnetic spectrum (from radio to hard X-rays, see also Fender, Belloni \& Gallo 2004 and Homan et al. 2005b) and from the timing properties of the X-ray light curve. Low-frequency Quasi-Periodic Oscillations (LFQPOs) with centroid frequency ranging from mHz to tens of Hz have been observed in the X-ray flux of many galactic BHCs since the '80s (see van der Klis 2006; McClintock \& Remillard 2006 and references therein). Three main types of LFQPOs, dubbed Type-A, -B and -C respectively, originally identified in the light curve of XTE J1550-564 (Wijnands et al. 1999; Remillard et al. 2002), have been seen in several sources (see Casella et al. 2005 and references therein). We summarize their properties in Table \ref{ABC_properties}. In the context of the state classification outlined above, it is possible to ascribe the three LFQPOs to different spectral conditions (see Table \ref{ABC_properties}, Homan et al. 2001, Homan \& Belloni 2005, Belloni et al. 2005). The type-C QPO is associated to the (radio loud) HIMS and to the low/hard state. It is a common QPO seen in almost all BHCs with a variable centroid frequency correlated with the count rate, a high fractional variability and a high coherence ($Q=\nu/$FWHM$\sim$10). The type-B QPO has been seen only in few systems, although it is being seen in a growing number of sources (see Casella et al. 2005 and references therein). It is a transient QPO associated to spectral transitions from the (radio loud) HIMS to the (radio quiet) SIMS. Its features are a $\sim$ fixed centroid frequency (around $\sim$6 Hz), lower fractional variability and $Q$ than type-C. Some authors (Fender, Belloni \& Gallo 2004; Casella et al. 2004) suggested that these spectral transitions are in turn associated to the emission and collimation of transient superluminal relativistic jets visible in radio band. These jets are seen in a number of sources (GRS 1915+105, XTE J1550-564, GX 339-4, XTE J1859+226, GRO J1655-40, etc.). However, not in all of these sources we could resolve the spectral transition to see the transient QPO. \begin{table} \label{tab:ABC} \centering \caption{Summary of type-A, -B and -C LFQPOs properties (from Casella et al. 2005).} \label{ABC_properties} \scriptsize \begin{tabular}{lccc} \hline \hline Properties & Type-C & Type-B & Type-A \\ \hline Frequency (Hz) & $\sim0.1-15$ & $\sim5-6$ & $\sim8$ \\ Q($\nu /$FWHM) & $\sim7-12$ & $\gtrsim6$ & $\lesssim3$ \\ Amplitude (\% {\it rms}) & 3-16 & $\sim2-4$ & $\lesssim3$ \\ Noise & strong flat-top & weak red & weak red \\ Phase lag$^{\mathrm{a}}$ @$\nu_{QPO}$ & soft$/$hard$^{\mathrm{b}}$ & hard & soft \\ Phase lag @2$\nu_{QPO}$ & hard & soft & ... \\ Phase lag @$\nu_{QPO}/2$ & soft & soft & ... \\ \hline \end{tabular} \begin{list}{}{} \item[$^{\mathrm{a}}$] With ``hard lag'' we mean that hard variability lags the soft one. \item[$^{\mathrm{b}}$] Trend towards soft lags for increasing QPO frequencies \end{list} \end{table} The spectral properties connected to type-A QPO are similar to those introduced for the type-B. This QPO has been seen in few systems (Casella et al. 2005). It is broader, weaker and less coherent than the type-B QPO. GRS 1915+105 is a transient BHC discovered on August 15 1992 with the WATCH instrument on board GRANAT (Castro-Tirado et al. 1992, 1994). It is the first galactic source observed to have apparently superluminal transient relativistic radio jets (Mirabel \& Rodriguez 1994), commonly interpreted as ejection of ultra-relativistic plasma, with a speed close to the speed of light (up to $\sim98$\%). Its radio variability was discovered to correlate with the hard X-ray flux (Mirabel et al. 1994). Thanks to VLA-radio observations of these jets, Rodriguez et al. (1995) estimated a distance of 12.5 kpc; more recent estimates attest a distance of $6.5 \pm 1.6$ kpc (Kaiser et al. 2005). The mass of the compact object was estimated through IR spectroscopic studies (Greiner, Cuby \& McCaughrean 2001, Harlaftis \& Greiner 2004) to be $14.0 \pm 4 M_{\odot}$ which unambiguously makes GRS 1915+105 a BHC. The various and rich phenomenology of this source was classified by Belloni et al. (2000): they analyzed 163 RXTE observations, showing that the complex behaviour of GRS 1915+105 can be described in terms of spectral transitions between three basic states, A, B and C (not to be confused with the name of the LFQPOs introduced before), that give rise to 12 variability classes. The non standard behaviour of GRS 1915+105 (it is a very bright transient source continuing the same outburst started in 1992) was interpreted as that of a source that spends all its time in Intermediate States (both in its hard and soft flavors), never reaching the LS or the quiescence (see e.g. Fender \& Belloni, 2004). GRS 1915+105 also shows strong time variability on time scales of fractions of second, revealing low- and high-frequency QPOs (see Morgan et al. 1997) whose properties (frequency and fractional variability) are tightly correlated with the spectral parameters (Morgan et al. 1997; Muno et al. 1999; Markwardt et al. 1999; Rodriguez et al. 2002a, 2002b; Vignarca et al. 2003). In particular, all LFQPOs observed from this system can be classified as type-C QPOs. Although GRS 1915+105 makes a large number of fast state transitions, which have been positively associated to radio activity and jet ejections, no type-B QPO has been observed to date. \begin{table*} \centering \caption{Log of the 5 RXTE/PCA observations analyzed in this work} \label{log_obs} \begin{tabular}{c c c c c c} \hline \hline N$^{\circ}$ & Obs. Id.& Date & Starting MJD & Exp. (s) & Classification\\ \hline 1 & 10408-01-35-00 & 1996 Sep 09 & 50348.271 & 9448& $\mu$\\ 2 & 20402-01-45-03 (\#1, 2, 3) & 1997 Sep 09 & 57700.250 & 10038& $\beta$\\ 3 & 20402-01-53-00 & 1997 Oct 31 & 50752.013 & 9656& $\beta$\\ 4 & 20402-01-53-01 \& 20402-01-53-02(\#1) & 1997 Nov 04-05 & 50756.412 & 6322&$\mu$\\ 5 & 20402-01-59-00 & 1997 Dec 17 & 50799.091 & 9784& $\beta$\\ \hline \end{tabular} \flushleft Obs. 4 is composed of two orbits from two separate observations\\ \# indicates the number of the RXTE orbit within the observation\\ Classification is from Belloni et al. (2000). \end{table*} In this paper we present the discovery with RXTE of the type-B QPO in the X-ray light curve of GRS 1915+105. The QPO was present during fast spectral transitions that we identify with the HIMS to SIMS transition observed in other BHCs. | \label{par:discussion} We analyzed 5 RXTE/PCA observations collected during the first two years of the mission. In the power density spectra of all five observations we detected several peaks which we identify with two different types of QPO already seen in many other BHCs: the type-C and type-B QPOs. This is the first identification of a type-B QPO in GRS 1915+105. To detect it, we looked in detail at spectral transitions in observations characterized by a fast and intense variability. Spectral transitions in GRS 1915+105 are usually very fast, often occurring on timescales of $\sim$ seconds. This is at variance with most of other black-hole binaries in which spectral transitions are observed to last hours or days. In order to study the spectral transitions in GRS 1915+105 we performed an energy-dependent timing analysis by averaging power spectra on the pattern the source recursively tracks in the 3-dimensional hardness-hardness-intensity diagram. Applying this method, we found a type-B QPO in all five observations. In all of them, we detected the type-B QPO together with the type-C, in region \# 3 of the HHID (see Tab. \ref{tab:zone}). In three of them, we also detected a type-B QPO alone, in region \# 4 of the HHID. Two of these observations belong to class $\mu$ and one to class $\beta$, which excludes any relation between the class of variability and the presence of the type-B QPO. This means that the presence of the long hard intervals (which differentiate the $\beta$-class from the $\mu$-class light curves) does not influence the fast timing properties of the source outside these intervals. We could not find any property (as e.g. hardness, rate) correlated with the presence of the type-B QPO alone in region {\em 4} of the HHID. In three observations we also found a type-B bump in region {\em 2}. No correlations were found neither between the presence of the type-B bump and the type-B QPO in region {\em 4} nor with the hardness and the count rate. \begin{figure} \begin{tabular}{c} \resizebox{7cm}{!}{\includegraphics{LAG_ALL_DEFINITIVO.ps}} \end{tabular} \vspace{-2.0cm} \caption{Phase lags of the detected QPOs in all observations. For each QPO we extracted the phase lag in a range centered at the QPO peak frequency and corresponding to the width itself ($\nu_p \, \pm \, FWHM/2$). Errors bar on the X-axis are not shown for clarity.} \label{fig:lag_all} \end{figure} \subsection{QPOs identification} \label{subpar:disc_identif} In order to identify the two types of QPO that we found in our data sets, we compare them with the known LFQPOs in BHCs. In particular we first analyze their position and behaviour in the Hardness-Intensity diagram (right panel of Figure \ref{fig:cd_hid_arrows}). When the source moves through regions \#{\em 1} and \#{\em 2} up to region \#{\em 3} the first QPO shows a behaviour very similar to that of type-C QPOs: its frequency is correlated with the count rate and is inversely correlated with the hardness. At a certain hardness, this QPO disappears. A second type of QPO is also detected: as in the case of type-B QPOs, this second QPO appears in a narrow frequency range (often around $\sim$6 Hz) and in a limited range in hardness. Its frequency and quality factor appear to be slightly correlated with the count rate, particularly when at its lowest frequencies (2.44-2.84 Hz). We interpret this as the first evidence of an increase of the coherence of the type-B QPO from $Q<2$ when at low frequencies to $Q>2$ when reaching frequencies around $\sim$6 Hz, possibly suggesting the presence of a resonance at this frequency (see Casella et al. 2004). The combined evolution of the QPOs in GRS 1915+105 is strongly reminiscent of the known behaviour of type-C and type-B QPOs in BHCs (see Casella et al. 2005 and references therein). To verify this identification, we plot in Figure \ref{fig:rms_1} the {\it rms} fractional variability of the detected QPOs as a function of their frequency for three energy bands. The two QPO types have a somewhat similar energy dependency (being stronger at high energy) but they clearly show different behaviours in these diagrams. In each of the three panels, two well-identified groups of points are evident. A comparison of these two groups with Figure \#3 of Casella et al. 2004 helps to classify the observed QPO in GRS 1915+105: the first group is diagonally spread across the plots, covering the whole frequency range between $\sim2$ and $\sim15$ Hz and a large range in {\it rms} (particularly at high energies, see the right panel of Figure \ref{fig:rms_1}). The second group is clustered both in frequency (between $\sim2.5$ and $\sim7$ Hz) and in fractional {\it rms}. The observed behaviour is clearly consistent with that known to be typical of type-C and type-B QPOs in BHCs (see Casella et al. 2004, 2005). \subsection{Phase lags} \label{subpar:disc_lag} The association between QPO-type and phase lag in literature is not conclusive: although an average behaviour can be identified (see Tab. \ref{tab:ABC}, Casella et al. 2005 and reference therein) there are a number of exceptions (see e.g. Belloni et al. 2005 and Homan et al. 2005a). Nevertheless we performed a phase-lag analysis in order to have a comprehensive view of the behaviour of the QPOs in GRS 1915+105. On the basis of the analyzed data it is not possible to characterize unambiguously the phase lag behaviour of any of the two types of QPO we observe: lags appear to vary between different observations, although both types show in average negative values of phase lag (see Fig. \ref{fig:lag_all}). This can possibly due to the fast variability of the analyzed light curves: we extracted power spectra over time intervals 2 seconds long, without applying any procedure to detrend the variability on longer time scales. However, this variability is very strong (see Fig. \ref{fig:mu_beta_class}), which results in a leakage at higher frequencies as strong as to actually dominate the phase-lag continuum. \subsection{GRS 1915+105 as a ``normal'' source} \label{subpar:disc_normal} The type-B QPO was detected in a few BHCs (see Casella et al. 2005), and associated to spectral transitions from HIMS to SIMS (Homan \& Belloni 2005, Belloni et al. 2005 and reference therein). Some authors (Gallo et al 2003; Fender, Belloni \& Gallo 2004) suggested a relation between these spectral transitions and the emission and collimation of transient relativistic radio jets: if we consider the type-B QPO as the signature of radio jet emission, we have this ``signature'' also in the prototypical galactic jet source. As already pointed out by Belloni et al. (2005), the non-detection of a type-B QPO in GRS 1915+105 was rather interesting, especially if you interpret the X-ray$/$radio correlation in this source and in other transients in the framework of the same model (Fender, Belloni \& Gallo 2004). In GRS 1915+105 the association between X-ray and radio activity is well known (Pooley \& Fender 1997, Mirabel et al. 1998, Fender \& Belloni 2004 and references therein). Belloni et al. 2005 suggested that the elusiveness of this QPO (in GRS 1915+105) could be due to high velocity of the movement of the source through the HID. The result presented in this work thus strengthens the interpretation of GRS 1915+105 in the framework of the same model of other BHCs: a type-B QPO appears in correspondence of spectral transition from the B-state to the A-state (Transition, Region {\em 3}, Stars in Figure \ref{licu_symbols}) and when the source is in the A-state (Region {\em 4}, Upward Triangles in Figure \ref{licu_symbols}). In the light curve in Figure \ref{licu_symbols} we see state oscillations CTAB TAB TAB that we can identify as fast passages between the HIMS and the SIMS of the other BHCs (Casella et al. 2004; Fender, Belloni \& Gallo 2004). The type-B QPO appears also in correspondence of the transition from the B-state to the C-state (Squares-Stars-Circles in Figure \ref{licu_symbols}), therefore not in an oscillation event but in a transition from a soft to a hard state. Furthermore, GRS 1915+105 is at present the heaviest black hole (although uncertainties on BH mass values are rather large, see McClintock \& Remillard 2006 and references therein) in which a type-B QPO has been observed at 6 Hz (when its quality factor $Q>2$). This strengthens the already known independence of this type of QPO on the mass of the compact object (see Casella et al. 2005). \subsection{X-ray$/$radio association} \label{subpar:disc_radio} Klein-Wolt et al. (2002) made a detailed study of simultaneous radio$/$X-ray observations of GRS 1915+105, focusing in particular on radio oscillation events, and found that $\beta$ class observations are usually associated to strong radio oscillations, while $\mu$ class observations are usually associated to a weak steady radio activity. These authors point out that the long hard intervals (which are the main macroscopic difference between the two variability classes) appear to be directly linked to the production of radio oscillations. We found the type-B QPO in both classes $\beta$ and $\mu$. We could not find any timing property (on time scales shorter than 2 seconds) clearly differentiating the two classes. This apparently weakens the link between the presence of the type-B QPO and radio activity. Unfortunately, the lack of radio coverage during the analyzed RXTE observations does not allow any conclusive result. In order to clarify this issue it will be fundamental to extend our analysis to RXTE observations for which a radio coverage is available. | 7 | 10 | 0710.3030 |
0710 | 0710.3340_arXiv.txt | We report the first determination of a distance bracket for the high-velocity cloud (HVC) complex~C. Combined with previous measurements showing that this cloud has a metallicity of 0.15 times solar, these results provide ample evidence that complex~C traces the continuing accretion of intergalactic gas falling onto the Milky Way. Accounting for both neutral and ionized hydrogen as well as He, the distance bracket implies a mass of 3--14\tdex6~\Msun, and the complex represents a mass inflow of 0.1--0.25~\Msunpyr. We base our distance bracket on the detection of \CaII\ absorption in the spectrum of the blue horizontal branch star SDSS\,J120404.78+623345.6, in combination with a significant non-detection toward the BHB star BS\,16034-0114. These results set a strong distance bracket of 3.7--11.2~kpc on the distance to complex~C. A more weakly supported lower limit of 6.7~kpc may be derived from the spectrum of the BHB star BS\,16079-0017. | \par The evolution of galaxies is strongly driven by the gas in the interstellar medium. There is strong evidence for the infall of new material that provides fuel for galaxy growth. This gas may originate in accreted satellite galaxies, as gas tidally pulled out of passing galaxies, or from pristine intergalactic gas. The cool, infalling clouds appear to be embedded in an extended (100--200~kpc radius) hot Corona (Sembach et al.\ 2003). Indirect evidence for infalling gas is provided by two arguments: (a) At the current rate of star formation, all of the ISM will be turned into stars within about a Gyr. (b) The narrowness of the distribution of metallicities of long-lived stars implies that the metallicity of the ISM remains more or less constant over a Hubble time, which can happen if there is a continuing inflow of low-metallicity material with a present-day rate of about 1~\Msunpyr. Item (b) is known as the ``G-dwarf problem'' (van den Bergh 1961). Using the infall hypothesis to solve it has been the subject of much theoretical work (see e.g.\ Pagel 1997 for a good summary). Continuing infall is essential in detailed numerical modeling of the chemical evolution of the Galaxy and the development of abundance gradients (e.g.\ Chiappini et al.\ 2001 and references therein). Infall of low-metallicity gas also seems necessary to reproduce the relatively high abundance of deuterium measured in the local interstellar medium (Linksy et al.\ 2006). \par Direct observational evidence for infalling low-metallicity gas is provided by the high-velocity clouds (HVCs; see reviews by Wakker \& van Woerden 1997; Richter 2006). Subsolar metallicities have now been determined for eleven clouds (see van Woerden \& Wakker 2004 for a summary). In particular, the metallicity of complex~C is well established as 0.15 times solar (see summary by Fox et al.\ 2004). Complex~C also has a high deuterium abundance (Sembach et al.\ 2004). Distance brackets have been more elusive, with just one known before 2006 (8--10~kpc for complex~A -- van Woerden et al.\ 1999a; Wakker et al.\ 2003). Thom et al.\ (2006) derive an 8.8~kpc upper limit for cloud WW\,35, while in a separate paper (paper~I, Wakker et al.\ 2007), we present new results for two HVCs (9.8--15.1~kpc for complex~GCP and 5.0--11.7~kpc for the Cohen Stream). In this letter we report a distance bracket for the HVC covering the largest sky area -- complex~C. We summarize our method in Sect.~2. The data are described in Sect.~3, the results in Sect.~4, while in Sect.~5 we summarize the implications. | \par Forty years after the first attempt (Prata \& Wallerstein 1967), we report the first successful detection of interstellar \CaII\ H and K absorption from HVC complex~C. This sets an upper limit on the distance of core CIII (left side in Fig.~1) of 11.2~kpc. For core CI (right side in Fig.~1) we find a lower limit of 3.7~kpc, possibly 6.7~kpc. Although the stars are 27\deg\ apart on the sky, it is still safe to conclude that complex~C is located at Galactocentric radius $<$14~kpc, and lies high above the Galactic plane ($z$=3--9~kpc). A more precise determination requires a lower limit for core CIII and an upper limit for CI. \par Integrating $N$(\HI) across the cloud, we estimate $M$(\HI) as 0.7--6\tdex{6}~\Msun. \Ha\ emission has also been detected (Tufte et al.\ 1998). We can assume either that the H$^+$ and \HI\ are thorougly mixed or that the H$^+$ originates in a photoionized skin around the cloud. In either case, the observed \Ha\ intensity suggests that there is roughly as much ionized as neutral gas. \par We can also estimate the mass inflow associated with complex~C, using a method described in paper~I. Including the neutral and ionized hydrogen, as well as a 40\% contribution from helium, we derive that complex~C represents about 0.1--0.25~\Msunpyr\ of infalling gas. This is a substantial fraction of the theoretically required amount of 1~\Msunpyr. Other HVCs may contribute the rest, but we have not yet determined distances and metallicities for the most likely candidates. \par From our results, we conclude that the mystery of the distances to the HVCs is beginning to be solved. The evidence shows that several HVCs are located in the upper reaches of the gaseous Galactic Halo and that they contribute significantly to the inflow of metal-poor gas onto the Galaxy. Once the mass inflow rate is constrained from observations of a sufficient number of HVCs, the next step will be to determine their three-dimensional structure, so that we can use their velocities and galactic location to derive orbits and solve the outstanding mystery of their ultimate origins. \bigskip Acknowledgements \par B.P.W., D.G.Y, R.W. and T.C.B. acknowledge support from grant AST-06-07154 awarded by the US National Science Foundation. T.C.B. also acknowledges NSF grants AST 04-06784 and PHY-02-16783; Physics Frontier Center/Joint Institute for Nuclear Astrophysics (JINA). \par Some of the data presented were obtained at the W.M. Keck Observatory, which is operated as a scientific partnership among the California Institute of Technology, the University of California and the National Aeronautics and Space Administration. The Observatory was made possible by the generous financial support of the W.M. Keck Foundation. \par The William Herschel telescope is operated on the Island of La Palma by the Isaac Newton group in the Spanish Observatorio del Roque de los Muchachos of the Instituto de Astrophysica de Canarias. \par Funding for the SDSS and SDSS-II 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. The SDSS Web Site is http://www.sdss.org/. The SDSS is managed by the Astrophysical Research Consortium for the Participating Institutions. The Participating Institutions are the American Museum of Natural History, Astrophysical Institute Potsdam, University of Basel, University of Cambridge, Case Western Reserve University, University of Chicago, Drexel University, Fermilab, the Institute for Advanced Study, the Japan Participation Group, Johns Hopkins University, the Joint Institute for Nuclear Astrophysics, the Kavli Institute for Particle Astrophysics and Cosmology, the Korean Scientist Group, the Chinese Academy of Sciences (LAMOST), Los Alamos National Laboratory, the Max-Planck-Institute for Astronomy (MPIA), the Max-Planck-Institute for Astrophysics (MPA), New Mexico State University, Ohio State University, University of Pittsburgh, University of Portsmouth, Princeton University, the United States Naval Observatory, and the University of Washington. | 7 | 10 | 0710.3340 |
0710 | 0710.1345_arXiv.txt | The results obtained from a study of the mass distribution of 36 spiral galaxies are presented. The galaxies were observed using Fabry-Perot interferometry as part of the GHASP survey. The main aim of obtaining high resolution H$\alpha$ 2D velocity fields is to define more accurately the rising part of the rotation curves which should allow to better constrain the parameters of the mass distribution. The H$\alpha$ velocities were combined with low resolution HI data from the literature, when available. Combining the kinematical data with photometric data, mass models were derived from these rotation curves using two different functional forms for the halo: an isothermal sphere and an NFW profile. For the galaxies already modeled by other authors, the results tend to agree. Our results point at the existence of a constant density core in the center of the dark matter halos rather than a cuspy core, whatever the type of the galaxy from Sab to Im. This extends to all types the result already obtained by other authors studying dwarf and LSB galaxies but would necessitate a larger sample of galaxies to conclude more strongly. Whatever model is used (ISO or NFW), small core radius halos have higher central densities, again for all morphological types. We confirm different halo scaling laws, such as the correlations between the core radius and the central density of the halo with the absolute magnitude of a galaxy: low luminosity galaxies have small core radius and high central density. We find that the product of the central density with the core radius of the dark matter halo is nearly constant, whatever the model and whatever the absolute magnitude of the galaxy. This suggests that the halo surface density is independent from the galaxy type. | Rotation curves are a fundamental tool for studying the dynamics and mass distribution in galaxies. The distribution of the total mass can then be compared with the distribution of visible light, assuming a certain mass-to-light (M/L) ratio. Observations have clearly established that dark matter is needed for explaining the rotation velocities observed in the outer parts of spirals. The dark matter is most often considered as being distributed in a spherical dark halo but its density profile, especially at small radii, is still a matter of debate (Blais-Ouellette et al. 2001, de Blok \& Bosma 2002, Swaters et al. 2003, Navarro 2004, Graham et al. 2006, Kusio de Naray 2006, Hayashi et al. 2007). However, mass models of spiral and dwarf galaxies have well-known degeneracies (Barnes et al. 2004) that prevent a unique mass decomposition, the most important being the unknown value of the stellar mass-to-light ratio (Dutton et al. 2005). Unfortunately, stellar population models (e.g. de Jong \& Bell 2006) still cannot predict accurately the (M/L) values of the stellar disks based on their colors. The main problem comes from the fact that the normalization of this relation (color vs M/L) depends critically on the shape of the stellar IMF at the low mass end. It is well known that the faint stars contribute significantly to the mass but not to the luminosity and color of the stellar disks. This means that it will prove difficult to tighten this relation and lessen the effect of the disk-halo degeneracy. Cosmological numerical simulations favor cuspy dark halos, although the value of the inner slope $\gamma$ of the radial density profile, where $\rho$ $\propto$ r $^{\gamma}$ (see equation 6 of section 4) depends on the authors, with $\gamma$ = -1.5 for Moore et al. 1999 as well as Fukushige \& Makino 2001 or -1.0 for Navarro, Frenk \& White 1997 (hereafter NFW). Recent simulations (Graham et al. 2006) have extrapolated inner logarithmic profile slope ranging from -0.2 to -1.5, with a typical value at 0.1 kpc around -0.7. Most observers conclude however that $\gamma$ is closer to 0.0 than to -1.0 and that mass models give better results with an isothermal (or pseudo-isothermal) sphere halo rather than a NFW profile (e.g. Blais-Ouellette et al. 2001, de Blok \& Bosma 2002, Swaters et al. 2003, de Blok et al. 2003, de Blok 2005, Kassin et al. 2006a). Most of these observations are based on the rotation curves of dwarf and Low Surface Brightness (LSB) galaxies (Kuzio de Naray et al. 2006). It is still unclear if this is true also for High Surface Brightness (HSB) systems and for all morphological types. One problem in this cusp-core debate is that numerical simulations mainly predict the halo density profile shape at the time of formation while galaxies are observed after many Gyrs of evolution. The problem is that internal dynamics and interaction between the dark halo and the luminous disk (e.g. adiabatic contraction: Dutton et al. 2005) and interaction with the environment (Maccio et al. 2007) may have altered the shape of the halo density profile. Moreover, the shape of the gravitational potential in CDM halos may explain the core-like halo density profiles seen in LSB systems. Hayashi et al. 2007 suggest that galactic disks may be forming in elliptical gravitational potential. This could result in significant non-circular motions in systems such as LSBs which would mimic constant density cores. Thus, taking into account the 3D shape of the dark mass distribution could reconcile the constant density cores observed in LSBs with the predicted cuspy mass profiles of CDM halos. Finally, it is unclear if CDM simulations, mainly obtained to compare with the large scale structure, have sufficient resolution to reliably probe the kpc scales necessary to compare with the observationally derived dark matter halo density profiles on the scale of a few halo core radii. As discussed by Navarro (2004), unfortunately rotation curves constraints are strongest where numerical simulations are the least reliable. In fact, rotation curves are usually compared with extrapolations of the simulation data that rely heavily on the applicability and accuracy of fitting formulae such as the NFW profile to regions that may be compromised by numerical artifact. This is especially true for LSB and dwarf galaxies (for instance Dutton et al. 2005). Blais-Ouellette et al. 1999 and Barnes et al. 2004 have shown the necessity of optical integral field spectroscopy to accurately determine the rotation curves in the inner parts of spiral and dwarf galaxies, for which the HI data are affected by beam smearing (Swaters et al. 2000, van den Bosch et al. 2000). While the use of two-dimensional data does not necessarily alter the halo parameters derived from optical long-slit data it however decreases the uncertainties by roughly a factor of 2 (de Naray et al. 2006). Blais-Ouellette et al. 2001 and Blais-Ouellette et al. 2004 also pointed out the great sensitivity of the mass distribution parameters to the inner rotation curve. The ideal rotation curve will therefore combine high resolution optical data, for the inner part, with radio data, for the outer part extending well beyond the luminous disk. | Our study, based on 36 galaxies of different morphological types, confirms the result already claimed by other authors about the shape of the dark matter halo in the center of spiral galaxies, namely that its density profile is probably closer to an isothermal sphere profile than to an NFW profile: DM halos are rather flat than cuspy. Our mass models (ISO and NFW) are compared when fitting the rotation curve as a whole, however a comparison of the chi2 values limited to the central regions would be still clearer since the inner slope of the RC found with the NFW model is systematically steeper and above the data points, compared with the ISO model. Thus far, most of the results found in the literature concerned dwarf or Low Surface Brightness galaxies. 9 galaxies of our sample are LSB (UGC 2304, 5272, 5279, 5789, 7524, 7699, 9465, 11707 and 12060, with types going from Scd to Im) and, for all of them, we systematically find better results with the ISO model (smaller $\chi^{2}$) than with NFW, except UGC 2304 for which both are equivalent. Interestingly, our study suggests that this holds for most spirals since the 36 galaxies of this study have morphological types ranging from Sab to Im. For almost all of them, the best fit is obtained with the ISO model (the reverse is found only for 4 galaxies out of 36). Also, no significant difference can be seen when comparing the quality of the fits obtained with the NFW and the ISO model as a function of the morphological type. However, a difference can be seen in the way the rotation curves are decomposed into several components, with the halo being the main component for late types and the disk (or disk + bulge) the main component for early types. This result, although needing to be confirmed with a larger sample, merely reflects the well known fact that later type galaxies are more dark matter dominated. Finally, we confirm different halo scaling laws seen previously by other authors such as C\^ot\'e et al (2000), Kormendy \& Freeman 2004 and Barnes et al. 2004. Among those, it appears clearly that low luminosity galaxies have small core radius and high central density, the product of the two parameters being nearly constant with absolute magnitude. This means that the galaxy halo surface density is independent of galaxy type or luminosity. Trends are also seen for the core radius as a function of luminosity and color but should be confirmed with a larger sample. | 7 | 10 | 0710.1345 |
0710 | 0710.4560_arXiv.txt | We study the effect of primordial nongaussianity on large-scale structure, focusing upon the most massive virialized objects. Using analytic arguments and N-body simulations, we calculate the mass function and clustering of dark matter halos across a range of redshifts and levels of nongaussianity. We propose a simple fitting function for the mass function valid across the entire range of our simulations. We find pronounced effects of nongaussianity on the clustering of dark matter halos, leading to strongly scale-dependent bias. This suggests that the large-scale clustering of rare objects may provide a sensitive probe of primordial nongaussianity. We very roughly estimate that upcoming surveys can constrain nongaussianity at the level $|\fnl|\lesssim 10$, competitive with forecasted constraints from the microwave background. | One of the fundamental predictions of standard (single-field, slow-roll) inflationary cosmology is that the density fluctuations in the early universe that seeded large-scale structure formation were nearly gaussian random (e.g.\ \cite{maldacena,Acquaviva:2002ud,Creminelli:2003iq,Lyth_Rodriguez,Seery_Lidsey}). Constraining or detecting non-gaussianity (NG) is therefore an important and basic test of the cosmological model. To the extent that it can be measured, gaussianity has so far been confirmed; the tightest existing constraints have been obtained from observations of the cosmic microwave background \cite{wmap3,Creminelli_wmap}. Recently, several inflationary models have been proposed which predict a potentially observable level of nongaussianity, see \eg \cite{ArkaniHamed:2003uz,Bartolo:2003jx,Lyth:2005du,Rigopoulos:2005ae, Allen:2005ye,Chen_DBI,Barnaby:2006cq,Barnaby:2006km,Barnaby:2007yb, Sasaki:2006kq,Chen:2006nt,Chen_Easther_Lim,Battefeld_Easther, Assadullahi:2007uw,Battefeld:2007en,Shandera} and \cite{Bartolo:2004if} for a review. Improved limits on NG would rule out some of these models; conversely, a robust detection of primordial nongaussianity would dramatically overturn standard inflationary cosmology and provide invaluable information about the nature of physical processes in the early universe. In this regard, there has been a resurgence in studying increasingly more sophisticated methods and algorithms to constrain (or, if we are lucky, detect) nongaussianity \cite{Babich:2005en,Babich_shape,Creminelli_estimators,Smith_Zaldarriaga,Fergusson_Shellard}. Nongaussianity manifests itself not only in the cosmic microwave background \cite{Falk_Ran_Sre,Luo_Schramm,Gangui_etal,Wang_Kam}, but also in the late-time evolution of large-scale structure. For example, detailed measurements of higher order correlations like the bispectrum or trispectrum of galaxy clustering could provide a handle on primordial nongaussianity \cite{Verde:2000vr,Scoccimarro:2003wn,Sefusatti:2007ih}. The abundance of galaxy clusters, the largest virialized objects in the universe, has also long been recognized as a sensitive probe of primordial NG \cite{Lucchin:1987yv,Robinson:1999se,Benson:2001hc,Matarrese:2000iz,verde01,Scoccimarro:2003wn,Komatsu:2003fd}. Because clusters are rare objects which form from the largest fluctuations on the tails of the density probability distribution, their abundance is keenly sensitive to changes in the shape of the PDF such as those caused by nongaussianity. Large statistical samples of massive clusters have already been compiled from wide-area optical imaging and spectroscopic surveys such as the Sloan Digital Sky Survey \cite{maxBCG,Koester:2007bg}, the Two-Degree Survey \cite{Eke:2004ve}, and from the Red Sequence Survey \cite{Yee:2007if} and from X-ray surveys using the Chandra and XMM-Newton observatories \cite{Willis:2005ag,Valtchanov:2003it}. Future missions, such as the Dark Energy Survey, Supernova/Acceleration Probe and Large Synoptic Survey Telescope, will detect and study tens of thousands of clusters, revolutionizing our understanding of cluster physics as well as providing important constraints on cosmology \cite{Haiman_Mohr_Holder,Majumdar_Mohr,Wang_Haiman, Battye_Weller,Lima_Hu_05, Marian_Bernstein,Takada_Bridle}. To exploit the potential of these upcoming surveys as probes of primordial nongaussianity, it is important to calibrate the effects of NG on the abundance and clustering of virialized objects. While no previous work has attempted to quantify the effects of NG on halo clustering, several groups over the past decade have constructed fitting formulae for the halo mass function \cite{Robinson_Baker,Robinson_Gawiser_Silk,MVJ}. All of this work, however, was analytic and relied on the validity of the Press-Schechter \cite{press-schechter} formalism, plus various further approximations. The resulting analytic estimates are, in general, rather cumbersome to compute and have questionable accuracy. As discussed below, the Press-Schechter model provides only a qualitative description of halo abundance, and fails to reproduce the halo mass function to within an order of magnitude over the mass and redshift range accessible to current and future cluster surveys. Therefore, analytic models for NG cluster abundance based on the Press-Schechter ansatz may not be sufficiently accurate. Given the high-quality data soon to be available, a much more precise calculation of cluster statistics will be required. Quite recently, two groups have attempted to quantify the mass function of clusters in NG models using N-body simulations \cite{Kang,Grossi}, reaching contradictory conclusions. In this paper, we use analytic arguments and numerical simulations to estimate the effect of NG on the abundance and clustering of virialized objects. Because N-body simulations can be expensive and there is a wide NG parameter space, we also strive to make our results useful to a cosmologist who is not necessarily equipped with the machinery or patience to run simulations or evaluate difficult analytic expressions. To this end, we provide a simple, physically motivated fitting formula for the halo mass function and halo bias, which we calibrate to our N-body simulations. Our main results are that the mass function and correlation function of massive halos can be significantly modified by primordial nongaussianity. We find a somewhat weaker effect of NG on the mass function than previous analytic estimates. We also show analytically and numerically that NG strongly affects the clustering of rare objects on large scales, implying that measurements of the large-scale power spectrum can place stringent bounds on NG. The plan of the paper is as follows. In Section \ref{anal} we derive analytic expressions for the abundance and clustering of rare peaks. In Section \ref{sec:method} we describe our N-body simulations, followed in Section \ref{sec:mf} by a discussion of our measured halo mass function, and our fitting formula for the mass function. In Section \ref{sec:pk} we present measurements of halo clustering within our simulations, and in Section \ref{sec:cosmo} we discuss cosmological implications of our findings. | We have quantified the effects of primordial nongaussianity on the abundance and power spectra of massive halos. Our two principal results are as follows. First, we have provided a new fitting formula for the halo mass function. The formula is based on matching halos in Gaussian and non-Gaussian simulations: for $\fnl>0$ the corresponding halos are more massive than in the Gaussian case, and vice versa. The formula is consistent with the measured mass function from our simulations to within $\sim10\%$ over the entire range of masses and redshifts that we consider. Being essentially a convolution of the Gaussian mass function and a Gaussian kernel (Eqs.~(\ref{eq:mf_conv})-(\ref{eq:rms_Mf_def})), the formula is also easy to use and does not require estimating the extreme tails of the nongaussian PDF of the density field. Our results also indicate that previous work based on Extended Press-Schechter type formulae overestimated the effects of nongaussianity on the abundance of halos by a factor of $\sim 2$ over the relevant mass scales. Secondly, we showed both analytically and numerically that nongaussianity (in the $\fnl$ model) leads to strong scale dependence of the bias of dark matter halos. We find remarkably good agreement between our analytic expression and our numerical results. Measurement of the power spectrum of biased objects therefore provides a new avenue to detect and measure nongaussianity. While cluster counts can constrain NG at a level comparable to existing CMB constraints, $|\fnl|\lesssim 100$, we found that future large-scale redshift surveys can potentially do much better, roughly $|\fnl|\lesssim 10$. We do not find significant degeneracies between $\fnl$ and dark energy parameters in our Fisher matrix calculations, either for mass function measurements or power spectrum measurements. More precise estimates will require considerably more sophisticated treatments than we have attempted in our illustrative examples above. We close this paper by considering, in light of our findings, the optimal methods for constraining NG of the $\fnl$ form. Measurements of the power spectrum would appear the most promising; observations of high redshift, highly clustered objects on large scales would allow the strongest constraints on the scale-dependent bias signature of $\fnl$. Fortunately, upcoming BAO surveys will likely provide the necessary observations of, e.g. luminous red galaxies (LRGs). Photometric surveys may also be useful in this regard. Since the effects of NG are most pronounced on large scales, rather than small scales, precise spectroscopic redshifts may not be necessary. Photometric redshifts with errors of order $\Delta z\approx 0.03$ have already been achieved for LRGs and for optically selected groups and clusters with prominent red sequences \cite{Padmanabhan:2004ic,Ilbert:2006dp,Yee:2007if}. At $z=0.5$, this corresponds to roughly 100 $h^{-1}$Mpc comoving, fairly small compared to the $\sim$Gpc scales where NG becomes most important. Since photometric surveys can cover wider areas more deeply than spectroscopic surveys, they may turn out to provide tighter bounds. Besides their abundance and clustering, the internal properties of massive halos may also be sensitive to nongaussianity. For instance, the concentrations and substructure content of massive halos have been found to depend upon primordial NG \cite{avila03}. Our simulations lacked sufficient force resolution to explore this in detail, but we note in passing that multiple groups find a tension between observations of massive lensing clusters and theoretical predictions for Gaussian perturbations \cite{rcs,arcs04,hennawi07,broadhurst08}. Another intriguing possibility for probing primordial NG is to use statistics of the largest voids in the universe. Just as the abundance and clustering of high density peaks are affected by nongaussianity, so are the same properties for deep voids (albeit with an opposite sign, c.f.\ Fig.~\ref{fig:slice_sims}). In a sense, because voids are not as nonlinear as overdense regions, their properties are more easily related to the initial Lagrangian underdensities whose statistics are straightforward to compute. Voids may be detected at high redshift as a deficit of Lyman-$\alpha$ forest absorption features in QSO spectra. The Sloan Digital Sky Survey (SDSS) has already measured spectra for high redshift QSO's over a roughly $\sim$8000 deg$^2$ area, corresponding to a volume of $\gtrsim 30 ({\rm Gpc}/h)^3$ \cite{sdss_lyaf}. Each QSO spectrum typically probes $\sim 400 h^{-1}$ Mpc, and the typical transverse separation between QSO sightlines in SDSS is $\sim 100 h^{-1}$ Mpc, (P.\ McDonald, priv.\ comm.) so measurements of the clustering of $\sim 10$ Mpc-sized voids on $\sim$ Gpc scales may already be feasible. Finally, we note that our conclusions are based on simulations implementing a very specific type of local primordial nongaussianity quantified by the $\fnl$ parameter. The validity of our conclusions in the context of other type of primordial nongaussianity is the subject of ongoing studies. | 7 | 10 | 0710.4560 |
0710 | 0710.2493_arXiv.txt | {The annihilations of WIMPs produce high energy gamma-rays in the final state. These high energy gamma-rays may be detected by IACTs such as the H.E.S.S. array of Imaging Atmospheric Cherenkov telescopes. Besides the popular targets such as the Galactic Center or galaxy clusters such as VIRGO, dwarf spheroidal galaxies are privileged targets for Dark Matter annihilation signal searches. H.E.S.S. observations on the Sagittarius dwarf galaxy are presented. The modelling of the Dark Matter halo profile of Sagittarius dwarf is discussed. Constraints on the velocity-weighted cross section of Dark Matter particles are derived in the framework of Supersymmetric and Kaluza-Klein models. The future of H.E.S.S. will be briefly discussed. \PACS{ {98.70.Rz}{Gamma-rays : observations} \and {95.35.+d}{Dark Matter} } % } % | \label{intro} H.E.S.S. (High Energy Stereoscopic System) is an array of four Imaging Atmospheric Cherenkov Telescopes (IACTs) located in the Khomas Highlands of Namibia at an altitude of 1800 m above sea level. Each telescope has an optical reflector of 107 m$^2$ composed of 382 round mirrors \cite{bernlohr}. The Cherenkov light is emitted by charged particles from electromagnetic showers initiated by the interaction of the primary gamma-rays in the Earth's upper atmosphere. The light is collected by the reflector which focuses it on a camera made of 960 fast photomultiplier tubes (PMTs) with individual field of view of 0.16$^{\circ}$ in diameter \cite{vincent}. Each PMT is equipped with a Winston cone to maximize the light collection and limit the background light. The total field of view of the H.E.S.S. instrument is 5$^{\circ}$ in diameter. The stereoscopic technique allows for an accurate reconstruction of the direction and energy of the primary gamma-rays as well as for an efficient rejection of the background induced by cosmic ray interactions. The energy threshold is about 100 GeV at zenith and the angular resolution is better than 0.1$^{\circ}$ per gamma-ray. The point source sensitivity is $\rm 2 \times 10^{-13} cm^{-2}s^{-1}$ above 1 TeV for a 5$\sigma$ detection in 25 hours \cite{crabe}.\\ The H.E.S.S. instrument is designed to detect very high energy (VHE) gamma-rays in the 100 GeV - 100 TeV energy regime and investigate their possible origin. Scenarios beyond the Standard Model of particle physics predict plausible WIMP (weakly interacting massive particle) candidates to account for the Cold Dark Matter. Among these are the minimal supersymmetric extension of the Standard Model (MSSM) or universal extra dimension (UED) theories. With R-parity conservation, SUSY models predict the lightest SUSY particle (LSP) to be stable. In various SUSY breaking scenarios, the LSP is the lightest neutralino $\rm \tilde{\chi}$. In Kaluza-Klein (KK) models with KK-parity conservation, the lightest KK particle (LKP) is stable \cite{KK}, the most promising being the first KK mode of the hypercharge gauge boson, $\rm \tilde{B}^{(1)}$. The annihilation of WIMPs in galactic halos may produce gamma-ray signals detectable with Cherenkov telescopes (for a review, see \cite{bertone}). Generally, their annihilations will lead to a continuum of gamma-rays with energy up to the WIMP mass resulting from the hadronization and decay of the cascading products, mainly from $\pi^0$'s generated in the quark jets. The gamma-ray flux from DM particle annihilations of mass m$_{DM}$ in a spherical halo can be expressed as: \begin{equation} \frac{d\Phi(\Delta\Omega,E_{\gamma})}{dE_{\gamma}} = \frac{1}{4\pi}\,\frac{\langle \sigma v \rangle}{m^2_{DM}}\frac{dN_{\gamma}}{dE_{\gamma}}\,\times\,\bar{J}\Delta\Omega \end{equation} which is a product of a particle physics term containing the velocity-weighted cross section $\langle \sigma v \rangle$ and the differential gamma-ray spectrum $dN_{\gamma}/dE_{\gamma}$, and an astrophysics term $\bar{J}$ given by: \begin{equation} \bar{J} = \frac{1}{\Delta\Omega}\int_{\Delta\Omega}\int_{l.o.s}\rho^2 ds \end{equation} This term corresponds to the line-of-sight-integrated squared density of the DM distribution which is averaged over the solid angle integration region spanned by the H.E.S.S. point spread function (PSF).\\ Plausible astrophysical targets have been proposed to search for DM, from local galactic objects to extragalactic objects such as galaxy clusters. This paper reports on three recent results on targets relevant to indirect dark matter search: the Galactic Center, the center of the VIRGO cluster (M87) and the Sagittarius dwarf spheroidal galaxy. | \label{conclusion} The H.E.S.S. collaboration has studied potential targets for DM annihilations. The TeV $\gamma$-ray energy spectrum measured by H.E.S.S. in the Galactic Center region is unlikely to be interpreted in terms of WIMP annihilations. Constraints on the velocity-weighted annihilation cross-section have been derived in the case of a NFW DM halo profile. Neither pMSSM nor KK models can be ruled out. The detection of a temporal variability of the VHE signal from the M87 nucleus in the VIRGO galaxy cluster excludes the bulk of the gamma-ray signal to come from DM annihilations. H.E.S.S. observed the Sagittarius dwarf spheroidal galaxy and no significant gamma-ray excess has been detected. Two DM halo modellings of Sgr have been investigated to derive contraints on the velocity-weighted annihilation cross-section. Some models can be ruled out in the case of a cored profile.\\ Dark matter searches will continue and searches with H.E.S.S. 2 will start in 2009. The phase 2 will consist of a new large 28 m diameter telescope located at the center of the existing array. With the availability of the large central telescope H.E.S.S. 2, the analysis energy threshold will be lowered down to less than 80 GeV and will allow to explore more supersymmetric models. | 7 | 10 | 0710.2493 |
0710 | 0710.2941_arXiv.txt | Although the occurrence of steady supercritical disk accretion onto a black hole has been speculated about since the 1970s, it has not been accurately verified so far. For the first time, we previously demonstrated it through two-dimensional, long-term radiation-hydrodynamic simulations. To clarify why this accretion is possible, we quantitatively investigate the dynamics of a simulated supercritical accretion flow with a mass accretion rate of $\sim 10^2 L_{\rm E}/c^2$ (with $L_{\rm E}$ and $c$ being, respectively, the Eddington luminosity and the speed of light). We confirm two important mechanisms underlying supercritical disk accretion flow, as previously claimed, one of which is the radiation anisotropy arising from the anisotropic density distribution of very optically thick material. We qualitatively show that despite a very large radiation energy density, $E_0\gsim 10^2L_{\rm E}/4\pi r^2 c$ (with $r$ being the distance from the black hole), the radiative flux $F_0\sim cE_0/\tau$ could be small due to a large optical depth, typically $\tau\sim 10^3$, in the disk. Another mechanism is photon trapping, quantified by ${\bm v}E_0$, where ${\bm v}$ is the flow velocity. With a large $|{\bm v}|$ and $E_0$, this term significantly reduces the radiative flux and even makes it negative (inward) at $r < 70 r_S$, where $r_S$ is the Schwarzschild radius. Due to the combination of these effects, the radiative force in the direction along the disk plane is largely attenuated so that the gravitational force barely exceeds the sum of the radiative force and the centrifugal force. As a result, matter can slowly fall onto the central black hole mainly along the disk plane with velocity much less than the free-fall velocity, even though the disk luminosity exceeds the Eddington luminosity. Along the disk rotation axis, in contrast, the strong radiative force drives strong gas outflows. | Recently, very bright objects which may be undergoing supercritical (or super-Eddington) accretion flows have successively been found. Good examples are ultraluminous X-ray sources (ULXs; \citeauthor{WMM01} \citeyear{WMM01}; \citeauthor{Ebisawa03} \citeyear{Ebisawa03}; \citeauthor{Okajima06} \citeyear{Okajima06}). These are pointlike off-center X-ray sources whose X-ray luminosity significantly exceeds the Eddington luminosity of a neutron star \citep{Fabbiano89}. Because of substantial variations, it is reasonable to assume that the ULXs are single compact objects powered by accretion flows \citep{Makishima00}. If so, there are two possibilities to account for large luminosities exceeding the Eddington luminosity for a mass of 100 $M_\odot$: sub-critical accretion onto an intermediate-mass black hole (IMBH) and supercritical accretion onto a stellar-mass black hole. We support the latter possibility, since through the fitting to several {\it XMM-Newton} EPIC data of ULXs, which have been claimed as good IMBH candidates, we have found evidence of supercritical flows \citep{Vierdayanti06}. Another interesting group is narrow-line Seyfert 1 galaxies \citep[see][for a review]{Boller04}. Because of their relatively small black hole masses, they have in general large Eddington ratios ($L/L_{\rm E}$ with $L$ and $L_{\rm E}$ being the luminosity and the Eddington luminosity, respectively), and some of them seem to fall in the slim-disk regimes (\citeauthor{Mineshige00} \citeyear{Mineshige00}; \citeauthor{Kawaguchi03} \citeyear{Kawaguchi03}; see also \citeauthor{Wang99} \citeyear{Wang99}) Despite growing evidence indicating the existence of supercritical accretion flows in the universe, theoretical understanding is far from being complete. It is well known that any spherically accreting object, irrespective of the nature of the central source, cannot emit above the Eddington luminosity, since otherwise significant radiative force will prevent accretion of the gas. If we examine detailed radiation-matter interactions in the interior, however, we notice that the situation is not so simple. Actually, radiation produced at the very center cannot immediately reach the surface (i.e., photosphere), since photons generated at the center should suffer numerous scatterings with accreting material and thus take a long time to reach the surface. If the matter continuously falls, and if the mass accretion timescale is shorter than the mean travel time for photons to reach the surface (the diffusion timescale), photons at the core may not be able to go out. This is the so-called photon-trapping effect \citep[e.g.,][]{Begelman78,HC91}. Here we define the trapping radius, inside which radiation-matter interaction is so frequent that photons are trapped within the accretion flow. Inside this trapping radius, therefore, radiative flux can be negative (i.e., inward). Thus, the apparent luminosity is reduced as compared with the case without photon trapping. Nevertheless, the concept of the Eddington luminosity is still valid, since far outside the trapping radius radiative flux should be outward and its absolute value should be less than $L_{\rm E}/4\pi r^2$, where $r$ is the distance from the central black hole, in the quasi-steady state. Radiation-hydrodynamic (RHD) simulations of spherically symmetric supercritical accretion flows have been performed by \citet{BK83}. The situation may differ in the case of disk accretion, since the radiation field is not isotropic due to inhomogeneous matter distribution. That is, matter can fall predominantly along the disk plane, whereas radiation can go out along its rotating axis, where matter density is low. In other words, the main directions of the inward matter flow and the outward radiative flux are not parallel to each other in disk accretion, leading to a situation in which the radiative force does not completely counteract the gravitational force. There is room for the possibility of a supercritical accretion flow with super-Eddington luminosity. Based on such an argument, many researchers have speculated about the occurrence of supercritical flows in disk accretion systems. \citet{SS73} discussed the possibility of supercritical disk accretion based on a one-dimensional steady model \citep[see also][]{MRT76}. They mentioned that the mass accretion rate would not be steady but oscillate if the mass accretion rate exceeded the critical rate; otherwise, a part of the accreting matter might be ejected from the disk as a disk wind. Radiatively driven outflows from supercritical disks were investigated by \citet{Meier79}, \citet{JR79}, and \citet{Icke80}. Despite a long history in the study of supercritical disk accretion flows, the occurrence of steady supercritical disk accretion has not yet been accurately verified. Similar simulations have been performed since the 1980s, but all of them calculated only the initial transient phase. Their conclusions then were not general, since the back-reactions (i.e., enhanced radiation pressure), which may inhibit steady flow, were not accurately evaluated. Although the radiative force predominantly has an effect in the vertical direction, it should also have an effect on the material within the disk. Hence, in the direction parallel to the disk plane, the situation may be similar to or more severe than the case of spherical accretion. This is because the radiative force, together with the centrifugal force, may possibly overcome the gravitational force. To study the possibility of supercritical disk accretion flows, precise and quantitative research treating both supercritical disks and outflows, and which also takes into consideration multi-dimensional effects, is needed. \citeauthor{O05} (\citeyear[][hereafter Paper I]{O05}) have confirmed the occurrence of quasi-steady supercritical disk accretion onto a black hole by two-dimensional RHD simulations. The motivation of the present study is to investigate the physical mechanisms which make supercritical disk accretion possible. For this purpose we examine quantitatively the flow motion and force fields via the radiative flux, rotation, and gravity of a supercritical disk accretion flow onto a central black hole, based on the two-dimensional RHD simulation data from Paper I. Through detailed inspection of the results, it will be possible to clarify the physics behind supercritical disk accretion flows. In \S 2 we plot the spatial distributions of several key quantities which control flow dynamics. A discussion is given in \S 3. | \subsection{Structure of the Supercritical Disk Accretion Flow} We summarize our simulation results in a schematic picture (see Fig. \ref{pic}). \begin{figure*} \epsscale{0.87} \plotone{Fig07_astroph.eps} \caption{Schematic picture of the supercritical disk accretion flow around a black hole. The gas motion is shown on the right: the high-velocity outflow ({\it solid arrows}) and the slow accretion flow ({\it dashed arrows}). The left side indicates the radiative fluxes in the comoving frame ({\it black arrows}) and the rest frame ({\it white arrows}). \label{pic} } \end{figure*} In this figure the radiative flux in the comoving frame, $F_0^r$, is positive (outward) except in the vicinity of the black hole. However, the radiative flux in the rest frame, $F^r (\sim F_0^r+4v_r E_0/3)$, is negative (inward) via photon trapping, $v_r E_0<0$, in the trapping region. In the vicinity of the black hole, both $F_0^r$ and $F^r$ are negative. Matter slowly accretes inside the disk, since the sum of the radiative and centrifugal forces is nearly balanced with the gravitational force. Here we stress again that the radiative force counteracts the gravitational force in spite of the trapping region because $F_0^r>0$. Radiatively driven high-velocity outflows appear above and below the disk. In the very vicinity of the black hole, the gas is accelerated inward by the radiative force and the gravitational force. As far as the physical quantities around the equatorial plane are concerned, the simulated profiles of the density, temperature, and radial as well as rotational velocities roughly agree with the prediction of the slim-disk model \citep{Abramowicz88}. Such features have already been shown in Figure 11 in Paper I. However, only about $10\%$ of the injected mass can accrete onto the black hole, and an almost equal amount of matter is ejected as high-velocity outflows. The mass accretion rate is not constant in the radial direction and decreases near the black hole (see Fig. 6 in Paper I). Thus, we conclude that the simulated flows do not perfectly agree with the slim-disk model with regard to the whole structure of the flow. \citet{HD07} have investigated local maximum values of the accretion rate in the supercritical disk accretion flows. In their paper they focused only on the force balance in the vertical and radial directions around the equatorial plane. Multi-dimensional effects were not taken into consideration. They revealed that the vertical radiative force limits the maximum accretion rate at the inner disk region, leading to a decrease of the accretion rate with a decrease of the radius. Their results imply that the disk loses mass via the radiatively driven outflows and the mass accretion rate decreases at the inner region. Such tendencies agree with our results. As shown in Figure \ref{vr}, our simulations show that radiatively driven outflows form above and below the disk. The mass accretion rate decreases with a decrease of the radius (see Fig. 6 in Paper I). \subsection{Dependency of the Mass Accretion Rate} In the present study, focusing on numerical simulations in which the mass input rate at the outer boundary is set to be $\dot{M}_{\rm input}=10^3L_{\rm E}/c^2$, we show that the radiative force is attenuated in the disk region via large optical thickness, which makes supercritical disk accretion possible. Such dilution of the radiative force would effectively operate even if the mass input rate (mass accretion rate) varied. In fact, simulations with mass input rates of $3\times 10^2L_{\rm E}/c^2$ and $3\times 10^3L_{\rm E}/c^2$ show that $E_0/\rho$ is almost independent of the mass input rate, although the radiation energy density goes up as the mass input rate increases. That is, the dynamics is not sensitive to the precise value of $\dot{M}_{\rm input}$, as long as it largely exceeds the critical value. So far, we have studied steady accretion flows. Although a highly supercritical disk ($\dot{M}_{\rm input}>3\times 10^2L_{\rm E}/c^2$) is quasi-steady, it has been revealed that a moderately supercritical disk [$\dot{M}_{\rm input}=(10-10^2)L_{\rm E}/c^2$] is unstable and exhibits limit-cycle behavior (\citeauthor{O06} \citeyear{O06}, \citeyear{O07}; see also \citeauthor{SH75} \citeyear{SH75}; \citeauthor{SS76} \citeyear{SS76}). The luminosity goes up and down around the Eddington luminosity. \citet{O07} has reported that the time-averaged mass, momentum, and kinetic energy output rates via the outflow, the mass accretion rate, and the disk luminosity increase as the mass input rate increases, $\propto \dot{M}_{\rm input}^{0.7} -\dot{M}_{\rm input}^{1.0}$ for $\alpha=0.5$ and $\propto \dot{M}_{\rm input}^{0.4} -\dot{M}_{\rm input}^{0.6}$ for $\alpha=0.1$. \subsection{Future Work} As we have already mentioned in \S 3.1, the sum of the accreting matter and the matter ejected as high-velocity outflows is $20\%$ of the injected mass, and $80\%$ of the injected matter is ejected from the computational domain as low-velocity outflows, whose velocities do not exceed the escape velocity at the outer boundary. Since such outflowing matter tends to be accelerated by the radiative force even at the outside of the computational domain, it would be blown away from the system. However, a part of the outflowing matter might return to the vicinity of the black hole through the disk region, since the radiative force does not exceed the gravity near the equatorial plane. Numerical simulations with larger computational domains would make this point clear. Whereas the resulting mass accretion rate onto the black hole is around $10^2 L_{\rm E}/c^2$ in the present simulations, \citet{HD07} have indicated that the mass accretion rate can increase up to $10^4 L_{\rm E}/c^2$. They have reported that the vertical force balance breaks down via a strong radiative force if the mass accretion rate exceeds this limit. However, even in such a case, the matter might accrete onto the black hole, although the strong radiative force would produce powerful outflows. To investigate the maximum value of the accretion rate is an outstanding issue. We should perform numerical simulations with larger computational domains, since the trapping region is expected to expand with the increase of the mass accretion rate. We reveal in the present paper that photons generated deep inside the disk are effectively trapped in the flow, leading to supercritical disk accretion. Although the magnetic fields are not solved in our simulations, magnetic buoyancy might play an important role in the transportation of matter, as well as photons, toward the disk surface (\citeauthor{Parker75} \citeyear{Parker75}; \citeauthor{SR84} \citeyear{SR84}; \citeauthor{SC89} \citeyear{SC89}). Magnetic buoyancy might lead to photon generation near the disk surface if the magnetic fields rise quickly without dissipation deep inside the disk. Thus, magnetic buoyancy would dilute the photon-trapping effect. A photon bubble instability, which is induced in the magnetized, radiation-pressure-dominated region, might also suppress the photon trapping \citep{Begelman02,Turner05}. In these cases the enhanced radiative force would more effectively accelerate the matter around the disk surface, working to decrease the mass accretion rate. However, the magnetic fields might prevent such acceleration if they strongly tie the matter near the disk surface with the disk matter. In the disk region the matter might easily accrete toward the black hole, since the radiation energy density decreases. Global radiation-magnetohydrodynamic (RMHD) simulations would make these points clear. Local RMHD simulations of accretion flows have been performed by \citet{Turner03} and \citet{HKS06}. In addition, it is thought that disk viscosity has magnetic origins (\citeauthor{HBS01} \citeyear{HBS01}; \citeauthor{MMM01} \citeyear{MMM01}; for a review see \citeauthor{Balbus03} \citeyear{Balbus03}). Hence, we should stress again that RMHD simulations are very important to more realistically investigate viscous accretion flows, although an $\alpha$-viscosity model is employed in the present study. | 7 | 10 | 0710.2941 |
0710 | 0710.2807_arXiv.txt | We have identified two moderately hot ($\sim$18000--22000\,K) white dwarfs, SDSS\,J1228+1040 and SDSS\,J1043+0855, which exhibit double-peaked emission lines in the CaII\,$\lambda\lambda$\,8600 triplet. These line profiles are unambiguous signatures of gaseous discs with outer radii of $\sim1R_\odot$ orbiting the two white dwarfs. Both stars accrete from the circumstellar material, resulting in large photospheric Mg abundances. The absence of hydrogen emission from the discs, and helium absorption in the white dwarf photospheres demonstrates that the circumstellar material is depleted in volatile elements, and the most likely origin of these gaseous rings are tidally disrupted rocky asteroids. The relatively high mass of SDSS\,J1228+1040 implies that planetary systems can not only form around $4-5\,M_\odot$ stars, but may also survive their post main-sequence evolution. | While more than 250 extra-solar planets orbiting main-sequence stars have been discovered, the destiny of planetary systems in the late stages of the evolution of their host stars is very uncertain, and so far no planet has been found around a white dwarf. Infrared excess detected around a number of white dwarfs has been interpreted as the signature of dust discs \citep[e.g.][]{zuckerman+becklin87-1, becklinetal05-1, kilicetal06-1, vonhippeletal07-1}. The photospheres of these white dwarfs are rich in metals \citep{zuckermanetal07-1}, indicating ongoing accretion from the circumstellar material. The likely origin of these debris discs are tidally disrupted asteroids \citep{jura03-1}, and hence they represent a close association with the planetary systems that the white dwarf progenitor stars may have had. However, while the infrared excess detected around these white dwarfs can be explained in terms of a dusty debris disc, the observations actually do not provide any strong constraint on the geometry of the source of the infrared light \citep[e.g.][]{reachetal05-1}. We summarise here our recent discovery of two white dwarfs in the SDSS spectroscopic data base which exhibit double-peaked emission lines of Ca\,II$\lambda\lambda$8600, unambiguously confirming a circumstellar disc-like structure \citep{gaensickeetal06-3, gaensickeetal07-1} | It has been suggested that planetary systems may survive the post-main sequence evolution of their host stars \citep{burleighetal02-1, villaver+livio07-1}, however, no planet has yet been discovered around a white dwarf. The detection of debris discs from rocky asteroids around white dwarfs, such as SDSS\,J1228+1040 and SDSS\,J1043+0855 and the cooler white dwarfs with dusty debris discs lends strong support to the survival hypothesis. It appears also entirely possible that these white dwarfs may still have planetesimal objects or planets. SDSS\,J1228+1040 is particular interesting, as its relatively high mass implies that its progenitor must have had a mass of $\sim4\,M_\odot$ \citep{dobbieetal06-1}, suggesting that also short-lived massive stars may be host to planetary discs. This is in accordance with the detection of a relatively massive debris disc around the young A2e star MWC\,480 \citep{manningsetal97-1}. | 7 | 10 | 0710.2807 |
0710 | 0710.2813_arXiv.txt | { We report on a sensitive search for mid-infrared molecular hydrogen emission from protoplanetary disks. We observed the Herbig Ae/Be stars UX Ori, HD 34282, HD 100453, HD 101412, HD 104237 and HD 142666, and the T Tauri star HD 319139, and searched for H$\,_2~0-0~S(2)~(J=4-2)$ emission at 12.278 micron and H$\,_2~0-0~S(1)~(J=3-1)$ emission at 17.035 micron with VISIR, ESO-VLT's high-resolution mid-infrared spectrograph. None of the sources present evidence for molecular hydrogen emission at the wavelengths observed. Stringent 3$\sigma$ upper limits to the integrated line fluxes and the mass of optically thin warm gas ($T=$ 150, 300 and 1000 K) in the disks are derived. The disks contain less than a few tenths of Jupiter mass of optically thin H$_2$ gas at 150 K, and less than a few Earth masses of optically thin H$_2$ gas at 300 K and higher temperatures. We compare our results to a Chiang and Goldreich (1997, CG97) two-layer disk model of masses 0.02 M$_{\odot}$ and 0.11 M$_{\odot}$. The upper limits to the disk's optically thin warm gas mass are smaller than the amount of warm gas in the interior layer of the disk, but they are much larger than the amount of molecular gas expected to be in the surface layer. If the two-layer approximation to the structure of the disk is correct, our non-detections are consistent with the low flux levels expected from the small amount of H$_2$ gas in the surface layer. We present a calculation of the expected thermal H$_2$ emission from optically thick disks, assuming a CG97 disk structure, a gas-to-dust ratio of 100 and T$_{\rm gas}$ = T$_{\rm dust}$. We show that the expected H$_2$ thermal emission fluxes from typical disks around Herbig Ae/Be stars are of the order of 10$^{-16}$ to 10$^{-17}$ erg s$^{-1}$ cm$^{-2}$ for a distance of 140 pc. This is much lower than the detection limits of our observations (5 $\times$ $10^{-15}$ erg s$^{-1}$ cm$^{-2}$). H$_2$ emission levels are very sensitive to departures from the thermal coupling between the molecular gas and dust in the surface layer. Additional sources of heating of gas in the disk's surface layer could have a major impact on the expected H$_2$ disk emission. Our results suggest that in the observed sources the molecular gas and dust in the surface layer have not significantly departed from thermal coupling (T$_{\rm gas}$/T$_{\rm dust}<$ 2) and that the gas-to-dust ratio in the surface layer is very likely lower than 1000. | Circumstellar disks surrounding low- and intermediate-mass stars in their pre-main sequence phase are the locations where planets presumably form. Such protoplanetary disks are composed of gas and dust. Their mass is initially dominated by gas (99\%), specifically by molecular hydrogen (H$_2$), which is the most abundant gas species. The dust constitutes only a minor fraction of the total disk mass, however, it is the main source of opacity. Consequently, most of what we know observationally about protoplanetary disks has been inferred from studies of dust emission and scattering (for recent reviews see Henning et al. 2006; Natta et al. 2007; Dullemond et al. 2007; Watson et al. 2007). In order to understand the structure and evolution of protoplanetary disks, it is necessary to study their gaseous content independently from the dust. For example, a basic physical quantity such as the disk mass is conventionally deduced from dust continuum emission at millimeter wavelengths assuming an interstellar gas-to-dust ratio of 100 (e.g., Beckwith et al. 1990; Henning et al. 1994). If dust is physically processed in the disk, as should be the case in order to form planets, the gas-to-dust ratio must change with time. The disk dissipation time scale - another fundamental quantity required to disentangle proposed giant planet formation scenarios (Pollack et al. 1996; Boss et al. 1998) - is deduced from observations of thermal infrared excess emission produced by dust grains (Haisch et al. 2001, 2005). Although recent studies (e.g., Sicilia-Aguilar et al. 2006) suggest a parallel evolution of the dusty and gaseous components, it still remains to be demonstrated that the gaseous disks disappear over the same time scale as the infrared excess. A variety of spectral diagnostics of the gas disk have been observed from the UV to the millimeter (see reviews by Najita et al. 2007; Dutrey et al. 2007). However, the only diagnostic that is potentially able to probe the warm gas in the regions where giant planets are thought to form is the mid-infrared (mid-IR) emission lines of H$_2$. UV and near-infrared diagnostics only probe the innermost regions of the disk (R $<$ few AU), and mm and sub-mm diagnostics are limited to probe the cold outermost regions of the disk (R $>$10 AU). \begin{table*} \caption{Summary of the observations.} % \label{table:observations} % \centering % \begin{tabular}{@{}l c l c c c c l c c c c c c c @{}} % \hline\hline % Star & $\lambda$ & Date & U.T. & $t_{exp}$ & Airmass $^a$ & V$_{\oplus\,{\rm rad}}\,^b$ & \multicolumn{2}{c}{Calibrator $^c$} & $t_{exp}$ & \multicolumn{2}{c}{Airmass} \\ & [$\mu$m] & & [hh:mm] & [s] & & [km s$^{-1}$] & \multicolumn{2}{c}{ } & [s] & \multicolumn{2}{c}{ } \\ \hline % UX Ori & 12.278 & 11 January 2006 & 02:28 & 3600 & 1.0 - 1.2 & 2.06 & HD 36167 & (P)(F) & 1000 & 1.1(P) & 1.2(F) \\ & 17.035 & 4 January 2007 & 01:54 & 3600 & 1.0 - 1.0 & 4.50 & HD 25025 & (P) & 600 & 1.0 (P) & ...\\ HD 34282 & 12.278 & 10 January 2006 & 04:01 & 3600 & 1.0 - 1.5 & 2.17 & HD 36167 & (P)(F) & 1000 & 1.1(P) & 1.7(F) \\ HD 100453 & 12.278 & 22 March 2006 & 07:27 & 3600 & 1.2 - 1.5 & 24.45 & HD 89388 & (P)(F) & 600 & 1.5 (P) & 2.0 (F) \\ & 17.035 & 27 March 2006 & 06:26 & 3600 & 1.3 - 1.7 & 22.96 & HD 89388 & (P) & 1000 & 1.5 (P) & ... \\ HD 101412 & 12.278 & 30 March 2006 & 04:40 & 3600 & 1.2 - 1.4 & 23.82 & HD 91056 & (P)(F) & 600 & 1.3 (P) & 1.7 (F) \\ & 17.035 & 30 March 2006 & 01:27 & 3600 & 1.3 - 1.2 & 24.02 & HD 91056 & (F) & 1000 & ... & 1.3 (F) \\ HD 104237 & 12.278 & 10 February 2006 & 05:43 & 3600 & 1.7 - 1.7 & 26.09 & HD 92305 & (P)(F) & 600 & 1.7 (P) & 1.8 (F) \\ & 17.035 & 12 February 2006 & 06:36 & 3600 & 1.6 - 1.7 & 26.15 & HD 92305 & (F) & 1000 & ... & 1.8 (F) \\ HD 142666 & 17.035 & 28 February 2006 & 06:28 & 3600 & 1.1 - 1.0 & 27.06 & HD 169916 & (P) & 1000 & 1.1 (P) & ...\\ HD 319139 & 17.035 & 30 March 2006 & 08:30 & 3600 & 1.1 - 1.0 & 22.47 & HD 169916 & (P) & 1000 & 1.2 (P) & ...\\ \hline \end{tabular} \flushleft $^a$ The airmass interval is given from the beginning to the end of the observations. $^b$ V$_{\oplus\,{\rm rad}}$ is the expected velocity shift of the spectra due to the reflex motion of the Earth around the Sun and the radial velocity of the star. $^c$ The standard stars were observed prior (P) and/or immediately following (F) the science observations. \end{table*} Molecular hydrogen is by far the most abundant molecular species in protoplanetary disks. Unfortunately, H$_2$ is one of the most challenging molecules to detect. Since H$_2$ is a homonuclear molecule, it lacks a permanent dipole moment and its transitions are thus electric quadrupole in nature. The small Einstein coefficients, characteristic of the quadrupole transitions, imply that H$_2$ emission features are very weak. In addition, in the case of protoplanetary disks, the H$_2$ lines are not sensitive to the warm gas in the optically thick regions where the dust and gas are at equal temperature. Practical observational challenges also have to be faced. The mid-IR H$_2$ emission from the disk needs to be detected on the top of a strong mid-IR continuum. From the ground, the mid-infrared windows are strongly affected by sky and instrument background emission, and the H$_2$ transitions at 12 and 17 $\mu$m lie close to atmospheric absorption lines highly dependent on atmospheric conditions. The advent of high spectral resolution spectrographs mounted on larger aperture telescopes, allows for the first time the study of H$_2$ emission from the ground, but the search is still limited to bright targets. From space, the problems of atmosphere absorption are alleviated and the $J = 2-0$ feature at 28 $\mu$m is visible. However, the beam sizes are large and the spectral resolution of space mid-infrared spectrographs are usually low when compared to ground-based facilities, therefore, they are not very appropriate for small line-to-continuum ratios. H$_2$ emission from protoplanetary disks in the mid-IR has been reported from ISO observations (Thi et al. 2001). However, subsequent ground-based efforts (Richter et al. 2002; Sheret et al. 2003; Sako et al. 2005) did not confirm the ISO detections. H$_2$ emission in the mid-IR has been searched towards debris disks using Spitzer (Hollenbach et al. 2005, Pascucci et al. 2006, Chen et al. 2006) with no detection reported. Most recently, Bitner et al. (2007) and Martin-Za\"idi et al. (2007) reported the detection of mid-IR H$_2$ emission in two Herbig Ae/Be stars (AB Aur and HD 97048) from the ground, and Lahuis et al. (2007) reported the detection of mid-IR H$_2$ emission in 6 T Tauri stars with Spitzer. Here, we report on a sensitive search for molecular hydrogen emission from protoplanetary disks. We observed six southern nearby (d$<$400 pc) Herbig Ae/Be stars and one T Tauri star, employing the Very Large Telescope (VLT) imager and spectrometer for the mid-infrared (VISIR; Lagage et al. 2004)\footnote{http://www.eso.org/instruments/visir}, ESO's VLT mid-infrared high-resolution spectrograph, and searched for H$_2~0-0~S(1)~(J=3-1)$~emission at 17.035~$\mu$m and H$_2~0-0~S(2)~(J=4-2)$ emission at 12.278~$\mu$m. The paper is organized as follows: in Sect. 2 we present the sample studied, and the details of how the observations were performed. In Sect. 3 we discuss the data reduction. In Sect. 4 we deduce upper limits to the H$_2$ fluxes, and using the optically thin approximation, we derive upper limits to the mass of warm (150 - 1000 K) gas in the disks. In Sect. 5 we discuss our results in the context of the Chiang and Goldreich (1997) two-layer disk model. Finally, we summarize our results and conclusions in Sect. 6. \begin{figure*} \centering \includegraphics[angle=0,width=0.95\textwidth]{7846fig1.eps} \caption{Spectra obtained for the H$_2$ 0--0 S(2) (J=4--2) line at 12.278 $\mu$m. The left panel shows a zoom to the -100 to 100 km s$^{-1}$ interval of the atmospheric corrected spectra. A Gaussian of {\it FWHM} = 15 km s$^{-1}$ and integrated line flux equal to the line-flux upper limits obtained is overplotted at the expected velocity shifted location (vertical dashed line, see Table 2). The central panel shows the full corrected spectra. Dotted lines show spectral regions strongly affected by telluric or standard star absorption features. The right panel shows the continuum normalized spectra of the standard star and the target before telluric correction. The uppermost right panel displays the sky spectrum from a half-chop cycle. The spectra are not corrected for the radial velocity of the targets. } \end{figure*} \begin{figure*} \centering \includegraphics[angle=0,width=0.95\textwidth]{7846fig2.eps} \caption{Spectra obtained for the H$_2$ 0--0 S(1) (J=3--1) line at 17.035 $\mu$m. The left panel shows a zoom to the -100 to 100 km s$^{-1}$ interval of the atmospheric corrected spectra. A Gaussian of {\it FWHM} = 15 km s$^{-1}$ and integrated line flux equal to the line-flux upper limits obtained is overplotted at the expected velocity shifted location (vertical dashed line, see Table 2). The central panel shows the full corrected spectra. Dotted lines show spectral regions strongly affected by telluric or standard star absorption features. The right panel shows the continuum normalized spectra of the standard star and the target before telluric correction. The uppermost right panel displays the sky spectrum from a half-chop cycle. The spectra are not corrected for the radial velocity of the targets. } \end{figure*} | We observed a sample of nearby pre-main sequence stars with evidence for cold ($T < 50$ K) gas disk reservoirs and searched for emission of the warm gas with $T > 150$ K, which is expected to be present in the inner region of these disks. None of the targets show any evidence for H$_2$ emission at 17.035 $\mu$m or at 12.278 $\mu$m. From the 3$\sigma$ upper limits of the H$_2$ line fluxes, we found stringent upper limits to the mass of optically thin warm H$_2$. The disks contain less than a few tenths of Jupiter mass of optically thin H$_2$ at 150 K, and less than a few Earth masses of optically thin H$_2$ at 300 K and higher temperatures. Assuming that T$_{\rm gas}$=T$_{\rm dust}$ and a gas-to-dust ratio of 100, we compared our results to models of disks employing a Chiang and Goldreich (1997) optically thick two-layer disk model of masses 0.02 M$_{\odot}$ and 0.11 M$_{\odot}$. The upper limits to the disk optically thin warm gas mass are smaller than the warm gas mass in the interior layer of the disk, but they are much larger than the amount of molecular gas expected to be in the surface layer. The amount of mass in the surface layer is very small ($<$ 10$^{-2}$ M$_J$) and almost independent of the total disk mass. We calculated the expected H$_2$ S(1) and H$_2$ S(2) line fluxes emitted from a two-layer disk for the low-mass and the high-mass cases assuming a distance of 140 pc, and LTE thermal emission. The predicted line fluxes of the two-layer disk model are of the order of $\sim$10$^{-16}-10^{-17}$ erg s$^{-1}$ cm$^{-2}$, much smaller than the detection limits of our observations (5 $\times$ $10^{-15}$ erg s$^{-1}$ cm$^{-2}$). {\it If the two-layer approximation to the structure of the disk is correct, we are essentially ``blind" to most of the warm H$_2$ in the disk because it is located in the optically thick interior layer of the disk. Our non-detections are explained because of the small flux levels expected from the little mass of H$_2$ present in the optically thin surface layer.} Naturally, the two-layer disk model is only an approximation of the real structure of a protoplanetary disk. The puffed-up inner rim, which could be an important contributor to the H$_2$ emission, is not included in our models, and in reality there will be a smooth transition zone between the disk hot surface layer and the cool disk interior, which again could contribute significantly to the H$_2$ emission. In addition, the surface layers of the disk are likely to have gas temperatures hotter than the dust temperatures or a gas-to-dust ratio higher than the canonical value of 100. Both effects could potentially increase the H$_2$ emission by significant amounts. We presented additional calculations in which the gas-to-dust ratio and the temperature of the molecular gas in the surface layer of the disk were increased. We showed that detectable S(1) and S(2) H$_2$ line flux levels can be achieved if T$_{\rm gas\,surf}$/T$_{\rm dust\,surf}>$ 2 and if the gas-to-dust ratio in the surface layer is greater than 1000. H$_2$ emission levels are very sensitive to departures from the thermal coupling between the molecular gas and dust in the surface layer. Our results suggest that in the observed sources the molecular gas and the dust in the surface layer have not significantly departed from thermal coupling and that the gas-to-dust ratio in the surface layer is very likely lower than 1000. A definitive interpretation of our results awaits the development of future, more sophisticated models. | 7 | 10 | 0710.2813 |
0710 | 0710.2955_arXiv.txt | We report on the discovery of three new pulsars in the first blind survey of the north Galactic plane (45$^{\circ}$ < l < 135$^{\circ}$ ; |b| < 1$^{\circ}$) with the Giant Meterwave Radio telescope (GMRT) at an intermediate frequency of 610 MHz. The timing parameters, obtained in follow up observations with the Lovell Telescope at Jodrell Bank Observatory and the GMRT, are presented. | Most pulsar surveys have been carried out with single dish telescopes where there is a trade-off between the collecting area and the beam-width, and consequently the rate of the survey. In a multi-element telescope such as the Giant Meterwave Radio Telescope (GMRT), a large number of smaller antennas can be combined to provide a high sensitivity and yet retain a relatively large beam-width. In this paper, we report on the discovery of three new pulsars in the first blind survey of the north Galactic plane (45$^\circ$ < l < 135$^\circ$ ; |b| < 1$^\circ$) with the GMRT at an intermediate frequency of 610 MHz, which represents the best trade-off between the increased flux density at low frequency for pulsars, interstellar scattering and dispersion and beam-width. The GMRT's multi-element nature was also exploited to determine the positions of the pulsars to an accuracy of 5 arcminutes and this technique is also described. | 7 | 10 | 0710.2955 |
|
0710 | 0710.0999_arXiv.txt | We investigate the fate of particle production in an expanding universe dominated by a perfect fluid with equation of state $p = \alpha\rho$. The rate of particle production, using the Bogolioubov coefficients, are determined exactly for any value of $\alpha$ in the case of a flat universe. When the strong energy condition is satisfied, the rate of particle production decreases as time goes on, in agreement to the fact that the four-dimensional curvature decreases with the expansion; the opposite occurs when the strong energy condition is violated. In the phantomic case, the rate of particle production diverges in a finite time. This may lead to a backreaction effect, leading to the avoidance of the big rip singularity, specially if $- 1 > \alpha > - \frac{5}{3}$. | The phenomenon of particle production in an expanding universe is due, essentially, to the fact that in curved space-time the vacuum is not unique \cite{birrel,jacobson}. As the universe evolves, and the curvature changes, the vacuum state also changes: the initial vacuum state, representing the state with no particle, becomes later a multiparticle state. The particle production is directly connected with the curvature of the universe. If the universe is spatially flat, the cosmic evolution leads asymptotically to a Minkowski space-time, where the phenomenon of particle production does not occur anymore. However, this is true only if the strong energy condition is verified: the energy density $\rho$ and the pressure $p$ must satisfy the relation $\rho + 3p > 0$. If the strong energy condition is violated the particle production should be zero initially, increasing as the universe evolves, in connection with the increase of the four-dimensional curvature with time when $\rho + 3p < 0$. \par In order to compute the particle production in a given cosmological model, it is necessary to fix an initial vacuum state. If the universe is always decelerating, there is no natural and unique choice for this, since all modes are initially outside the Hubble radius feeling strongly the curvature of the space-time. However, a natural choice for the initial vacuum state can be done for the case of an accelerating universe, since all physical modes are initially well inside the Hubble radius, and they behave as in a Minkowski space-time. This is one of the reasons why the primordial inflationary scenario is so successful theoretically: it encodes naturally a mechanism for quantum fluctuations, which will be later the seeds for the large scale structure of the universe \cite{jerome}. \par Observations indicate that we live today in an inflationary phase, since the universe is accelerating \cite{spergel,uzan}. Hence, the mechanism of particle production may play again a very important role today. Suppose we define a quantum state for a given field today. The question we want to address concerns the rate of particle production. In particular, we would like to know if such particle production is so important that it can alters, by back-reaction, the later evolution of the universe. The back-reaction due to quantum effects has already studied in the case of the cosmological constant \cite{ford2}. The back-reaction may be particularly relevant for the case the universe is dominated by a phantom fluid since, classically, in such a situation the universe will inevitably evolve towards a new singular state, called the big rip. We remember that the observational data favors somehow a phantom scenario today \cite{caldwell}. A phantom fluid has many special features. Local configurations of a phantom fluid leads to regular black holes \cite{kirill}. In cosmology, a universe dominated by a phantom fluid implies a future singularity that will occur in a future finite proper time \cite{frampton}. \par We will solve the rate of the particle production when the universe is flat and dominated by an equation of state $p = \alpha\rho$, for an arbitrary value of $\alpha$. A simplified scenario, not considering the different phase transitions that occur in the real universe, will be adopted. The calculations will be performed for a massless scalar field, for which the corresponding Klein-Gordon equation is solved. Through the quantization of the field, the rate of particle may be determined using the technique of Bogolioubov's coefficients. The general solution for this problem is expressed in terms of Bessel functions. The rate of particle production is determined exactly for any value of $\alpha$. The connection of this rate with the strong energy condition is established. It is shown that for a phantom fluid the rate of particle production diverges as the big rip is approached. This may lead to a back-reaction effect changing the effective equation of state of the universe, implying perhaps the avoidance of the big rip itself: the particle produced does not necessarily obeys the phantom equation of state and may become the dominant component of the universe if $ - 1 > \alpha > - \frac{5}{3}$. Hence, the back-reaction of the particle production may be an ineluctable mechanism to the avoidance of the big rip if the pressure is not excessively negative. The limiting case, $\alpha = - \frac{5}{3}$, has already been found before in a complete different context, that is, in the evolution of classical scalar perturbations in phantom cosmological models \cite{finelli,fabris}. \par This paper is organized as follows. In the next section we review the formalism of particle creation in a curved space-time. In section $3$ we apply the formalism to a perfect fluid cosmological model, and we determine the rate of particle creation for any value of $\alpha$. In section $4$ we present our conclusions. Even if many steps of the formalism used and of the calculation performed are quite well known, we will present them with some details in order to trace the main features of the final results. | We have evaluate the rate of creation of massless scalar particle in a universe dominated by a perfect fluid whose equation of state is given by $p = \alpha\rho$. An analytic expression has been found in terms of Hankel's functions. Since we have not considered a complete cosmological scenario, with a sequence of different phases, with an initial inflationary phase (as in standard cosmological model), the calculation performed here makes sense, strictly speaking, only when the strong energy condition is violated, that is, $\alpha < - 1/3$. For such a case, there is a natural initial vacuum state, from which the particle occupation number can be determined. However, we formally extended the calculation for any value of $\alpha$. It has been found that for those "dark energy" scenarios the rate of particle creation diverges as $t \rightarrow \infty$, that is, in the future infinity. \par We have payed special attention to the phantom scenario. The reason is that a universe dominated by a phantom fluid may develop a singularity in a finite future time, the so-called big rip. In this case, we have found that it is possible that the energy density associated to the particles created (which obey in present case an equation of state of the type $p_s = \rho_s$) becomes dominant over the phantom fluid if $- 1 > \alpha > - 5/3$. Hence, the big rip can be avoided if the pressure is not deeply negative. \par It is interesting to remark that a similar critical point, $\alpha = - 5/3$, has been found in the case of classical scalar perturbation \cite{fabris}. In that case, however, the evolution of scalar perturbation may destroy the conditions of homogeneity (necessary for the big rip) if $\alpha < - 5/3$, that is, if the pressure is negative enough. The result found here is exactly the opposite: quantum effects can be operative in the sense of destroying the conditions for the big rip if the pressure is not negative enough. It must be stressed, however, that the evaluation made in the present work must be complemented by a study of more general quantum fields and by a deeper thermodynamical analysis of the energy balance between the phantom fluid and the created particles as the big rip is approached. \newline \vspace{0.5cm} \newline {\bf Acknowledgements}:\\ We thank CNPq (Brazil) and the french/brazilian scientific cooperation CAPES/COFECUB (project number 506/05) for partial financial support. We thank specially J\'er\^ome Martin for his criticisms on the text. | 7 | 10 | 0710.0999 |
0710 | 0710.1167_arXiv.txt | A symplectic integrator algorithm suitable for hierarchical triple systems is formulated and tested. The positions of the stars are followed in hierarchical Jacobi coordinates, whilst the planets are referenced purely to their primary. The algorithm is fast, accurate and easily generalised to incorporate collisions. There are five distinct cases -- circumtriple orbits, circumbinary orbits and circumstellar orbits around each of the stars in the hierarchical triple -- which require a different formulation of the symplectic integration algorithm. As an application, a survey of the stability zones for planets in hierarchical triples is presented, with the case of a single planet orbiting the inner binary considered in detail. Fits to the inner and outer edges of the stability zone are computed. Considering the hierarchical triple as two decoupled binary systems, the earlier work of Holman \& Wiegert on binaries is shown to be applicable to triples, except in the cases of high eccentricities and close or massive stars. Application to triple stars with good data in the multiple star catalogue suggests that more than 50 \% are unable to support circumbinary planets, as the stable zone is non-existent or very narrow. | \label{sec:intro} It is well known that many of the stars in the solar neighbourhood exist in multiple systems. As the number of planetary surveys increases, planets are regularly being found not only in single star systems, but binaries and triples as well. For example, recently a hot Jupiter has been claimed in the triple system HD 188753 (Konacki 2005; but see also Eggenberger et al. 2007), and \citet{DB07} lists 33 binary systems and 2 other hierarchical triples known to harbour exoplanets. As the majority of work on planetary dynamics has been for single star systems, the dynamics of bodies in these multiple stellar systems is of great interest. At first sight, it might not seem likely to expect long term stable planetary systems to exist in binary star systems, let alone triples, but numerical and observational work is starting to show otherwise. In recent years, much study has been devoted to the stability of planets and planetesimal discs in binary star systems. There are several investigations of individual systems (e.g. work on $\gamma$ Cephei by \citealt{Dv03}, \citealt{Ha06} and \citealt{VE06}) as well as substantial amounts of work on more general limits for stability of smaller bodies in these systems. Notably, \citet{HW99} approach this problem by using numerical simulation data to empirical fit critical semimajor axes for test particle stability as functions of binary mass ratio and eccentricity. These general studies are of great use in the investigation of observed systems and their stability properties, giving an effective and fast method of placing limits on stability in the large parameter space created by observational uncertainties. The aim of this work is to investigate test particle stability in hierarchical triple star systems, and to see if any similar boundaries can be defined for them. To do this, a new method for numerically integrating planets in such systems is presented, following the ideas for binary systems presented by \citet{CQDL02}. Although there have been empirical studies of the stability of hierarchical star systems themselves, there do not appear to have been studies of small bodies in such systems. There is a great deal of literature concerning periodic orbits in the general three and four body problems (see e.g. \citealt{Sz67}), but these are almost entirely devoted to considerations of planetary satellites in single star systems, for example satellite transfer in the Sun-Earth-Moon systems. Also, while periodic orbits prove that stable solutions exist in these problems, they are of little practical use in determining general stability limits. The problem of planetary orbits in triple systems is more complex than for those in binary systems, as there are many different orbital configurations possible relative to the three stars. These are considered in Section~\ref{sec:orbits}, and a method for classifying them is described in order to simplify the discussion in this paper. The derivation of a method to numerically integrate such planets is then given in Section~\ref{sec:maths}. In Section~\ref{sec:stats} is a brief overview of the statistical properties of known triple systems, as a basis for deciding the parameters of the systems used in the numerical simulations. The results of numerical investigations into stability properties are then presented in Section~\ref{sec:sp} for one of the possible orbital types. | \label{sec:conclusion} The main achievement of this paper is the formulation of a symplectic integrator algorithm suitable for hierarchical triple systems. This extends the algorithm for binary systems presented by \citet{CQDL02}. The positions of the stars are followed in hierarchical Jacobi coordinates, whilst the planets are referenced purely to their primary. Each of the five distinct cases, namely circumtriple orbits, circumbinary orbits and circumstellar orbits around each of the stars in the hierarchical triple, requires a different splitting of the Hamiltonian and hence a different formulation of the symplectic integration algorithm. Here, we have given the mathematical details for each of the five cases, and presented a working code that implements the algorithm. As an application, a survey of the stability zones for circumbinary planets in hierarchical triples is presented. Here, the planet orbits an inner binary, with a more distant companion star completing the stellar triple. Using a set of numerical simulations, we found the extent of the stable zone which can support long-lived planetary orbits and provided fits to the inner and outer edges. The effect of low inclination on this boundary is minimal. A reasonable first approximation to a behaviour of a hierarchical triple is to regard it as a superposition of the dynamics of the inner binary and a pseudo-binary consisting of the outer star and a point mass approximation to the inner binary. If it is considered as two decoupled binary systems, then the earlier work of Holman \& Wiegert (1999) on binaries is applicable to triples, except in the cases of high eccentricities and close or massive stars. The implication here is that the addition of a stable third star does not distort the original binary stability boundaries. As mentioned, \citet{MW06} have shown that overlapping sub-resonances are the cause of the boundary in the binary case. It is reasonable to expect that in triples the same process is responsible, and the similarities between the binary and triple results support this theory. It is also expected that there is a regime in which the resonances from each sub-binary start to overlap as well, further destabilising the test particles. Evidence of this is the deviation from the binary results when the stars are close, massive and very eccentric, when resonances would be both stronger and wider. Since the parameter space investigated was chosen to reflect the observed systems it would seem to be a reflection on the characteristics of known triple stars that most lie in the decoupled regime. The relatively constant nature of the stellar orbits in the simulations is however a consequence of the test particle orbits being destabilised long before the stars are close enough to interact. By extension of all these arguments, it is expected that the binary criteria can be used to fairly accurately predict the stability zones in any hierarchical stellar system, no matter the number of stars. The results presented here can be used to estimate the number of known hierarchical triple systems that could harbour S(AB)-P planets. \citet{To97} lists 54 systems with semimajor axis, eccentricity and masses for both the inner and outer components. The mutual inclinations of most are not well known, but there are nine systems listed for which this angle can be determined. For five of these it is less than $15^\circ$, two are around $40^\circ$ and two are retrograde. Although a small sample it suggests that there are systems that fall within the low inclination regime investigated here. Using the criteria of \cite{HW99} and those found here for the position of the inner and outer critical semimajor axis the size of the coplanar stable region for each of these triples can be calculated. This can be considered an upper limit, since it is likely that very non-coplanar systems and those with significant eccentricities for both binary components will further destabilise planets. Of the 54 systems 13 are completely unstable to circumbinary planets according to the four parameter fits (compared to 11 using \citet{HW99}'s criteria). Figure \ref{fig:zone} shows a plot of the width of the stable region for the remaining systems. Interestingly, the majority seem to have very small stable zones, with 16 smaller than an au. Whether this is a feature of triple systems, or an observation bias is not apparent. It does indicate though that circumbinary planets are unlikely to exist in at least 50 \% of observable systems. \begin{figure} \centering \includegraphics[width=\textwidth]{fig14.eps} \caption{ Widths of the circumbinary stability zones for known triple systems in au and as a percentage of the area between the two sub-binaries, calculated using the four parameter fits derived in Section~\ref{sec:sp}. Note that using the criteria of \citet{HW99} gives almost identical results.} \label{fig:zone} \end{figure} | 7 | 10 | 0710.1167 |
0710 | 0710.1021_arXiv.txt | \noindent We test statistically the hypothesis that radio pulsar glitches result from an avalanche process, in which angular momentum is transferred erratically from the flywheel-like superfluid in the star to the slowly decelerating, solid crust via spatially connected chains of local, impulsive, threshold-activated events, so that the system fluctuates around a self-organised critical state. Analysis of the glitch population (currently 285 events from 101 pulsars) demonstrates that the size distribution in individual pulsars is consistent with being scale invariant, as expected for an avalanche process. The measured power-law exponents fall in the range $-0.13\leq a \leq 2.4$, with $a\approx 1.2$ for the youngest pulsars. The waiting-time distribution is consistent with being exponential in seven out of nine pulsars where it can be measured reliably, after adjusting for observational limits on the minimum waiting time, as for a constant-rate Poisson process. PSR J0537$-$6910 and PSR J0835$-$4510 are the exceptions; their waiting-time distributions show evidence of quasiperiodicity. In each object, stationarity requires that the rate $\lambda$ equals $- \epsilon \dot{\nu} / \langle\Delta\nu\rangle$, where $\dot{\nu}$ is the angular acceleration of the crust, $\langle\Delta\nu\rangle$ is the mean glitch size, and $\epsilon\dot{\nu}$ is the relative angular acceleration of the crust and superfluid. Measurements yield $\epsilon \leq 7 \times 10^{-5}$ for PSR J0358$+$5413 and $\epsilon \leq 1$ (trivially) for the other eight objects, which have $a < 2$. There is no evidence that $\lambda$ changes monotonically with spin-down age. The rate distribution itself is fitted reasonably well by an exponential for $\lambda \geq 0.25\,{\rm yr^{-1}}$, with $\langle \lambda \rangle = 1.3^{+0.7}_{-0.6}\,{\rm yr^{-1}}$. For $\lambda < 0.25\,{\rm yr^{-1}}$, its exact form is unknown; the exponential overestimates the number of glitching pulsars observed at low $\lambda$, where the limited total observation time exercises a selection bias. In order to reproduce the aggregate waiting-time distribution of the glitch population as a whole, the fraction of pulsars with $\lambda > 0.25\,{\rm yr^{-1}}$ must exceed $\sim 70$ per cent. | } Glitches are tiny, impulsive, randomly timed increases in the spin frequency $\nu$ of a rotation-powered pulsar, sometimes accompanied by an impulsive change in the frequency derivative $\dot{\nu}$. They are to be distinguished from timing noise, a type of rotational irregularity where pulse arrival times wander continuously, although there is evidence that timing noise is the cumulative result of frequent microglitches in certain pulsars \citep{cor85,dal95}. At the time of writing, 285 glitches in total have been detected in 101 objects ($\sim 6\%$ of the known radio pulsar population), the majority in the last four years, facilitated by the Parkes Multibeam Survey, refined multifrequency ephemerides, and better interference rejection algorithms \citep{hob02,kra03,kra05,lew05,jan06}. Efforts to analyse the data statistically have focused on the correlation of glitch activity with age \citep{mck90,she96,ura99,lyn00,wan00} and Reynolds number \citep{peralta06,mel07}, the post-glitch relaxation time-scale \citep{wan00,won01}, the size distribution \citep{mor93a,mor93b,peralta06}, and the correlation between glitch sizes and waiting times \citep{wan00,won01,mid06,peralta06}. \citet{hob02} reviewed the role of observational selection effects. Most glitching pulsars ($65\%$) have been seen to glitch once, but a minority glitch repeatedly; the current record holder is PSR J1740$-$3015, with 33 glitches. Of those objects which glitch repeatedly, most do so at unpredictable intervals, but two (PSR J0537$-$6910 and Vela) are quasiperiodic; Vela, in particular, has been likened to a relaxation oscillator \citep{lyn96}. The fractional increase in $\nu$ spans seven decades ($3\times 10^{-11} \leq \Delta\nu \leq 2\times 10^{-4}$) across the glitch population and as many as four decades in a single object (e.g.\ $7\times 10^{-10} \leq \Delta\nu \leq 2\times 10^{-6}$ in PSR J1740$-$3015). The spin-down age $\tau_{\rm c}= -\nu/(2\dot{\nu})$ of glitching pulsars spans four decades, from $1\times 10^3\,{\rm yr}$ to $3\times 10^7\,{\rm yr}$. In many respects, therefore, the glitch phenomenon is {\em scale invariant}. This striking property invites physical interpretation. Theories of pulsar glitches have focused mainly on the local microphysics of the superfluid in the stellar interior and its coupling to the solid crust, for example the strength of vortex pinning \citep{and75,jon98}, the rate of vortex creep \citep{lin96}, or the conditions for exciting superfluid turbulence \citep{per05,per06,mel07,and07}. Ultimately, however, the local microphysics must be synthesized with the global, {\em collective} dynamics in order to make full contact with observational data. (Likewise, a practical model of earthquakes must synthesize the microphysics of rock fracture with the macrodynamics of interacting tectonic plates.) For example, if approximately $10^{16} (\Delta\nu/1\,{\rm Hz})$ vortices unpin from crustal lattice sites in sympathy during a glitch, they must communicate rapidly across distances much greater than their separation. How? And why does the number that unpin fluctuate so dramatically (by up to four orders of magnitude) from glitch to glitch in a single pulsar, while always amounting to a small fraction ($\Delta\nu/\nu$) of the total? Such collective, scale invariant behavior is a generic feature of a class of natural and synthetic far-from-equilibrium systems, called self-organized critical systems, that are discrete, interaction dominated, and slowly driven, and that adjust internally via erratic, spatially connected {\em avalanches} of local, impulsive, threshold-activated, relaxation events \citep{jen98}. Such systems fluctuate around a stationary state towards which they evolve spontaneously, in which global driving balances local relaxation on average over the long term. The archetype of a self-organized critical system is the sandpile \citep{bak87}. In this paper, we study pulsar glitches as an avalanche process, as first proposed by \citet{mor93a}. After reviewing self-organised criticality in \S\ref{sec:gli2}, we define the statistical sample on which our study is based (\S\ref{sec:gli3}) and analyze the observed distribution of glitch sizes (\S\ref{sec:gli4}) and waiting times (\S\ref{sec:gli5}). Some implications for glitch physics are explored in \S\ref{sec:gli6}. We only include radio pulsars in the sample, to preserve its homogeneity, even though glitches have now been observed in anomalous X-ray pulsars (magnetars) as well \citep{dal03,kas03}. | } In this paper, we analyze the size and waiting-time distributions of pulsar glitches, taking advantage of the enlarged data set produced by the Parkes Multibeam Survey. We conclude that the data are consistent with the hypothesis that pulsar glitches arise from an avalanche process. In each of seven pulsars with $N_{\rm g} > 5$, the size distribution is consistent with being scale invariant across the observed range of $\Delta\nu$ (up to four decades), and the waiting-time distribution is consistent with being Poissonian. These features are natural if the system is driven globally at a constant rate (as the pulsar spins down), and each glitch corresponds to a locally collective, threshold activated relaxation of one of the many spatially independent, metastable stress reservoirs in the system (e.g. via a vortex unpinning or crust cracking avalanche). In two pulsars, PSR J0537$-$6910 and PSR J0835$-$4510, the dynamics may include a second, quasiperiodic component, comprising $\sim 20\%$ of the events. The size and waiting-time distributions of the quasiperiodic component are narrowly peaked, as expected for rare, system-spanning avalanches, which relax a large fraction of the total stress accumulated in the system. This two-component behavior is observed widely in self-organised critical systems, including experiments on magnetic flux vortices in type II superconductors, which are closely analogous to neutron star superfluids \citep{fie95}. The power-law exponent of the size probability density function differs from pulsar to pulsar, spanning the range $-0.13 \leq a \leq 2.4$. Calculating $a$ theoretically from first principles is a deep problem which has not yet been solved for other self-organised critical systems, let alone glitching pulsars, although some progress has been made on two-dimensional sandpiles using renormalization group techniques \citep{pie94,jen98}. In the mean-field approximation, which is exact in four or more dimensions, theoretical calculations on sandpiles (and other systems in their universality class) yield $a = 1.5$, whereas three-dimensional cellular automata output $a=1.3$ \citep{jen98}. The size distribution transmits two important lessons concerning the microphysics of glitches. First, the fact that $a$ differs from pulsar to pulsar implies that the strength and level of conservation of the local (e.g.\ pinning and intervortex) forces also differs \citep{ola92,fie95}. By contrast, in equilibrium critical systems like ferromagnets, $a$ depends only on the dimensionality of the system and its order parameter and is therefore universal. Second, except for the two pulsars which show quasiperiodicity, $a$ appears to vary smoothly with spin-down age, with $a\approx 1.2$ for the youngest pulsars (e.g.\ the Crab). Figure \ref{fig:gli9} depicts the trend between $a$ and $\tau_{\rm c}$. It is suggestive; after all, local pinning forces do depend on temperature and hence $\tau_{\rm c}$. Interestingly, however, there is no clear trend between $a$ and $\nu$, even though the mean vortex spacing (and hence intervortex force) is proportional to $\nu^{1/2}$. It will pay to study these trends more thoroughly as more glitch data is collected. An avalanche process predicts a specific relation between the distributions of glitch sizes $\Delta\nu$ and lifetimes $T$ (as opposed to waiting times $\Delta t$). Specifically, in a self-organized critical state, the lifetime probability density function is also a power law, $p(T) \propto T^{-b}$, with \begin{equation} b = 1 + (a-1)\gamma_2 / \gamma_3~. \label{eq:gli16} \end{equation} The constants $\gamma_2$ and $\gamma_3$ are defined such that the cardinality of an avalanche scales with its linear extent ($L$) as $L^{\gamma_2}$ and its lifetime (i.e.\ duration) scales as $L^{\gamma_3}$ \citep{jen98}. Both $\gamma_2$ and $\gamma_3$ depend on the effective dimensionality of the local forces and can be calculated numerically using a cellular automaton. In two dimensions, avalanches are compact, not fractal, and one has $\gamma_2=2$; in three dimensions, one has $2 < \gamma_2 < 3$. At present, radio timing experiments cannot measure $T$; most glitches are detected as unresolved, discontinuous, spin-up events with $T < 120\,{\rm s}$ \citep{mcc90}. \footnote{In the Crab, some spin-up events seem to be resolved, e.g.\ at epochs MJD 50260 ($T \approx 0.5\,{\rm d}$) and MJD 50489 (secondary spin up, $T \approx 2\,{\rm d}$) \citep{won01}. If these are rare but otherwise standard glitches originating from the long-$T$ tail of the lifetime distribution, it is puzzling that other, shorter, but still resolved (and presumably more common) spin-up events are not observed, e.g.\ with $T\sim 0.1\,{\rm d}$ or $0.01\,{\rm d}$. Alternatively, the events at MJD 50620 and MJD 50489 may have been triggered by a different physical mechanism.} In the future, however, single- and/or giant-pulse timing experiments with more sensitive instruments (e.g.\ the Square Kilometer Array) will test this prediction. If confirmed, it will independently corroborate the avalanche hypothesis. The mean glitching rates of the nine pulsars studied here are fairly narrowly distributed, spanning the range $0.35\,{\rm yr^{-1}} \leq \lambda \leq 2.6\,{\rm yr^{-1}}$. The probability density function for $\lambda$ is adequately fitted by an exponential, as for solar flare avalanches \citep{whe00}, with $\langle\lambda\rangle = 1.3_{-0.6}^{+0.7}$ {\rm yr}$^{-1}$, or by an exponential with a lower cutoff, at $\lambda_{\rm min} \approx 0.25$ {\rm yr}$^{-1}$. A theoretical derivation of $\langle\lambda\rangle$ from first principles is currently lacking, although estimates of how long it takes to crack the crust locally predict reasonable rates, if the critical strain angle approaches that of imperfect terrestrial metals \citep{alp96,mid06}. Figure \ref{fig:gli10} plots $\lambda$ versus $\tau_c$ for the nine pulsars examined individually in this paper. There is no significant trend. The data are consistent with the notion that old pulsars glitch less frequently than young pulsars \citep{she96}, but they are equally consistent with the notion that the glitching rate is independent of age. Many authors have searched for a correlation between waiting time and the size of the next glitch. Such a correlation appears to be absent from the data, e.g.\ Figure 17 in \citet{wan00} and Figure 10 in \citet{won01}. At first blush, this is surprising: the vortex unpinning and crust fracture paradigms, which are driven by the accumulation of differential rotation and mechanical stress respectively, seem to be natural candidates for a `reservoir effect'. Avalanche dynamics resolves this apparent paradox. The reservoir effect does operate locally, but the star contains many reservoirs, insulated from each other by relaxed zones, whose storage capacities evolve stochastically in response to the slow driver and avalanche history. During a glitch, a single reservoir (often small but sometimes large) relaxes at random via an avalanche, releasing its stored $\Delta\nu$ (and destabilizing neighboring reservoirs in preparation for the next glitch). Some of the $\Delta\nu$ is accumulated since the previous glitch, but the remainder is `borrowed' from earlier epochs, when some other reservoir relaxed instead. All self-organized critical systems share these dynamics; the waiting time is uncorrelated with the size of the next avalanche \citep{jen98}. The only exceptions are large, system-spanning avalanches, which always have roughly the same sizes and waiting times, and which account for $\sim 20\%$ of the glitches in PSR J0537$-$6910 and PSR J0835$-$4510. A corollary of the previous paragraph is that the total $\Delta\nu$ released in glitches up to some epoch is less than the total crust-superfluid differential rotation accumulated since that epoch, viz. \begin{equation} \sum_{i=1}^{N_{\rm g}} \Delta\nu_i \leq \epsilon |\dot{\nu}| \sum_{i=1}^{N_{\rm g}} \Delta t_{i}, \label{eq:gli17} \end{equation} where $\epsilon\dot{\nu}$ is the relative angular acceleration of the crust and superfluid due to electromagnetic spin down. The `staircase' described by (\ref{eq:gli17}) has been noted previously \citep{she96,lyn00}, both in quasiperiodic glitchers like PSR J0537$-$6910 [e.g.\ Figure 8 in \citet{mid06}], where the reservoir effect is obvious, and in Poisson glitchers like PSR J0534$+$2200, [e.g.\ Figure 12 in \citet{won01}], where the trend is more subtle because it reverts to the mean over long times, not after every glitch. Upon dividing (\ref{eq:gli17}) by $N_{\rm g}$, and averaging over long times, the inequality becomes an equality (provided there is no secular accumulation of differential rotation in the system) and we recover (\ref{eq:gli5}). It is fundamentally impossible to measure $\epsilon$ for individual pulsars with current data, because $\langle \Delta \nu \rangle$ is dominated by large (and therefore rare) glitches for $a < 2$. It is therefore wrong to assume stationarity over a typical, $40$-yr observation interval. Consequently, we are prompted to reassess the familiar correlation between activity and spin-down age \citep{she96}. Our definition of $\epsilon\dot{\nu}$ is identical to $\dot{\nu}_{\rm glitch}$ in \citet{lyn00} (but for individual objects, not in aggregate) and $A_{\rm g}$ in \citet{won01}. It is closely related to the original activity parameter defined by \citet{mck90}, which equals $N_{\rm g} \nu^{-1} \epsilon |\dot{\nu}|$. For PSR J0358$+$5413, we measure $\epsilon\leq 7 \times 10^{-5}$, lower than the {\it aggregate} value $0.017\pm0.002$ measured by \citet{lyn00} for objects with $\tau_{\rm c} > 10\,{\rm kyr}$ (binned by semi-decades in $\dot{\nu}$). \footnote{The aggregate value $\dot{\nu}_{\rm glitch}$ \citep{lyn00}, binned over semi-decades in $\dot{\nu}$, effectively averages together different pulsars. While this approach reduces the formal error bar on $\dot{\nu}_{\rm glitch}$, its physical interpretation is less straightforward, given the likelihood that $\epsilon$ is different in different pulsars.} Interpreted in terms of the vortex unpinning model, this result suggests that $0.007$--$2$ \% of the angular momentum outflow during spin down may be stored in metastable reservoirs on average over time. On the other hand, five other objects have $0.04 \leq \epsilon_{\rm max} \leq 0.8$, under the questionable assumption that the maximum physical size is $\Delta \nu_{\rm upper} = 2 \times 10^{-4}$ in all pulsars. Our data are therefore inadequate to update usefully the value $A_{\rm g}/|\dot{\nu}| = 1\times 10^{-5}$ measured by \citet{won01} for PSR J0534$+$2200. In the context of vortex unpinning, it has been argued that the aggregated $\epsilon$ measured by \citet{lyn00} partly corroborates the hypothesis that younger pulsars are still in the process of forming their capacitive elements, e.g.\ by creating pinning centers through crust fracture, while older pulsars have mostly completed the task \citep{alp96,won01}. However, the full picture is more complicated. Vela's quasiperiodic avalanches point to a richly connected network of reservoirs \citep{alp96}, yet its aggregated value $\dot{\nu}_{\rm glitch}$ is relatively low. On the other hand, the other quasiperiodic glitcher, PSR J0537$-$6910, is relatively young ($\tau_{\rm c} = 4.9\,{\rm kyr}$); how did it form a richly connected reservoir network so quickly? And, if its network is so richly connected, why is its aggregated $\dot{\nu}_{\rm glitch}$ value so low? Likewise, PSR J0358$+$5413 is the oldest object in the sample ($\tau_{\rm c} = 560 \,{\rm kyr}$), yet its $\epsilon$ value arguably points to a dearth of capacitive elements, characteristic of a young object. There are no obvious grounds (e.g.\ quasiperiodicity) on which to treat PSR J0358$+$5413 as exceptional. Do all pulsars glitch eventually? It has been speculated in the past that there is something special physically about the minority of pulsars that do glitch. While it is impossible to reject this hypothesis unequivocally with the data at hand, the results presented here suggest that all pulsars are capable of glitching. However, most do so infrequently (low $\lambda$) and hence have not been detected during the last four decades of timing experiments. We find that up to $\sim 30 \%$ of the pulsar population can glitch at rates lower than $\lambda_{\rm min}= 0.25$ {\rm yr}$^{-1}$ and still conform with the measured aggregate waiting-time distribution. Once verified, the claimed Poissonian nature of the glitch mechanism can be invoked to exclude broad classes of glitch theories, e.g.\ those that rely on `defects' or `turbulence' at special locations (like the pole), or that involve a pair of dependent events (A.\ Martin, private communication). It is important to interpret aftershocks carefully in this light \citep{won01}. In self-organized critical systems, the excess number of avalanches following a large avalanche (over and above the Poissonian baseline following a small avalanche) scales inversely with the time elapsed, a property known as Omori's law for earthquakes \citep{jen98}. In this paper, we do not analyze post-glitch relaxation times and glitch-activated changes in $\dot{\nu}$ in the context of avalanche processes, e.g.\ the correlation between $\Delta\dot{\nu}$ and the transient component of $\Delta\nu$ \citep{won01}. We also assume implicitly that the quantized superfluid vortices in the vortex unpinning model are organized in a rectilinear array, even though recent work suggests that meridional circulation destabilizes the array and converts it into a turbulent tangle \citep{per05,per06}. Further study of these matters is deferred to future work. We acknowledge the computer time and system support supplied by the Australian Partnership for Advanced Computation (APAC) and the Victorian Partnership for Advanced Computation (VPAC). We thank Andre Trosky for illuminating conversations on self-organized criticality and cellular automata. This research was supported by a postgraduate scholarship from the University of Melbourne. It makes use of the Australia Telescope National Facility Pulsar Catalogue \citep{man05}, which can be accessed on-line at {\tt http://www.atnf.csiro.au/research/pulsar/psrcat}. | 7 | 10 | 0710.1021 |
0710 | 0710.3815_arXiv.txt | The Newtonian solid-mechanical theory of non-compressional spheroidal and torsional nodeless elastic vibrations in the homogenous crust model of a quaking neutron star is developed and applied to the modal classification of the quasi-periodic oscillations (QPOs) of X-ray luminosity in the aftermath of giant flares in SGR 1806-20 and SGR 1900+14. A brief outline is given of Rayleigh's energy method which is particular efficient when computing the frequency of nodeless elastic spheroidal and torsional shear modes as a function of multipole degree of nodeless vibrations and two input parameters -- the natural frequency unit of shear vibrations carrying information about equation of state and fractional depth of peripheral seismogenic layer. In so doing we discover that the dipole overtones of both spheroidal and torsional nodeless vibrations possess the properties of Goldstone soft modes. It is shown that obtained spectral formulae reproduce the early suggested identification of the low-frequency QPOs from the range $30\leq \nu \leq 200$ Hz as torsional nodeless vibration modes $\nu(_0t_\ell)$ of multipole degree $\ell$ in the interval $2\leq \ell \leq 12$. Based on this identification, which is used to fix the above mentioned input parameters of derived spectral formulae, we compute the frequency spectrum of nodeless spheroidal elastic vibrations $\nu(_0s_\ell)$. Particular attention is given to the low-frequency QPOs in the data for SGR 1806-20 whose physical origin has been called into question. Our calculations suggest that these unspecified QPOs are due to nodeless dipole torsional and dipole spheroidal elastic shear vibrations: $\nu(_0t_1)=18$ Hz and $\nu(_0s_1)=26$ Hz. | The discovery of QPOs of X-ray luminosity in the aftermath of giant flares SGR 1806-20 and SGR 1900+14 (Israel et al 2005; Strohmayer \& Watts 2006; Israel 2007; Watts \& Strohmayer 2007) with concomitant interpretation of QPOs as caused by quake-induced differentially rotational seismic vibrations has stimulated remarkable developments in the magnetar asteroseismology (e.g. Piro 2005, Samuelsson \& Andersson 2007a, 2007b; Lee 2007; Levin 2007; Watts \& Reddy 2007; Sotani et al 2007; Bastrukov et al 2007a and references therein). Following the above interpretation and presuming the dominant role of the elastic restoring force, the focus of most theoretical works is on computing the frequency spectra of odd-parity torsional mode of shear vibrations and less attention is paid to the even parity spheroidal elastic mode. However, from the viewpoint of modern global seismology (Lay \& Wallace 1995; Aki \& Richards 2003), the spheroidal vibrational mode in a solid star and planet has the same physical significance as the toroidal one in the sense that these two fundamental modes owe their existence to one and the same restoring force (e.g. McDermott, Van Horn, Hansen 1988; Bastrukov, Weber, Podgainy 1999, Bastrukov et al 2007b). In this light there is a possibility that, by not considering both these modes on an equal footing, we may miss discovering certain essential novelties which are consequences of solid mechanical laws governing seismic vibrations of superdense matter of neutron stars. Adhering to this attitude and continuing the investigations recently reported by Bastrukov et al (2007a), we derive here spectral equations for the frequency of both spheroidal and torsional elastic nodeless vibrations in the solid crust of quaking neutron star and examine what conclusions can be drawn regarding low-frequency QPOs whose physical nature still remain unclear. In Sec.2 by use of the energy variational method we derive spectral formulae for the frequencies of the nodeless spheroidal and torsional elastic vibrations locked in the finite-depth seismogenic layer. Particular attention is given to the dipole spheroidal and torsional vibrations possessing properties of Goldstone's soft modes. In sec.3, the obtained spectral formulae are applied to a modal analysis of available data on the above mentioned QPOs. The obtained results are highlighted in Sec.4. | The exact spectral formulae, which has been obtained here within the framework of Newtonian, non-relativistic, solid-mechanical theory of seismic vibration for the first time, are interesting in its own right from the viewpoint of general theoretical seismology (e.g. Lay, Wallace 1995) because they can be utilized in the study of seismic vibrations of more wider class of solid celestial objects such as Earth-like planets. One of the remarkable findings of our investigation is that the dipole overtones of nodeless elastic shear vibrations trapped in the finite-depth crust of seismically active neutron star possess properties of Goldstone soft modes. It is shown that obtained spectral equations are consistent with the existence treatment of low-frequency QPOs in the X-ray luminosity of flares SGR 1900+14 and SGR 1806-20 as caused by quake-induced torsional nodeless vibrations (Samuelsson \& Andersson 2007a; 2007b). What is newly disclosed here is that previously non-identified low-frequency QPOs in data for SGR 1806-20 can be attributed to nodeless dipole torsional and spheroidal vibrations, namely, $\nu(_0t_1)=18\,\mbox{Hz}$ and $\nu(_0s_1)=26\,\mbox{Hz}$. | 7 | 10 | 0710.3815 |
0710 | 0710.5508_arXiv.txt | We analyse a sample of 33 extensive air showers (EAS) with estimated primary energies above $2\cdot 10^{19}$~eV and high-quality muon data recorded by the Yakutsk EAS array. We compare, event-by-event, the observed muon density to that expected from CORSIKA simulations for primary protons and iron, using SIBYLL and EPOS hadronic interaction models. The study suggests the presence of two distinct hadronic components, ``light'' and ``heavy''. Simulations with EPOS are in a good agreement with the expected composition in which the light component corresponds to protons and the heavy component to iron-like nuclei. With SYBILL, simulated muon densities for iron primaries are a factor of $\sim 1.5$ less than those observed for the heavy component, for the same electromagnetic signal. Assuming two-component proton-iron composition and the EPOS model, the fraction of protons with energies $E>10^{19}$~eV is $0.52 ^{+0.19}_{-0.20}$ at 95\% confidence level. | 7 | 10 | 0710.5508 |
||
0710 | 0710.4162_arXiv.txt | Estimates of inflationary parameters from the CMB $B$-mode polarization spectrum on the largest scales depend on knowledge of the reionization history, especially at low tensor-to-scalar ratio. Assuming an incorrect reionization history in the analysis of such polarization data can strongly bias the inflationary parameters. One consequence is that the single-field slow-roll consistency relation between the tensor-to-scalar ratio and tensor tilt might be excluded with high significance even if this relation holds in reality. We explain the origin of the bias and present case studies with various tensor amplitudes and noise characteristics. A more model-independent approach can account for uncertainties about reionization, and we show that parametrizing the reionization history by a set of its principal components with respect to $E$-mode polarization removes the bias in inflationary parameter measurement with little degradation in precision. | \label{sec:intro} Temperature and polarization power spectra of the cosmic microwave background (CMB) are consistent with predictions of the simplest inflationary models~\cite{Gut81,AlbSte82,Lin82,Sat81}: a nearly flat geometry, superhorizon correlations as probed by the spectrum of acoustic peaks, and primordial scalar perturbations that are adiabatic, Gaussian, and close to scale-invariant~\cite{HuWhi96c,SpeZal97,HuSpeWhi97,Peietal03,Speetal07}. One of the key remaining signatures of inflation, tensor perturbations (i.e.\ gravitational waves)~\cite{KamKosSte97,SelZal97}, has yet to be detected. Depending on the amplitude of the tensor perturbations, which is not well constrained theoretically, it may be possible to measure the angular power spectrum of the inflationary gravitational waves in the $B$-mode component of the CMB polarization on large scales. Non-detection of the tensor spectrum does not necessarily rule out inflation, but upper limits on $r$ can be used to exclude particular models of inflation and limit its energy scale. Many experiments have been proposed to search for this signal~\cite{Planck,CMBTaskForce,Oxletal04,LawGaiSei04,Ruhl:2004kv,Mafetal05,Yooetal06,Kogut:2006nd,Macetal07,Polenta07}. Measurement of tensor perturbations in the $B$-mode polarization power spectrum would test models of inflation by constraining inflationary parameters. These parameters include the tensor-to-scalar ratio, $r$, and the tensor spectral index, $n_t$, which are related by a consistency relation under the simplest single-field slow-roll inflationary scenarios. If the tensor spectrum can be detected, precise measurements over a wide range of scales could test the consistency relation. CMB constraints on $r$ and $n_t$ depend on the ability to accurately determine the large-scale power in $B$-modes due to tensor perturbations, independent of the effects of other cosmological parameters. On the largest scales, the tensor $B$-mode spectrum depends not only on inflationary parameters but also on the reionization history of the universe~\cite{Zal97}. The main impact of reionization on the spectrum is through the total optical depth, $\tau$. The 3-year \wmap\ measurements of $E$-mode polarization determine $\tau$ to an accuracy of about $30\%$~\cite{Pagetal07,Speetal07}, and future CMB experiments should constrain $\tau$ at the~$5-10\%$~level~\cite{Holetal03,KeaMil06,Planck,MorHu07b}. However, the detailed evolution of the reionization history also significantly affects the large-scale polarization spectra. Uncertainty in this history leads to added uncertainty in inflationary parameters. Moreover, incorrect inferences due to an oversimplified treatment of reionization may bias estimates of inflationary parameters. For unbiased estimation of the optical depth from the $E$-mode reionization peak, the solution is to use a complete, principal-component-based description of reionization effects when estimating parameters from CMB polarization data~\cite{HuHol03,MorHu07b}. In this paper, we extend this approach to tensor $B$-mode polarization and show that it is equally if not more effective in ensuring accurate measurements with little loss in precision. The outline of the paper is as follows. We discuss the effects of reionization and inflationary parameters on the polarization power spectra and the large-scale degeneracy between these parameters in \S~\ref{sec:param}. A brief overview of the principal component parametrization of the reionization history follows in \S~\ref{sec:pcs}. In \S~\ref{sec:mcmc}, we describe our Markov Chain Monte Carlo analysis of simulated polarization data and give the resulting constraints on $\tau$, $r$, and $n_t$, which we discuss further in \S~\ref{sec:discuss}. \begin{figure} \centerline{\psfig{file=f1.eps, width=3.0in}} \caption{$B$-mode tensor spectra illustrating the degeneracy between $\tau$ and $n_t$ for large-scale measurements, with angular power spectra plotted in the upper panel and fractional deviations from the base model in the lower panel. For the base model (\emph{solid}), $r=0.03$, $\tau=0.1$, and $n_t=-0.00375$. The other two models have $\{\tau,n_t\}=\{0.12,-0.00375\}$ (\emph{long dashed}) and $\{0.12,0.13\}$ (\emph{short dashed}), with a pivot scale of $k_{\rm pivot}=0.01~{\rm Mpc}^{-1}$. The reionization history here is assumed to be instantaneous. Cosmic variance of $\cltens$ for the base model, which excludes the variance from lensing, is shown by the shaded band in the lower panel. } \label{fig:degen} \end{figure} | \label{sec:discuss} The value of the optical depth to reionization estimated from the CMB $E$-mode polarization spectrum on large scales can be biased by adopting a model that has insufficient freedom to describe the true reionization history. Likewise, the use of simple reionization models can bias inflationary parameters such as the tensor-to-scalar ratio and tensor tilt that depend on the large-scale amplitude of the $B$-mode spectrum of primordial gravitational waves. In each case, the problem can be solved by using a more general parametrization of the reionization history. We have shown that using a small but complete set of the principal components of the reionization history effectively yields unbiased constraints on both reionization and inflationary parameters. Measurements of $r$ and $n_t$ are only affected by the assumed form of the reionization history if the reionization peak of the tensor $B$-mode spectrum at the very largest scales is needed to precisely constrain the parameters. If, instead, good constraints can be obtained using only the $B$-mode recombination peak at intermediate scales, then assumptions about reionization do not affect tests of the consistency relation between $r$ and $n_t$. They would instead appear as false evidence for running of the tensor tilt in violation of slow-roll expectations. Measurement of the recombination peak however is inhibited by experimental noise and contamination from $E$-mode power converted to $B$-mode power by gravitational lensing, both of which become more important at smaller scales. To study the potential impact of reionization on parameter constraints from $B$-mode polarization, we have employed a Markov Chain Monte Carlo analysis of simulated CMB polarization power spectra and compared results for two descriptions of reionization: a simple, one-parameter, instantaneous reionization model, and a parametrization using principal components of the reionization history with respect to the $E$-mode polarization power spectrum. By varying the properties of the simulated polarization power spectra, including the fiducial tensor-to-scalar ratio and the noise spectrum, we have determined over what range of scales CMB polarization data is most important for constraining inflationary parameters in various scenarios. In particular, the question of whether the large-scale reionization peak of $\cltens$ or the smaller-scale recombination peak is more important determines the severity of bias in inflationary parameters when reionization is modeled incorrectly. If the tensor-to-scalar ratio is near the current upper limit of $r\sim 0.3$ and measurements of $B$-mode polarization are limited only by cosmic variance, then the spectrum on scales $20 \lesssim \ell \lesssim 500$ dominates constraints on $r$ and $n_t$ and incorrect assumptions about reionization do not strongly bias the results. If the true tensor-to-scalar ratio is more than a factor of a few smaller than this upper bound, however, then lensing $B$-modes limit the information that can be extracted from the recombination peak of the tensor spectrum alone. Furthermore, all-sky experiments in the foreseeable future are likely to have noise that exceeds the lensing signal, making tests of inflation even more reliant on the reionization peak of the tensor $B$-modes on large scales. In all of these cases, a general parametrization of reionization such as that provided by principal components allows the use of the $B$-mode reionization peak for inflationary parameter constraints without significantly worsening the errors on those parameters. \vfill | 7 | 10 | 0710.4162 |
0710 | 0710.4448_arXiv.txt | The first high resolution \textit{Spitzer} IRS 9-37$\mu$m spectra of 29 Seyfert galaxies (about one quarter) of the 12$\mu$m Active Galaxy Sample are presented and discussed. The high resolution spectroscopy was obtained with corresponding off-source observations. This allows excellent background subtraction, so that the continuum levels and strengths of weak emission lines are accurately measured. The result is several new combinations of emission line ratios, line/continuum and continuum/continuum ratios that turn out to be effective diagnostics of the strength of the AGN component in the IR emission of these galaxies. The line ratios [NeV]/[NeII], [OIV]/[NeII], already known, but also [NeIII]/[NeII] and [NeV]/[SiII] can all be effectively used to measure the dominance of the AGN. We extend the analysis, already done using the 6.2$\mu$m PAH emission feature, to the equivalent width of the 11.25$\mu$m PAH feature, which also anti-correlates with the dominance of the AGN. We measure that the 11.25$\mu$m PAH feature has a constant ratio with the H$_2$ S(1) irrespective of Seyfert type, approximately 10 to 1. Using the ratio of accurate flux measurements at about 19$\mu$m with the two spectrometer channels, having aperture areas differing by a factor 4, we measured the source extendness and correlated it with the emission line and PAH feature equivalent widths. The extendness of the source gives another measure of the AGN dominance and correlates both with the EWs of [NeII] and PAH emission. Using the rotational transitions of H$_2$ we were able to estimate temperatures (200-300K) and masses (1-10 $\times$ 10$^{6}$ M$_{\sun}$), or significant limits on them, for the warm molecular component in the galaxies observed. Finally we used the ratios of the doublets of [NeV] and of [SIII] to estimate the gas electron density, which appears to be of the order of n$_e$ $\sim$ 10$^{3-4}$ cm$^{-3}$ for the highly ionized component and a factor 10 lower for the intermediate ionization gas. | Mid-infrared (mid-IR) spectroscopy provides a powerful tool to investigate the nature and physical processes in active galactic nuclei (AGNs) and in the starburst dominated regions frequently associated to them. Because of the large variety of fine structure lines present in the mid-IR, covering a wide range in ionization/excitation conditions and gas density \citep[e.g.][]{sm92} mid-IR spectroscopy of the Narrow Line Regions (NLR) in AGNs can add information not available from classical optical spectroscopy, especially when dust extinction is high. Furthermore, the electronic transitions responsible for the infrared emission lines of various elements are less sensitive to uncertainties in temperature than the corresponding optical lines. Moreover the brightest H$_2$ rotational lines, that can be used to quantify the presence and excitation of warm molecular gas, as well as a prominent Polycyclic Aromatic Hydrocarbons \citep{pule89}, hereafter PAH, feature at 11.25 $\mu$m are in the mid-IR. The first detailed mid-IR spectroscopic studies of AGNs and Starburst galaxies using multiple ionic transitions of various elements were performed by various authors \citep[][for a review]{stu02, spi05, ver03, ver05} with the \textit{Short Wavelength Spectrometer} (SWS) \citep{deg96} onboard of the \textit{Infrared Space Observatory} (ISO) \citep{kes96}. However, the improved sensitivity of the Infrared Spectrometer (IRS) \citep{hou04} on the \textit{Spitzer Space Telescope} now enables a detailed investigation of the nature and physical processes in large samples of galactic nuclei from nonstellar (AGN) and stellar (starburst) power sources. Even at low resolution, the IRS spectra of the first few classical AGNs already showed the diversity of the mid-IR spectral features: silicate absorption and emission, PAH emission and strong fine structure lines \citep{wed05}. \citet{buc06} examined 51 low resolution IRS spectra of 12$\mu$m selected Seyfert galaxies \citep{rms93}, exactly the sample we are considering in our study. They report a few major findings: (1) the sample contains a very wide range of continuum types, with no more than about 3 galaxies being closely similar to one another, however principal component analysis applied to their data suggests that the relative contribution of starburst emission may be the dominant cause of variance in the observed spectra; (2) the starburst component in the sample objects does not contribute more than 40\% of the total IR flux density; (3) Seyfert 1's have higher ratios of infrared to radio emission \citep[see also][]{rme96}; (4) the Seyfert 2 galaxies typically show stronger starburst contributions than Seyfert 1's. \citet{Gor07} found a strong correlation between the [NeV]14.3$\mu$m and the [NeIII]15.5$\mu$m lines in Narrow Line Regions (NLR) of AGNs, spanning 4 orders of magnitude in luminosity. This would imply a very narrow range in ionization parameter (-2.8$<$logU$<$-2.5) for simple constant density photoionization models. \citet{dud07} discussed the ratio of the doublets of [NeV] and [SIII] in a heterogeneous sample of active galaxies. Finally ULIRG galaxies, more than Seyfert galaxies, have been so far the object of systematic spectroscopic studies with Spitzer IRS \citep{hig06,arm07,des07,far07}. In this article we present the first high resolution IRS spectra of 29 galaxies from the original list of Seyfert galaxies of the "12$\mu$m Galaxy Sample" (12MGS) \citep{rms93}, an IRAS-selected all-sky survey flux-limited to 0.22 Jy at 12$\mu$m. The sample selection is briefly described in \S 2, the observations are described in \S 3, the results are presented in \S 4, the diagnostics diagrams using line ratios, equivalent widths and other observed quantities are presented and discussed in \S 5 and a comparison between Seyfert 1's and Seyfert 2's is given in \S 6. The conclusions are given in \S 7. | We have identified a family of IRS observables which quantify the proportion of the total IR emission coming from a Seyfert nucleus, all of which are intercorrelated with each other. The ratios of ionic fine structure lines [NeV]/[NeII] and [OIV]/[NeII] were already proposed to measure the importance of the AGN component. We also see that [NeIII]/[NeII] and OIV or [NeV]/[SiII]35$\mu$m can be used to quantify the AGN dominance. It was also known from ISO that the equivalent width of the 6.2$\mu$m (and 7.7$\mu$m) PAH features is inversely related to the AGN dominance; we find that the same holds for the equivalent width of the 11.25$\mu$m PAH feature. We also discovered two additional IRS observables: the equivalent width of [NeII]12.8$\mu$m and the extendness of the 19$\mu$m continuum, which also quantify the dominance of the AGN component compared with the emission from the underlying spiral galaxy. All of these observables are correlated with each other, since they are measuring this same astrophysical quantity, and they are all correlated with the hardness of the far-IR continuum, since a more dominant AGN component is already known to be correlated with hotter dust. There is no clear indication that recent star formation is much more important on average in the Seyfert 2's in our sample, compared with the Seyfert 1's. Although the Seyfert 1's generally tend to have more dominant AGN than the Seyfert 2's, there is a strong overlap between these two classes. The relatively small difference between the averages of these observables for Seyfert 1's and Seyfert 2's indicates that those two AGN categories are not extremely different in the mid-IR range. Thus for example, the AGN in Seyfert 2's are clearly observable in their 10$\mu$m continuum emission, and in [NeIII], [OIV] and [NeV], to almost the same degree as in Seyfert 1's, which is not consistent with some proposed torus models for Seyfert 1/2 unification. Finally, we do not fully confirm the observational claims of \citet{dud07} : 75 - 80\% of our Seyfert 1's and Seyfert 2's show [SIII] or [NeV] densities larger than the low-density limit. In fact, after applying plausible aperture corrections to the [SIII] line ratio, only three Seyfert 2's and one Seyfert 1 have a [SIII] line ratio below the density limit. A similar result is also found for the [NeV] ratio, for which we have two Seyfert 1's (that became four including the upper limits) and 2 Seyfert 2's below the low density limit. These few cases do not in our view require enormous dust extinction values. | 7 | 10 | 0710.4448 |
0710 | 0710.2617.txt | We consider the propagation of cosmic rays in turbulent magnetic fields. We use the models of magnetohydrodynamic turbulence that were tested in numerical simulations, in which the turbulence is injected on large scale and cascades to small scales. Our attention is focused on the models of the strong turbulence, but we also briefly discuss the effects that the weak turbulence and the slab Alfv\'enic perturbations can have. The latter are likely to emerge as a result of instabilities with in the cosmic ray fluid itself, e.g., beaming and gyroresonance instabilities of cosmic rays. To describe the interaction of cosmic rays with magnetic perturbations we develop a non-linear formalism that extends the ordinary Quasi-Linear Theory (QLT) that is routinely used for the purpose. This allows us to avoid the usual problem of 90 degree scattering and enable our computation of the mean free path of cosmic rays. We apply the formalism to the cosmic ray propagation in the galactic halo and in the Warm Ionized medium (WIM). In addition, we address the issue of the transport of cosmic rays perpendicular to the mean magnetic field and show that the issue of cosmic ray subdiffusion (i.e., propagation with retracing the trajectories backwards, which slows down the diffusion) is only important for restricted cases when the ambient turbulence is far from what numerical simulations suggest to us. As a result, this work provides formalism that can be applied for calculating cosmic ray propagation in a wide variety of circumstances. | The propagation and acceleration of cosmic rays (CRs) is governed by their interactions with magnetic fields. Astrophysical magnetic fields are turbulent and, therefore, the resonant and non-resonant (e.g., transient time damping, or TTD) interaction of cosmic rays with MHD turbulence is the accepted principal mechanism to scatter and isotropize cosmic rays (see Schlickeiser 2002). In addition, efficient scattering is essential for the acceleration of cosmic rays. For instance, scattering of cosmic rays back into the shock is a vital component of the first order Fermi acceleration (see Longair 1997). At the same time, stochastic acceleration by turbulence is entirely based on scattering. It is generally accepted that properties of turbulence are vital for the correct description of CR propagation. Historically, the most widely used model is the model composed of slab perturbations and 2D MHD perturbations (see Bieber, Smith, \& Matthaeus 1988). The advantage of this empirical model is its simplicity and the ability to account for the propagation of CRs in magnetosphere given a proper partition of the energy between the two types of modes. Numerical simulations (see Cho \& Vishniac 2000, Maron \& Goldreich 2001, M\"uller \& Biskamp 2000, Cho, Lazarian \& Vishniac 2002, Cho \& Lazarian 2002, 2003, see also book of Biskamp 2003, as well as, Cho, Lazarian \& Vishniac (2003) and Elmegreen \& Scalo 2004 for reviews), however, do not show 2D modes, but instead show Alfv\'enic modes that exhibit scale-dependent anisotropy consistent with predictions in Goldreich \& Sridhar (1995, henceforth GS95). The approach in the latter work makes productive use of the earlier advances in understanding of MHD turbulence, that can be traced back to Iroshnikov (1963) and Kraichnan (1965) work and the classical work that followed (see Montgometry \& Turner 1981, Higdon 1984, Montgomery, Brown \& Matthaeus 1987). A careful analysis shows that there is no big gap between the Reduced MHD and the GS95 model. In fact, it was shown in Lazarian \& Vishniac (1999) that the numerical results in Matthaeus et al. (1998) are consistent with GS95 predictions. While particular aspects of the GS95 model, e.g., the particular value of the spectral index, are the subject of controversies\footnote{To address quantitatively these controversies, we need much better numerical resolution. For instance, hydrodynamic turbulence simulations in Kritsuk et al. (2007) showed that only starting with the 1028$^3$ numerical cubes the bottleneck effects stop dominating the measured spectral slope. While the simulations in Kowal \& Lazarian (2007) show that the bottleneck is less important for their MHD code, the exact value of the spectral slope is still uncertain. At the same time, the particular theoretically-predicted features of turbulence, for instance, the existence of the scale-dependent anisotropy, can be reliably established, while the exact scaling of this dependence, e.g., like $k_{\|}\sim k_{\bot}^{2/3}$ as in GS95 model or $k_{\|}\sim k_{\bot}^{\alpha*}$, where $\alpha*<2/3$ (see Beresnyak \& Lazarian 2006), also require higher resolution simulations.} (see M\"uller \& Biskamp 2000, Boldyrev 2005, 2006, Beresnyak \& Lazarian 2006, Gogoberidze 2006, Mason et al. 2007), we think that, at present, GS95 model provides a good starting for developing models of CR scattering. as was done in Chandran (2000), Yan \& Lazarian (2002, 2004, henceforth YL02, Paper I, respectively), Brunetti \& Lazarian (2007) etc. In particular, the latter three papers used the decomposition of MHD turbulence over Alfv\'en, slow and fast modes as in Cho \& Lazarian (2003) and identify the fast modes as the major source of CR scattering in interstellar and intracluster medium. However, while the turbulence injected on large scales may correspond to GS95 model and its extensions to compressible medium (Lithwick \& Goldreich 2001, Cho \& Lazarian 2002, 2003), one should not disregard the possibilities of generation of additional perturbations by CR themselves. Indeed, the slab Alfv\'enic perturbation can be created, e.g., via streaming instability (see Wentzel 1974, Cesarsky 1980) or kinetic gyroresonance instability (see its application for CR transport in Lazarian \& Beresnyak 2006). These perturbations, that are present for a range of CRs energies (e.g., $\lesssim 100$GeV for the instabilities above in ISM) owing to non-linear damping arising from ambient turbulence (YL02, Paper I, Farmer \& Goldreich 2004, Lazarian \& Beresnyak 2006), should also be incorporated into the comprehensive models of CR propagation and acceleration. At present, the propagation of the CRs is an advanced theory, which makes use both of analytical studies and numerical simulations. However, these advances have been done within the turbulence paradigm which is being changed by the current research in the field. As we discussed above, instead of the empirical 2D+slab model of turbulence, numerical simulations suggest anisotropic Alfv\'enic modes (an analog of 2D, but not an exact one, as the anisotropy changes with the scale involved) + fast modes or/and slab modes. This calls for important revisions of the CR propagation, which is the subject of the current paper. The perturbations of turbulent magnetic field are usually accounted for by direct numerical scattering simulations (Giacalone \& Jokipii 1999, Mace et al 2000, Qin at al. 2002) or by quasi-linear theory, QLT (see Jokipii 1966, Schlickeiser 2002). The problem with direct numerical simulations of scattering is that the present-day MHD simulations have rather limited inertial range. At the same time, creating synthetic turbulence data which would correspond to scale-dependent anisotropy in respect to the local magnetic field (which corresponds, e.g., to GS95 model) is challenging and has not been practically realized, as far as we know. While QLT allows easily to treat the CR dynamics in a local magnetic field system of reference, a key assumption in QLT, that the particle's orbit is unperturbed, makes one wonder about the limitations of the approximation. Indeed, while QLT provides simple physical insights into scattering, it is known to have problems. For instance, it fails in treating $90^o$ scattering (see Jones, Birmingham \& Kaiser 1973, 1978; V\"olk 1973, 1975; Owens 1974; Goldstein 1976; Felice \& Kulsrud 2001) and perpendicular transport (see K\'ota \& Jokipii 2000, Matthaeus et al. 2003). Indeed, many attempts have been made to improve the QLT and various non-linear theories have been attempted (see Dupree 1966, V\"olk 1973, 1975, Jones, Kaiser \& Birmingham 1973, Goldstein 1976). Currently we observe a surge of interest in finding way to go beyond QLT. Those include recently developed nonlinear guiding center theory (see Matthaeus et al. 2003), weakly nonlinear theory (Shalchi et al. 2004), second-order quasilinear theory (Shalchi 2005a) (see also Shalchi 2006, Webb et al. 2006, Qin 2007, Le Roux \& Webb 2007). At the same time, most of the analysis so far has been confined to traditional 2D+slab models of MHD turbulence. Following the reasoning above, we think that it is important to extend the work to the non-linear treatment of CR scattering to models MHD turbulence that are supported by numerical simulations. Propagation of CRs perpendicular to the mean magnetic field is another important problem in which QLT encounters serious difficulties. Compound diffusion, resulting from the convolution of diffusion along the magnetic field line and diffusion of field line perpendicular to mean field direction, has been invoked to discuss transport of cosmic rays in the Milky Way (Getmantsev 1963; Lingenfelter, Ramary \& Fisk 1971; Allan 1972). The role of compound diffusion in the acceleration of CRs at quasi-perpendicular shocks were investigated by Duffy et al. (1995) and Kirk et al. (1996). Indeed, the idea of CR transport in the direction perpendicular to the mean magnetic field being dominated by the field line random walk (FLRW, Jokipii 1966, Jokipii \& Parker 1969, Forman et al. 1974) can be easily justified only in a restricted situation where the turbulence perturbations are small and CRs do not scatter backwards to retrace their trajectories. If the latter is not true, the particle motions are subdiffusive, i.e., the squared distance diffused growing as not as $t$ but as $t^{\alpha}$, $\alpha<1$, e.g., $\alpha=1/2$ (K\'ota \& Jokipii 2000, Mace et al 2000, Qin at al. 2002, Shalchi 2005b). If true, this could indicate a substantial shift in the paradigm of CR transport, a shift that surely dwarfs a modification of magnetic turbulence model from the 2D+slab to a more simulation-motivated model that we deal here. It was also proposed that with substantial transverse structure, {\it i.e.}, transverse displacement of field lines, perpendicular diffusion is recovered (Qin et al 2002). Is it the case of the MHD turbulence models we deal with? How realistic is the subdiffusion in the presence of turbulence? The answer for this question apparently depends on the models of turbulence chosen. In this paper we again seek the answer for this question within domain of numerically tested models of MHD turbulence. There are three major thrusts of the paper:\\ I. Extend QLT by taking into account magnetic mirroring effect on large scales.\\ II. Describe CR propagation in Milky Way (e.g., calculate CR mean free path for different phases of ISM).\\ III. Address the problem of perpendicular transport of CR. In what follows, we discuss the cosmic ray transport in incompressible turbulence in \S2. We shall describe the \S2.1 dispersion of guiding center of CRs and introduce the broadened resonance function to replace the $\delta$ function in QLT, following which we shall discuss the scattering in strong and weak incompressible turbulence respectively in \S2.2 and \S2.3. Then we shall consider the scattering by fast modes in \S3 and apply the analysis to ISM and get mean free path for different phases of ISM (\S4). In \S5, we shall study the perpendicular transport of cosmic rays on both large and small scales. We shall also discuss the applicability of the subdiffusion. Discussion and summary are provided in \S6 and \S7 respectively. | The present paper extends our study in Paper I. As in Paper I we mostly deal with the magnetic perturbations that are part of the large scale turbulent cascade, which is consistent with the Big Power Law in the sky observed via radio-scattering and scintillation technique (Armstrong et al. 1995). In both papers we use the description of the MHD turbulence that follows from numerical simulations. In Paper I we have the CR scattering calculated in the selected interstellar environments making use of Quasi-Linear Theory (QLT). Because of the limitations of the QLT, we could not provide calculations of the mean free path in Paper I, which limited the utility of the study. In this paper we extended the non-linear approach suggested in V\"olk (1975) to treat the scattering, which allows us to calculate the mean free paths that arise from CR interactions with the fast modes. In doing so, similar to Paper I, we take into account damping of the fast modes in the presence of the field wondering induced by the Alfv\'enic modes. Our results show that in WIM and halo of our Galaxy, confinement of bulk CRs are mostly due to the compressible modes. We obtain CR mean free paths about a few parsec, consistent with what observations indicate. The major difference with earlier picture is the dependencies of CR transport parameters on the medium properties. The dependence appears as a result of damping of the fast modes. For low energy CRs ($\lesssim 100$GeV), if dominated by viscous damping, the mean free path of CRs would decrease with energy; with collisionless damping, however, CRs' mean free path stays almost a constant. Field line wandering in general increases the damping of the fast modes and reduces the scattering efficiency of CRs. For higher energy CRs, the influence of damping is limited, and their mean free path increases with energy. The dependencies on the turbulence damping and therefore the phase properties should have various implications from ratio of secondary to primary elements, diffuse Galactic $\gamma$ ray emission, to the CMB synchrotron foreground. With precise measurements, the understanding of CMB is now constrained by our understanding of the foreground. The variation of CR index over the Galaxy may paralyze the synchrotron templates. Such variations can be addressed on the basis of the more elaborate CR propagation theory. The importance of this study goes beyond the interstellar medium. For instance, Brunetti \& Lazarian (2007), treated acceleration of CRs for plasma in clusters of galaxies appealing to the fast modes, which is the approach to CRs similar to that in Paper I. We believe that the non-linear treatment may be useful for such cases as well. In addition, stochastic acceleration by the MHD turbulence is a promising mechanism for generating high energy particles during solar flares (see, e.g., Petrosian \& Liu 2004, and references therein). An application to the acceleration of CRs in solar flares will be given in Yan, Lazarian \& Petrosian (2007, in preparation). In our treatment we attempted to use the scalings that (a) are consistent with numerical calculations and (b) whose amplitudes we can estimate with a sufficient degree of precision. Therefore our present study does not deal with scattering of CRs by the fast modes on the scales $l>LM_A^2$, $M_A<1$, i.e., on the scales where the Alfv\'enic turbulence in the weak regime. It was suggested by Chandran (2005) that the weak fast modes at small pitch angles tend to steepen due to the coupling with the Alfv\'en modes. When the resulting scaling of the fast modes becomes clearer, our approach will be applicable to them. We have not quantitatively dealt in the present paper with the case of the slab Alfv\'en modes created by instabilities\footnote{Streaming instability (see Cesarsky 1980) is an example of such instability. However, the instability is suppressed by both ion-neutral damping (Kulsrud \& Pierce 1969) and the ambient turbulence (YL02, Farmer \& Goldreich 2004, Paper I, Lazarian \& Beresnyak 2006). Another example is the gyroresonance instability discussed in the context of CRs in Lazarian \& Beresnyak (2006).}. The CR scattering by the perturbations created by those modes may dominate over the gyroresonance with the fast modes, especially for CRs of low energies, i.e., whose gyroresonance with the fast modes is inefficient due to the fast modes damping (see estimates in Lazarian \& Beresnyak 2006). Progress in quantitative description of the non-linear stages of the instabilities that can create slab modes should enable comprehensive models that include both the fast modes and the slab modes. In addition, we addressed the issue of perpendicular diffusion, the issue that we have not dealt with in Paper I. We found, that similar to the case of thermal diffusion discussed in Lazarian (2006), the diffusion of CRs depends on the Alfv\'enic Mach number $M_A$. We found that the suppression of the perpendicular diffusion compared to the parallel one scales as $M_A^4$ for $M_A<1$. Approaching the issue of subdiffusion, we found that it is negligible for CRs in the Alfv\'enic turbulence. | 7 | 10 | 0710.2617 |
0710 | 0710.4354_arXiv.txt | We present deep WIYN H$\alpha$ imaging of the dwarf irregular starburst galaxy NGC 1569, together with WIYN SparsePak spatially-resolved optical spectroscopy of the galactic outflow. This leads on from our previous detailed analyses of the state of the ISM in the central regions of this galaxy. Our deep imaging reveals previously undetected ionized filaments in the outer halo. Through combining these results with our spectroscopy we have been able to re-define the spatial extent of the previously catalogued superbubbles, and derive estimates for their expansion velocities, which we find to be in the range 50--100~\kms. The implied dynamical ages of $\lesssim$\,25~Myr are consistent with the recent star- and cluster-formation histories of the galaxy. Detailed decomposition of the multi-component H$\alpha$ line has shown that within a distinct region $\sim$$700\times 500$~pc in size, roughly centred on the bright super star cluster A, the profile is composed of a bright, narrow (FWHM $\lesssim$ 70~\kms) feature with an underlying, broad component (FWHM $\sim$ 150~\kms). Applying the conclusions found in our previous work regarding the mechanism through which the broad component is produced, we associate the faint, broad emission with the interaction of the hot, fast-flowing winds from the young star clusters with cool clumps of ISM material. This interaction generates turbulent mixing layers on the surface of the clouds and the evaporation and/or ablation of material into the outflow. Under this interpretation, the extent of the broad component region may indicate an important transition point in the outflow, where ordered expansion begins to dominate over turbulent motion. In this context, we present a multi-wavelength discussion of the evolutionary state of the outflow. | Outflows powered by the collective injection of kinetic energy and momentum from massive stars and supernovae (SNe) can drastically affect the structure and subsequent evolution of galaxies. Thus, a good understanding of the feedback mechanisms between massive stars, star clusters and the ISM is fundamental. In particular, dwarf galaxies are thought to be strongly affected by the effects of feedback since their smaller gravitational potentials mean that supernova-heated gas can escape more easily \citep{larson74}. Although the ejection of the ISM potentially could have significant consequences on the star-formation rate within these low-mass systems \citep{dekel86}, more recent work suggests that ejection of hot gas through bubble blow-out may not be as efficient as first thought \citep{deyoung94, martin98}. It is therefore important to study such systems to understand how gas is removed and what effects this has in the evolution of the galaxy. NGC 1569 (UGC 3056, Arp 210, VII Zw 16, IRAS 4260+6444) is a nearby \citep[2.2~Mpc;][]{israel88}, low metalicity \citep*[0.25~\Zsun;][]{devost97, kobulnicky97} dwarf irregular galaxy that has recently undergone a period of starburst activity. The most recent burst is thought to have peaked between $\sim$10--100~Myr ago with an average star-formation rate of $\sim$0.5~\Msun~yr$^{-1}$ \citep{greggio98}. At some point near the end of this event, the two well-known super-star clusters (SSCs) A and B were formed (\citealt{arp85}; \citealt*{oconnell94}; \citealt{demarchi97}), and together with the slightly older cluster 30 \citep{hunter00, origlia01}, appear to dominate the energetics of the central regions. H\one\ observations of NGC 1569 \citep{stil98, stil02, muhle05} clearly show morphological and kinematical signatures caused by the starburst. Firstly, the H\one\ kinematics are ``strongly disrupted'' within the central 900~pc, with little or no evidence for the disc rotation seen at lager radii. Furthermore, large seemingly tidal structures are seen, including a so-called bridge connecting the galaxy to a low-mass H\one\ cloud \citep[the companion;][]{stil98}, a large H\one\ arm extending to the west of the disc, and a very faint filamentary H\one\ stream wrapping around the south of the disc \citep{muhle05}. A `hot spot' (a region of velocity crowding) on the western edge of the disk was found by \citet{muhle05}, who interpreted it as the impact location of infalling gas from the companion. This provides a compelling explanation of how the starburst event was triggered. H$\alpha$ images of this galaxy show an equally chaotic, complex structure to the warm ionized component, exhibiting many filaments, bubbles and loops. Many of these were identified by \citet*{hunter93} from deep H$\alpha$ imaging. Later kinematical studies found that these filaments form part of a cellular outflow structure, comprising large-scale expanding superbubbles on the northern and southern sides of the disc (\citealt*{tomita94}; \citealt{heckman95, martin98}). By analysing spectra from two long-slits placed parallel to the major and minor axes, \citet{heckman95} found evidence of shocks in the outer regions of the ionized halo. The ratios of [O\one]/H$\alpha$, [S\two]/H$\alpha$ and [N\two]/H$\alpha$ \citep[used often to diagnose and trace the conditions within ionized gas;][]{veilleux87, dopita00} were all found to be high in the outer filaments, indicating either an increase in the importance of shocks, or that photoionization becomes less important with distance as the ionizing radiation from the central starburst becomes diluted. The existence of shocked gas is supported by high-resolution X-ray observations of NGC 1569. \citet*{martin02} examined the X-ray properties of the outflow with \textit{Chandra}, and found significant soft (0.3--0.7~keV), diffuse emission coincident with the H$\alpha$ morphology. Although they found the X-ray colour variations to be inconsistent with a free-streaming wind, they concluded that the X-ray emission probably originates in the halo shock generated by the outflowing gas, possibly from the mixing layers between the shock and the bubble interior. In order to properly characterise the structure of the outflow, it is essential to study its kinematics. From optical long-slit spectroscopy, \citet{martin98} found the outflow to be composed of numerous superbubbles. In general, she finds the redshifted component of the split-line profile to be stronger in the south and weaker in the north, suggesting an inclined outflow aspect. This is consistent with more recent X-ray absorption measurements \citep{martin02} and H\one\ observations \citep[from which an inclination angle of $\sim$60$^{\circ}$ was derived][]{stil02}. Although it is unclear whether the two SSCs, A and B, alone are sufficiently powerful to provide enough mechanical energy and ionizing radiation to drive the whole outflow and power the galaxy's diffuse ionized medium \citep{martin97, martin02}, what is clear from the H$\alpha$ morphology is that energy is being injected throughout the central starburst zone of the disc from multiple sources. A detailed look at the spectral line profiles, however, shows that near SSC A, ``distinctly non-Gaussian'' H$\alpha$ emission can be found, exhibiting weak but very broad wings \citep{heckman95}. Broad emission line wings have been detected in other nearby starburst galaxies (e.g.~\citealt{izotov96, homeier99, marlowe95, mendez97}; \citealt*{sidoli06}; \citealt{westm07c}). Due to mismatches in spectral and spatial resolution and in the specific environments observed, the nature of the energy source for these broad lines has been contested, and has resulted in the proposal of number of possible explanations. However, a detailed IFU (integral field unit) study of the ionized ISM conditions in four regions within the 200~pc region surrounding SSC A by \citet[][hereafter \citetalias{westm07a}]{westm07a} and \citet[][hereafter \citetalias{westm07b}]{westm07b} have shed a considerable amount of light on this problem. By accurately decomposing the emission line profiles across each of the IFU fields, we found the line shape to be, in general, composed of a bright narrow feature (intrinsic FWHM $\sim$ 50~\kms) superimposed on a fainter broad component (FWHM $\sim$ 200--400~\kms). By mapping out their individual properties, we identified a number of correlations between the line components that allowed us to investigate in detail the state of the ionized ISM. We concluded that the broad underlying component is most likely produced in a turbulent mixing layer \citep{slavin93, binette99} on the surface of cool gas knots, set up by the impact of the fast-flowing cluster winds \citep{pittard05}. Our analysis revealed a very complex environment with many overlapping and superimposed components, but surprisingly no evidence for organised bulk motions. We concluded that the four regions sampled are all located well within the wind energy injection zone \citep{shopbell98} at the very roots of the outflow, and that the collimation processes required to transform the turbulent motions into an organised net outflow forming the large-scale superbubbles must occur between 100--200~pc from the central star clusters. With this in mind, we have obtained new deep H$\alpha$ imaging of NGC 1569, together with spatially-resolved spectra of the outer halo, to investigate in detail the morphology and kinematics (including the line profile shapes) of the warm ionized component of the outflow at these large radii. In Section~\ref{sect:data} we present the observations and in Section~\ref{sect:maps} we map out the properties of the line components and discuss the results of the spectroscopy. Since a number of SparsePak fibres are coincident with or lie adjacent to the Gemini GMOS/IFU fields presented in Papers I and II, we compare the results obtained with the two instruments in Section~\ref{sect:comparison}. In Section~\ref{sect:disc} we discuss the state of the outflow, including the conditions in the inner and outer halo, and we summarise our findings and conclusions in Section~\ref{sect:concs}. | \label{sect:concs} We have presented WIYN MiniMo deep H$\alpha$ imaging of NGC 1569 covering a field-of-view of $9.5\times 10.7$~arcmins. The depth and large dynamic range of the observations have enabled the identification of previously undetected faint ionized filaments in the halo and the study of the gas morphology right down into the bright, central regions of the disc. We have also presented WIYN SparsePak ``formatted field unit'' observations covering the outer galactic wind flow of NGC 1569 in four pointings with integration times of 4--4.5 hours per field. The large diameter of the SparsePak fibres makes this instrument ideal for probing the faint ionized gas found in the halos of galaxies. This light-collecting power allowed us to choose a high-resolution spectrograph set-up, enabling us to characterise the line profile shapes of the important nebular diagnostic lines of H$\alpha$ and [S\two] to an accuracy limited only by the S/N achieved. We now summarise our main findings. \begin{itemize} \item We find H$\alpha$ emission out to radii of $\sim$1.5~kpc from the disc, and detect emission in almost every SparsePak fibre over the combined FoV. The presence of such an extensive system of ionized filaments results from the ongoing starburst that is supplying both mechanical energy to eject material and the Lyman continuum luminosity to keep it photoionized. \item Through detailed Gaussian line fitting, we find that within a distinct region $\sim$$700\times 500$~pc in size, roughly centred on the location of SSC A, the nebular emission line profiles are composed of a bright, narrow (FWHM $\lesssim$ 70~\kms) component with an underlying, broad component (FWHM $\sim$ 150~\kms). At larger radii, we find two narrow components to the H$\alpha$ line, each representing one half of a split-line profile. \item By comparing our results to observations of the central regions directly surrounding SSC A \citepalias{westm07a, westm07b}, we conclude that the physical mechanisms that give rise to the underlying broad emission seen within this zone must be the same as within the regions directly surrounding SSC A sampled by Gemini GMOS/IFU observations. The broad emission is most likely to result from turbulent mixing layers on the surface of cool gas clumps set up by the impact of the hot, fast-flowing cluster winds, and from evaporation and/or ablation of material from the clumps. \item The extent of this broad component region, coincident with the point at which we start observing signatures of large-scale bubble expansion, may indicate a transition point where ordered expansion begins to dominate over turbulent motion. Further observations are needed to investigate this in more detail. \item By combining our deep H$\alpha$ imaging and spectroscopy, we redefine the spatial extents of superbubbles A and B, and confirm their published \citep{martin98} expansion velocities (90 and 85~\kms, respectively). We estimate the dynamical ages of these bubbles to be $\sim$10--15~Myr. Contrary to what has been previously suggested \citep{martin98}, we find no kinematic or morphological evidence to suggest that either of these two superbubbles have ruptured and are venting their interiors into the galactic halo. \item Our data indicate that the halo of NGC 1569 contains only 4 superbubbles. Following the terminology introduced by \citet{martin98}, we propose that her superbubble F should should encompass the whole south-eastern bubble complex, where the velocity ellipses identified by \citet{martin98} and used to define shells G and F, are simply one level of a `hierarchy of structure'. Furthermore, we propose that superbubble E should encompass what were previously referred to as shells D and E and the large north-eastern X-ray spur \citep{martin02}. \item We derive new measurements of the expansion velocity, $v_{\rm exp}$ (calculated from the difference in the radial velocities between the two split-line components), for the superbubble complexes E and F of $v_{\rm exp}\sim50$~\kms{} and $\lesssim$100~\kms, respectively. Assuming a diameter of $\sim$1.3~kpc for these two structures implies dynamical ages of $\lesssim$25~Myr. \item The derived ages of the supershells are consistent with the recent cluster formation history of NGC 1569 \citep{anders04}, implying that each shell is associated with a specific star-forming event, such as a young massive star cluster that can provide a large mechanical luminosity from its many type II supernovae. \item The consistent reversal of strengths between the blue and red components in the northern and southern outflows provides evidence of preferred outflow directions approximately perpendicular to the inclined and flattened H\one\ disc \citep{stil02}. \item In addition to characterising the H$\alpha$ line profile, we have also measured [S\two] derived electron densities and [S\two]$\lambda$6717+$\lambda$6731/H$\alpha$ line ratios. We find that much of the ionized gas in the broad component region and in the outer-wind regions is at or below the low density limit, as is expected for a rarefied outflow. We find the highest [S\two]/H$\alpha$ ratios are associated with the faintest H$\alpha$ fluxes and the largest galactocentric distances. log([S\two]/H$\alpha$) ratios $>$0 are found in the superbubbles E, B and A, indicating that in these regions the gas emission may be significantly shock-excited. \end{itemize} In summary, the outflow in NGC~1569 appears to consist of several superbubbles in various phases of development, as noted by \citet{martin98}. This situation is further complicated by the disturbed state of the H\one\ that includes features well out of the main plane of the galaxy (Fig.~\ref{fig:xray_HI}b). Our data confirm this model and thus indicate that the evolution of the outflow will be determined by the development of the superbubbles. In superbubbles A, B, and F ionized gas arcs seen in H$\alpha$ and the X-ray morphology suggest that much of the hot gas still is confined, albeit moving at velocities that are comparable to those needed to escape. Thus the NGC~1569 outflow does not currently appear to be in the form of an approximately steady state galactic wind, even though it may eventually lead to mass loss from the system. It therefore differs from the well known M82 outflow, whose outer regions can be modelled by a supersonic galactic wind \citep[e.g.][]{suchkov94, shopbell98, zirakashvili06}. | 7 | 10 | 0710.4354 |
0710 | 0710.2003_arXiv.txt | The apparent alignment of the cosmic microwave background multipoles on large scales challenges the standard cosmological model. Scalar field inflation is isotropic and cannot account for the observed alignment. We explore the imprints, a non-standard spinor driven inflation would leave on the cosmic microwave background anisotropies. We show it is natural to expect an anisotropic inflationary expansion of the Universe which has the effect of suppressing the low multipole amplitude of the primordial power spectrum, while at the same time to provide the usual inflationary features. | 7 | 10 | 0710.2003 |
||
0710 | 0710.0837_arXiv.txt | \citet{chang06} reported millisecond duration dips in the X-ray intensity of Sco X-1 and attributed them to occultations of the source by small trans-Neptunian objects (TNOs). We have found multiple lines of evidence that these dips are not astronomical in origin, but rather the result of high-energy charged particle events in the {\it RXTE} PCA detectors. Our analysis of the {\it RXTE} data indicates that at most 10\% of the observed dips in Sco X-1 could be due to occultations by TNOs, and, furthermore, we find no positive or supporting evidence for any of them being due to TNOs. We therefore believe that it is a mistake to conclude that any TNOs have been detected via occultation of Sco X-1. | \label{sec:intro} \citet{chang06} found statistically significant 1-2 millisecond duration dips in the count rate during X-ray observations of the bright X-ray source Sco X-1 carried out with the Proportional Counter Array (PCA) on the {\it Rossi X-ray Timing Explorer} ({\it RXTE}) and attributed them to occultations of the source by small objects orbiting the Sun beyond the orbit of Neptune, i.e., trans-Neptunian objects (TNOs). In all, \citet{chang06} found some 58 dips in approximately 322 ks of Sco X-1 observations. Given that the {\it RXTE} spacecraft moves through the diffraction-widened shadows of any TNOs at a velocity of $\sim30$ km s$^{-1}$, dips of $\sim$2 ms duration should correspond to a TNO size of $\sim$60 m. If the identification of these dips with occultations by TNOs is correct, the dips would provide extremely valuable information on the number and distribution of solar system objects of $\sim20$-100 m in size. We have found evidence that these dips are produced by electronic dead-time as a result of high-energy charged particle events in the RXTE PCA detectors. Preliminary reports of our results were given by \citet{tajatel, tajast}; herein we give a more detailed and complete report. | \label{sec:disc} The observed dips have widths and depths that are approximately what one might expect to be produced by occultations by TNOs, even though much wider dips would be detectable in principle (given appropriate depths). Thus we are obligated to seriously consider the hypothesis that some or all of the observed dips are the product of TNO occultations. However, close examination of the {\em RXTE} PCA data reveals six signatures that independently indicate that few and possibly none of the observed dips are due to occultations by TNOs. The signatures are (1) the numbers of SM1-other events during the dips; (2) the numbers of VLE events during the dips; (3) the absence of the expected diffraction sidelobes; (4) the temporal asymmetry of the dips; (5) the almost total lack of dips longer than $\sim$1 ms; and (6) the lack of correlation between dip duration and depth. We discuss each of these in turn. (1) {\bf SM1-Other Events:} Fig.~\ref{fig:xraylc} shows that there is, on average for 201 dips, a large excess of SM1-other events at or near the times of the dips. On average, individual dips should show an excess of SM1-other events at the $2.7\, \sigma$ level. In Fig.~\ref{fig:otherhist} we show a histogram of the number of SM1-other events in the 1/8-s time bins corresponding to the dips expressed in standard deviations from the mean. The mean value obtained from the histogram is $\sim2.7\, \sigma$, as expected. If one makes the reasonable assumption that the numbers of SM1-other events should not be affected by true occultations (other than by negligible increases due to reductions in the electronic dead time), then one may estimate the maximum fraction, $f_{\rm occ}$, of the observed dips that represent genuine occultation events that is consistent with this distribution of SM1-other events. We constructed the following simple function with which to fit the histogram, and thereby constrain $f_{\rm occ}$: \begin{equation} {\rm probability} = \frac{1}{\sqrt{2\pi}}f_{\rm occ} e^{-C^2/2} + \frac{1}{\sqrt{2\pi \sigma_{cr}^2}}(1-f_{\rm occ})e^{-(C- \overline{C}_{cr})^2/2\sigma_{cr}^2} \end{equation} where $C$ is the number of excess SM1-other counts (in units of standard deviations of the counts per bin in each PCA light curve), and $\overline{C}_{cr}$ is the mean of $C$ for those dips which are not the results of occultation events and which we take to be $\approx2.7/(1-f_{\rm occ})$. The distribution of the numbers of excess SM1-other counts is wider than what would be expected from a Poisson distribution with a mean equal to the slightly increased (on average) number of SM1-other events per bin; the width of this component of the fitting function is adjusted by means of the parameter $\sigma_{cr}$. Fits of the function to the histogram in Fig.~\ref{fig:otherhist} were carried out with $\chi^2$ fits using both Gaussian and Cash (1979) statistics. If we neglect the tail of the distribution at high numbers of excess SM1-other events, i.e., at $>9\sigma$, we obtain formally acceptable fits with values of $f_{\rm occ}$ in the range 0.0 to 0.12 and values of $\sigma_{cr}$ in the range 1.85 to 2.35 (based on Gaussian statistics; the limits represent the formal joint 95\% confidence range). Using Cash statistics, we obtain formally acceptable fits with values of $f_{\rm occ}$ in the range 0.0 to 0.11 and values of $\sigma_{cr}$ in the range 1.65 to 2.15 (95\% confidence). These results indicate that fewer than 11\% of the 203 dips might be the product of TNO occultations. \begin{figure*} \epsscale{0.85} \plotone{f7.eps} \caption{Histogram of the numbers of excess SM1-other events in 1/8-s time bins corresponding to the times of 201 of the 203 dips. The number of excess SM1-other events is given in terms of the square root of the mean number of SM1-other events per bin away from the time of the dip. The solid smooth curve represents the best fit with $f_{\rm occ} = 0$, $\sigma_{cr} = 2.1$ and the dashed curve represents a formally {\em un}acceptable fit with $f_{\rm occ} = 0.24$, $\sigma_{cr} = 1.8$ (see text). } \label{fig:otherhist} \end{figure*} (2) {\bf VLE events:} Figure~\ref{fig:xraylc} also shows that there is an excess of VLE events around the times of the dips. The difference between the background rate and that in the 1/8-s bin containing the dips is very close to 4 (actually $3.9 \pm 0.5$) extra VLE events per dip. The peak in Fig.~\ref{fig:xraylc} is significant at the $7 \sigma$ level. If there is precisely one VLE per operating PCU for each non-TNO dip, then we would expect on average an excess of 4.67 VLEs per non-TNO dip. If only the non-TNO dips contribute to the excess VLE events then there is an upper limit to $f_{\rm occ}$ that is consistent with the observations. If we further allow that the statistical mean excess number of VLE events per dip may have been as small as 2.9, then a simple calculation gives the limit $f_{\rm occ} \lesssim 0.38$ (95\% confidence). This limit is weaker than for the SM1-other events, and, furthermore, is compromised by the possibility that more than one VLE event could be produced in a operating PCU in a single cosmic-ray induced dip. (3) {\bf Lack of diffraction sidelobes:} In Fig.~\ref{simprof} we showed an average profile of simulated dips that had been inserted into actual PCA data. It should be compared to averages of the actual measured dip profiles in Figs.~\ref{fig_superpos_202} and \ref{fig_superpos_109}. The average model dip profile shows a clear bump of $\sim8$\% amplitude on either side of the dip due to diffraction, whereas the averages of the actual profiles show no significant evidence for diffraction sidelobes. Thus, we conclude that the fraction $f_{\rm occ}$ of legitimate TNO occultations can be no larger than $\sim$30\%, otherwise diffraction sidelobes likely would have been detected. Again, while this is a clear strike against the dips being due to TNOs, the limiting statistically significant constraint that can be set due to the lack of diffraction sidelobes is not as significant as for the SM1-other events. (4) {\bf Asymmetry:} A comparison of the simulated with the actual dip profiles (as in [3]) above, clearly shows a marked asymmetry for the real dip events. This is physically implausible if the dips are the product of occultation events and therefore testifies against a TNO origin for most of the dip events. We estimate that the statistical significance of the asymmetry is $\sim 6\,\sigma$. Unfortunately, there is no direct way to use this information to constrain the fraction of legitimate TNO occultations. The problem is that we do not know, a priori, how large the asymmetry is, on average, for non-TNO dips. Therefore, we can not tell how `diluted' the non-TNO events are by potentially real ones. Nonethless, this marked asymmetry is another solid indication that few of the dips are the product of TNO occultations. (5) {\bf Lack of dips longer than $\sim$1 ms:} From Fig.~\ref{fig_realwd} we can see that all of the dips, except for a single event, have RMS widths $\sigma < 1.1$ ms. In Section~\ref{sec:data} we described a computer simulation of the production, detection, and analysis of dips caused by TNO occultations. For a relative speed between the {\it RXTE} satellite and the shadows of the putative TNOs of $v_{\rm rel} = 25$ km s$^{-1}$ we find that the fraction of recovered simulated dips with $\sigma > 1.1$ ms is $\sim$27\%. For $v_{\rm rel} = 35$ km s$^{-1}$, $9$\% of the dips have $\sigma > 1.1$ ms. We estimate that the average relative velocity between {\it RXTE} and the shadows of any TNOs was not higher than $v_{\rm rel} \sim 30$ km s$^{-1}$. For this speed, $16$\% of the dips are characterized by $\sigma > 1.1$ ms. Therefore, if {\em all} of the dips are the result of TNO occultations the number of longer-duration dips should be $\sim$30, whereas the observed number is actually 1. On the other hand, if only 15\% of the dips are due to TNO occultations, then we would expect only $\sim 5$ dips with $\sigma > 1.1$ ms. This expected number is marginally statistically consistent, i.e., at $\sim$5\% confidence, with the detection of one dip with $\sigma > 1.1$ ms. Therefore, we conclude that the lack of longer dips allows an upper limit of 15\% to be set on the fraction, $f_{\rm occ}$, of potentially real TNO occultations. (6) {\bf Lack of correlation between width and depth:} If the dips were due to TNO occultations of Sco X-1, we would expect a strong correlation between the widths of the dips and their depths. This results from the fact that diffraction produces shallow occultations for the smaller size occulters, while it produces deeper more geometric-shadowing-like occultations for the larger occulters. As can be seen from the distribution of dip widths vs. depths in Fig.~\ref{fig_realwd}, there is no such correlation, with almost all of the dips confined to a narrow range of widths (between 0.4 and 0.8 ms) and depths that range all the way from 45\% to nearly 100\%. Thus, the fact that the dips we detect include a significant number, i.e., $\sim$20\%, that are both narrow ($\sigma < 0.7$ ms) and deep (minimum normalized count rate below 0.2) whereas only $\sim2$\% of the `detected' simulated dips (for $v_{\rm rel} \sim 30$ km s$^{-1}$) are this narrow and deep, indicates that $\lesssim10$\% of the dips might be due to TNO occultations. Given the effects of statistical fluctuations on the observed number of narrow deep dips and the fact that the simulation is based upon somewhat uncertain parameters, it is more reasonable to use these numbers to set an upper limit of $\sim20$\%. Summarizing the results from approaches (1) through (6) above, we find limits on the fraction of valid TNOs to be $f_{\rm occ} <11\%$, $< 38\%$, $< 30\%$, $< Q\%$, $<15\%$, $<20\%$, respectively, where ``Q'' denotes that a formal limit could not be set, but the approach provides an important independent indication that the dips are, for the most part, not the result of TNO occultations. We believe that the combined upper limit on $f_{\rm occ}$ due to the joint application of all six approaches is simply the minimum value achieved by the most sensitive of these, i.e., the constraints cannot be combined. The reason, in short, is that the effects we explore serve only to statistically limit the number of events which could be due to TNOs rather than to identify specific qualifying events. Therefore, our final limit is simply $f_{\rm occ} \lesssim 10\%$. One might argue, as did \citet{chang07}, that since $\sim10$\% of the observed dips cannot be formally eliminated as being due to TNOs, they serve as viable potential candidates for TNO detections. However, we argue that if 90\% of the dips can be securely eliminated as TNO occultations, and there are six different and independent indicators that point in the direction of a common cause due to cosmic ray interactions in the detector, then it is most plausible that {\em all} of the dips have this common origin. While our results cast serious doubt on whether any true occultation events have been detected, one cannot yet conclude with a high degree of confidence that no such events have been detected. Further investigations of the dip phenomenon and its possible causes would be of interest. We are working to obtain a new measurement of, or upper limit on, the rate of occurrence of occultations of Sco X-1 by analyzing the data that are being obtained in a new series of {\it RXTE} observations of Sco X-1 with high-time-resolution information on VLE events. | 7 | 10 | 0710.0837 |
0710 | 0710.3252.txt | % The role of optical Fe\,{\sc iii} absorption lines in B-type stars as iron abundance diagnostics is considered. To date, ultraviolet Fe lines have been widely used in B-type stars, although line blending can severely hinder their diagnostic power. Using optical spectra, covering a wavelength range $\sim$ 3560 -- 9200 \AA, a sample of Galactic B-type main-sequence and supergiant stars of spectral types B0.5 to B7 are investigated. A comparison of the observed Fe\,{\sc iii} spectra of supergiants, and those predicted from the model atmosphere codes {\sc tlusty} (plane-parallel, non-LTE), with spectra generated using {\sc synspec} (LTE), and {\sc cmfgen} (spherical, non-LTE), reveal that non-LTE effects appear small. In addition, a sample of main-sequence and supergiant objects, observed with FEROS, reveal LTE abundance estimates consistent with the Galactic environment and previous optical studies. Based on the present study, we list a number of Fe\,{\sc iii} transitions which we recommend for estimating the iron abundance from early B-type stellar spectra. | %----------------------- Iron lines dominate the spectra of many astrophysical objects, such as novae \citep{{mck97},{hat07}}, photoionized H II regions \citep{{rub97},{rod02},{est02}} and active galactic nuclei \citep*{{sig03},{sig04},{zha06}}. The atomic processes for iron and other iron-group ions have been the subject of numerous investigations, for example the IRON Project \citep{hum93} which considers applications in astrophysical and laboratory plasmas \citep{pra96a}. Absorption lines of iron provide important metallicity diagnostics for both stars and galaxies, and also play a key role in investigating star formation histories through element ratios, such as [$\alpha$/Fe] \citep{gil98}. However, there is a lack of robust Fe abundance determinations in external galaxies, e.g. the Magellanic Clouds (see for example, \citealt*{{rol02},{tru02},{tru07}} and \citealt{mok07}), due to the complex Grotian diagrams for Fe, the low metallicity environment of the Magellanic Clouds and the reliability of the currently available atomic data. Early B-type stars are important for studying the chemical composition of our own and other galaxies \citep{{kil92},{duf98}}. In the optical spectra of B-type stars, iron lines due to a number of ionization stages are observed (see for example, \citealt*{{gie92},{len93},{sma97},{mor06}}). Absorption features arising from Fe\,{\sc iii} are primarily detected \citep{har70}, with Fe\,{\sc ii} also found in later B-type stars \citep{pin93}. Fe\,{\sc iv} lines are not expected in the optical spectra of B-type stars due to their intrinsic weakness in this temperature range. The optical Fe\,{\sc iii} line spectrum has not been widely employed to determine abundances, as it has often been believed to be too weak to provide reliable measurements \citep{ken94a}. However, it has been used for chemical composition studies of several bright, narrow-lined main-sequence B-type stars, such as $\zeta$\,Cas, $\gamma$\,Peg, $\iota$\,Her, $\tau$\,Sco and $\lambda$\,Lep \citep{{sni69},{pet70},{har70},{pet76},{pet85}}. These Galactic objects have sufficiently high Fe content, coupled with narrow metal absorption lines (due to their low projected rotational velocities), so that, even with relatively poor quality optical spectra, Fe\,{\sc iii} features can be detected. However, the quality of the available spectra did not allow the study of Fe\,{\sc iii} lines in objects of lower metallicity. Therefore, more recently the optical wavelength range has been largely overlooked in favour of the ultraviolet domain \citep{{swi76},{pet85},{dix98}}, and in particular the very rich spectral region around 1900 \AA\ \citep*{{tho74},{heb83},{ken94a},{gri96},{moe98}}. On the other hand, due to the high density of absorption features and resultant blending, continuum placement in the ultraviolet is difficult, and hence significant errors may be present in the derived abundances \citep{moe98}. Due to instrumental advances in more recent years, it has become routine to obtain high resolution and signal-to-noise (S/N) spectra, and the use of the optical Fe\,{\sc iii} lines as a diagnostic has been revisited (e.g. \citealt{{cun94},{kil94}}). A number of main-sequence objects have subsequently been re-examined, for example $\gamma$\,Peg and $\iota$\,Her \citep*{{pin93},{zon98}}, producing results consistent with the earlier optical analyses. In addition, the higher S/N ratios of the available spectra have made it possible to detect the weak optical Fe\,{\sc iii} lines in a number of lower metallicity objects, such as AV\,304 in the Small Magellanic Cloud \citep{rol03}, and the globular cluster post-AGB stars ZNG-1 in M\,10 \citep{moo04}, Barnard\,29 in M\,13 and ROA\,5701 in $\omega$\,Cen \citep{tho07}. A number of B-type stars have been studied using both ultraviolet and optical spectra, but the corresponding abundance estimates are generally in poor agreement. Estimates from the ultraviolet spectra are consistently lower than those from the optical, for example in the post-AGB stars Barnard\,29, ROA\,5701 \citep{tho07}, BD\,+33${^\circ}$2642 \citep*{nap94} and HD\,177566 \citep{{ken94a},{ken94b}}. Young Galactic objects including $\gamma$\,Peg \citep{moe98} and $\iota$\,Her \citep{gri96}, plus stars in known metallicity environments such as the Magellanic Clouds \citep{duf07}, also display similar discrepancies. Here, high resolution spectra for a number of narrow-lined Galactic B-type main-sequence and supergiant stars, covering a range of spectral types, are analysed using LTE and non-LTE model atmosphere techniques. Details on the observations, models and Fe\,{\sc iii} line selection can be found in Sections \ref{sec_obs} and \ref{sec_data}. An LTE approximation is considered, and the reliability of such an assumption, along with an assessment of the individual Fe\,{\sc iii} lines, is discussed in Section \ref{sec_discuss}. %----------------------- | \label{sec_conc} %----------------------- Due to their intrinsic weakness, Fe\,{\sc iii} absorption lines have not been widely considered for use in chemical composition studies. Instead, the very rich ultraviolet spectral region has been favored. However, the results found in this paper suggest that the optical region can provide consistent results. The stars here display abundance estimates that agree with the Galactic metallicity, and are consistent with previous studies using optical spectra, where available. By contrast, previous determinations from ultraviolet spectra have followed the trends observed in other studies (for example \citealt{{tho07},{duf07}}), providing lower abundance estimates, in these cases by approximately 0.5 dex. Although our study has concentrated on B-type stars found in the Milky Way, the optical Fe\,{\sc iii} lines examined here can be applied to studies of B-type stars in other galaxies, provided suitable S/N spectra are employed. For example, \citet{tru02} analysed a sample of B-type supergiants in M\,31, obtaining an Fe abundance from the Fe\,{\sc iii} line at 4419 \AA. \citet{rol02} analysed a sample of OB-type main-sequence stars from the LMC, finding an Fe abundance for one object (PS\,34-16), while \citet{rol03} investigated a B-type dwarf from the SMC (AV\,304), finding agreement with other giant stars. More recently, \citet{tru07} used similar methods to those detailed here and obtained Fe abundances for B-type stars in the Galaxy, LMC and SMC, using the two Fe\,{\sc iii} lines at 4419 and 4430 \AA. The results found were consistent with the present accepted metallicities of these systems. These studies indicate that the optical Fe\,{\sc iii} lines can provide reliable abundance indicators in different galaxies. Our comparison of stars analysed using the model atmosphere codes {\sc cmfgen} and {\sc tlusty} generally shows little difference in the abundance estimates, indicating that the different physical assumptions, in particular non-LTE effects, are small for this species. Therefore, the results suggest that the optical Fe\,{\sc iii} absorption line spectrum may be used with confidence in chemical composition studies, and an LTE analysis provides reliable results. The atomic data of \citet{nah96}, employed both in this paper and by \citet{cro06}, appear to provide appropriate abundances, although there are some features, such as those at 4005 and 4273 \AA, whose log\,{\it gf} values may be incorrect. Comparing the values in Table \ref{tab_atdata} shows that, for some features, e.g. the 4166.88, 4419.60 and 5272.90 \AA\ lines, there are large differences between the atomic data from the Kurucz database and \citet{nah96}. Further work is required to refine the atomic data for this species. \begin{table} \caption{Recommended Fe\,{\sc iii} lines for use as abundance diagnostics.} \label{tab_touse} \begin{tabular}{@{}lcccc} \hline Line & Spectral Type &Line & Spectral Type \\ (\AA) & Range &(\AA) & Range \\ \hline 4419 & B0.5--B7 &5156 & B0.5--B7 \\ 4431 & B1--B7 &5272 & B1.5--B5 \\ 5063 & B1.5--B7 &5282 & B2--B7 \\ 5086 & B0.5--B7 &5299 & B1.5--B4 \\ 5127 & B0.5--B7 &5302 & B1.5--B4 \\ \hline \end{tabular} \end{table} In Table \ref{tab_touse} we list recommended Fe\,{\sc iii} lines which we believe, based on the present study, will provide reliable diagnostics for the iron abundance in early B-type stars. Lines have been selected based on the following criteria: \begin{itemize} \item Relatively strong, isolated feature, free from known blends. \item Yields an abundance estimate within $\pm$ 0.2 dex of the mean iron abundance for all well observed lines. \end{itemize} \begin{figure*} \includegraphics[angle=0,width=0.8\textwidth]{figure3_a.eps} \caption{Observed Fe\,{\sc iii} lines in HD\,79447 as listed in Table \ref{tab_touse}, including the 5193 \AA\ line. Overplotted are theoretical fits (smooth line) to the Fe\,{\sc iii} lines, calculated using the average abundance estimate for the star of 7.63 dex.} \label{fig_felines} \end{figure*} The range of spectral types over which it is advisable to use the lines as an abundance diagnostic is also listed in Table \ref{tab_touse}. This spectral type range is not the same for all lines, generally due to the fact that some of the weaker transitions are only detected over a restricted span of spectral types. Observations for all of the lines in this Table are shown in Fig. \ref{fig_felines} for HD\,79447. The line at 5193 \AA\ has been included in the Figure, but not the Table, because its derived abundance is more than 0.2 dex larger than the mean value for three of the stars studied here, namely HD\,108002 (B1), HD\,142768 (B1.5) and HD\,53138 (B3). However, the feature, along with others observed (see Tables \ref{tab_tcab}, \ref{tab_sab} and \ref{tab_mab}) may be suitable for diagnostic use, depending on the quality of the spectra used. We note that the stars included in this study have been selected due to being narrow lined and are typical of their individual luminosity classes. However, if objects with larger $\upsilon$\,sin{\it i} values were employed, blending between Fe lines and other metal features may occur, thus, care should be taken when using objects with larger projected rotational velocities. | 7 | 10 | 0710.3252 |
0710 | 0710.2145_arXiv.txt | \footnote{To appear in proceedings of the Puerto Vallarta Conference on ``New Quests in Stellar Astrophysics II: Ultraviolet Properties of Evolved Stellar Populations'' eds. M. Chavez, E. Bertone, D. Rosa-Gonzalez \& L. H. Rodriguez-Merino, Springer, ASSP series.} Recent HST/ACS images of M82 covering the entire galaxy have been used to detect star clusters. The galaxy is known to contain a young population (age $<10^7$~yr) in its starburst nucleus, surrounded by a post-starburst disk of age $<10^9$~yr. We detect more than 650 star clusters in this galaxy, nearly 400 of them in the post-starburst disk. These data have been used to derive the luminosity, mass and size functions separately for the young nuclear, and intermediate-age disk clusters. In this contribution, we discuss the evolutionary status of these clusters, especially, on the chances of some of these clusters surviving to become old globular clusters. | \label{sec:1} Super star clusters (SSCs) and globular clusters (GCs) represent the youngest and the oldest stellar aggregates known in the Universe. The environments in which these two kinds of clusters are found are vastly different --- SSCs are found in violent star-forming regions, whereas GCs are found in the halos of galaxies. Yet, the similarity in their compactness and mass, is a reason compelling enough to think of an evolutionary connection between them. The growing popularity of the hierarchical model of galaxy formation in the years following the discovery of SSCs, and the possibility of observing the epochs of galaxy (and GC) formation at high redshifts, have also generated interest in looking for a common origin for these two seemingly different classes of clusters. In order to investigate the relation between the two types of clusters, it is important to analyze the survival of SSCs for a Hubble time. Star clusters are vulnerable to a variety of disruption processes that operate on three different timescales \citep[see][for more details] {Fal01, Mai04, deG07}. On short timescales ($t\sim10^7$~yr), the exploding supernovae and the resulting superwinds are responsible for cluster expansion and disruption, a process popularly dubbed as infant mortality. On intermediate timescales ($10^7<t<{\rm few} \times 10^8$~yr), the mass-loss from evolving stars leads to the disruption of the clusters. On even longer timescales ($t> {\rm few} \times 10^8$~yr), stellar dynamical processes, especially evaporation due to two-body scattering, and tidal effects on a cluster as it orbits around the galaxy, known as gravitational shocks, come into play in the removal of stellar mass from clusters. The GCs represent those objects that have survived all these processes, whereas young SSCs are just experiencing them. Intermediate age SSCs are the ideal objects to investigate the influence of disruption processes on the survival of star clusters. Almost all the star formation in the disk of M82 took place in a violent disk-wide burst around 100--500~Myr ago, following the interaction of M82 with the members of M81 group \citep{May06}. Cluster formation is known to be efficient during the burst phase of star formation \citep{Bas05}, and hence we expect large number of clusters of intermediate age ($\sim100$~Myr) in its disk. Hence, M82 offers an excellent opportunity to assess the evolutionary effects on the survival of star clusters, and to look for a possible evolutionary connection between the SSCs and GCs. | Luminosity and Mass functions of star clusters in M82 follow power-law functions, with the power law index showing a tendency for flattening of the profile with age. In other words, there is a deficiency of low-mass clusters among the older clusters. We also find the mean size of the older clusters to be smaller as compared to the younger clusters for masses $<10^5$~\msun. These two results together imply the selective destruction of loose clusters. The tidal forces experienced by the clusters as they orbit around the galaxy lead to exactly such a destruction process. If this process continues in M82, the LF of surviving clusters can mimic the presently observed LF of the Galactic GCs, provided the clusters move around the galaxy in highly elliptical orbits, with perigalactic distance as small as 350~pc. The resulting LF contains 85 clusters with the function peaking at the same luminosity as for the Galactic GCs at 5~Gyr age, and fainter by $\sim$0.5~mag at 10~Gyr. On the other hand, if the clusters move in nearly circular orbits, the LF will retain the power-law form, with the number of surviving clusters even higher. \vspace*{0.3cm} This work is partly supported by CONACyT (Mexico) research grants 42609-F and 49942-F. We would like to thank the Hubble Heritage Team at the Space Telescope Science Institute for making the reduced fits files available to us. | 7 | 10 | 0710.2145 |
0710 | 0710.0189_arXiv.txt | I review the evolutionary connection between low-mass X-ray binaries (LMXBs) and pulsars with binary companions (bPSRs) from a stellar binary evolution perspective. I focus on the evolution of stellar binaries with end-states consisting of a pulsar with a low-mass ($<1.0 \msun$) companion, starting at the point the companion's progenitor first initiates mass transfer onto the neutron star. Whether this mass transfer is stable and the physics driving ongoing mass transfer partitions the phase space of the companions's initial mass and initial orbital period into five regions. The qualitative nature of the mass-transfer process and the binary's final end-state differ between systems in each region; four of these regions each produce a particular class of LMXBs. I compare the theoretical expectations to the populations of galactic field LMXBs with companion-mass constraints and field bPSRs. I show that the population of accreting millisecond pulsars are all identified with only two of the four LMXB classes and that these systems do not have readily identifiable progeny in the bPSR population. I discuss which sub-populations of bPSRs can be explained by binary evolution theory and those that currently are not. Finally I discuss some outstanding questions in this field. | Since the discovery of the class prototype \citep{backer82}, there has been a posited evolutionary connection between millisecond radio pulsars (MSPs) and low-mass X-ray binaries (LMXBs)---mass transferring binaries with a neutron star (NS) accretor and donor companion with a mass $M_2 \lesssim 1 \msun$ \citep{alpar82}. The central idea behind this connection is that LMXBs can provide the long-lived phase ($\sim \mathrm{Gyr}$) of moderate mass transfer rates ($\Mdot \lesssim \Mdot_{\mathrm{Edd}} \approx 10^{-8} \msun\,\mathrm{yr}^{-1}$, where $ \Mdot_{\mathrm{Edd}}$ is the Eddington mass-transfer rate) thought necessary to spin-up the NS to spin periods of $\Pspin < 10 \mathrm{ms}$ as observed in the MSP population. \begin{figure} \includegraphics[height=.425\textheight]{proc_eps_sys_bPSRs_lmxbs.eps} \caption{The galactic field population of radio pulsars with binary companions (filled circles) in the $M_2$-$\Porb$ plane. The data are from the ATNF catalog \citep{manchester05}. Plotted are each system's minimum $M_2$, with horizontal lines extending to the system's median $M_2$ (assuming $i$ is randomly distributed). Filled circle size indicates each pulsar's $\Pspin$ (see plot legend). Symbols circumscribing a filled circle indicates the binary's eccentricity (again see plot legend). For comparison, LMXBs with independent $M_2$ estimates are plotted with open stars and filled triangles, the latter indicating the minimum $M_2$ for \emph{accreting MSP} systems. For the open stars, the dash-dotted lines indicate the \emph{total} estimated $M_2$ range in each system. For the accreting MSPs, the dash-dotted lines extend to the median $M_2$. This sample of LMXB systems represents a union of systems with $M_2$ estimates in the \citet{ritter98} catalog and a targeted literature search on LMXB systems where $\Pspin$ has been determined. Thus, this plot likely does not present in total the current census of LXMBs with $M_2$ estimates. } \label{fig:bPSRs_sys} \end{figure} Observational evidence for millisecond variability in LMXBs has been growing steadily in the form of detections of kilohertz quasi-periodic oscillations (kHz QPOs), X-ray burst oscillations, and accretion-powered oscillations \citep[see][]{chakrabarty05}. In terms of support for the LMXB-MSP connection, pride of place has been given to the accretion-powered pulsations systems (also known as accreting MSPs) since the pulsation period in these systems are identifiable directly with $\Pspin$ \citep{chakrabarty05}. However, recent work has also shown that $\Pspin$ is of order the oscillation periods in the kHz QPO and X-ray burst oscillation sources \citep{chakrabarty03,strohmayer03,wijnands03,linares05}, establishing that these systems also harbor a rapidly rotating NS. Thus, it is now well established that NSs in LMXBs can be spinning rapidly enough to produce radio MSPs once the LMXB phase ends. However, understanding fully the evolutionary connection between LMXBs and MSPs requires not only explaining the NS's spin evolution but also accounting for other properties seen in the radio MSP population. In particular, this includes the distribution in orbital period, $\Porb$ of MSPs that retain a remnant binary companion, how this remnant's $M_2$ correlates with $\Porb$, the distribution of binary eccentricity, $e$, and the production of \emph{isolated} MSPs. Indeed, the ideal test of accretion-torque theory would be accomplished by understanding the evolution in the LMXB-phase well enough to correlate final $\Pspin$ with these other quantities. This is a rather ambitious goal since the $\Pspin$ evolution depends not only on the secular $\Mdot$ evolution but also on the efficiency with which matter accretes onto the NS, whether accretion onto the NS occurs sporadically due to disk instabilities, how $\Pspin$ evolves during mass-transfer outbursts and when unstable disks are in quiescent phases, and how each LMXB transitions into an MSP system. Turning from where one would like to be to where we are now, my goal for this contribution is to approach the LMXB-MSP connection from the vantage point of stellar binary evolution theory. To do so, I will expand the view somewhat and review our understanding of NS-main sequence (MS) binaries whose evolutionary endpoints are NSs with a low-mass ($M_2 \lesssim 1.0 \msun$) binary companion. In doing so, the focus of the discussion will shift from solely the MSP population to making connections between NS-MS binaries and various populations of radio pulsars in binaries, bPSRs (one would like to simply say ``binary pulsars'' here, but the discovery of \emph{the} Binary Pulsar \citep{burgay03} has lead to this term often causing confusion; I'll leave it to the reader to decide whether the scientific windfall from this system compensates sufficiently for necessitating such unwieldy terminology). The starting point for this discussion is Figure \ref{fig:bPSRs_sys}, which shows the location of galactic field bPSRs in the $\Porb$-$M_2$ plane by the filled circles \citep{manchester05}. Only field sources are included so as to compare theory to a sample of bPSRs whose properties have not been influenced by dynamic interactions. The filled circles indicate each bPSR's minimum-$M_2$; the horizontal lines extend to each system's median $M_2$ (corresponding to a binary inclination of $i = 60^\circ$). Spin period and $e$ are encoded via filled circle size and circumscribed symbols, respectively. Filled triangles show the same information for accreting MSP systems \citep{chakrabarty98,galloway02,galloway05,kaaret06,krimm07,markwardt02,markwardt03}. Open stars indicate other field LMXB systems with $M_2$ determinations \citep{bhattachar06,casares06,cominsky89,cornelisse07,finger96,heinz01,hinkle06,jonker01,parmer86,pearson06,reynolds97}; for these latter systems, the dashed-dotted horizontal lines show the estimated range of $M_2$ (not its median value). In the following, I'll compare the predictions of binary evolution theory for how systems evolve in this $\Porb$-$M_2$ plane to the location of systems in Fig. \ref{fig:bPSRs_sys}. While this will allow tentative positive identifications of the evolutionary connections discussed above, it will also serve to highlight sub-populations of bPSRs whose formation is \emph{not} currently explained by binary evolution theory. I'll discuss the evolution of NS-MS binaries, focusing on how initial conditions lead to four different classes of X-ray binaries. I'll compare between this theory and the observations, pointing out where agreement between the two is better and worse. Finally, I'll close by discussing several open questions related to the evolution of LMXBs and bPSR formation. For other discussions and reviews of the bPSR population see \citep{lorimer05} and \citep{phinney94}. Also see \citep{flamb05} for a conference proceeding that reviews NS spin evolution under accretion. | 7 | 10 | 0710.0189 |
|
0710 | 0710.5881_arXiv.txt | We report observations of a radio burst that occurred on the flare star AD~Leonis over a frequency range of 1120-1620~MHz ($\lambda\approx$18--27~cm). These observations, made by the 305~m telescope of the Arecibo Observatory, are unique in providing the highest time resolution (1~ms) and broadest spectral coverage ($\Delta \nu/\nu=0.36$) of a stellar radio burst yet obtained. The burst was observed on 2005 April 9. It produced a peak flux density of $\sim 500$ mJy and it was essentially 100\% right-circularly polarized. The dynamic spectrum shows a rich variety of structure: patchy emission, diffuse bands, and narrowband, fast-drift striae. Focusing our attention on the fast-drift striae, we consider the possible role of dispersion and find that it requires rather special conditions in the source to be a significant factor. We suggest that the emission may be due to the cyclotron maser instability, a mechanism known to occur in planetary magnetospheres. We briefly explore possible implications of this possibility. | The use of radio dynamic spectra has played a central role in identifying and clarifying the physical mechanisms at work in the solar corona \citep[see][for reviews]{1985srph}. The application of similar techniques to active stars has long been an important goal, but it has been hampered by limitations in available instrumentation. Past studies of radio emission from M dwarf flare stars led to the discovery of extreme stellar radio bursts, characterized by close to 100\% circularly polarized emission with brightness temperatures in excess of 10$^{14}$K and durations less than a few tens of milliseconds \citep{gudel1989,bastian1990}. However, these spectroscopic investigations of the coherent radio bursts on flare stars have typically been limited by relatively long integration times \citep{bb1987,gudel1989} and/or limited frequency bandwidth ratios $\Delta\nu/\nu$ \citep[usually just a few percent; e.g.,][]{bastian1990, abada1997a}. The necessary combination of high time resolution and a large frequency bandwidth ratio has only been available infrequently \citep{stepanov2001, zaitsev2004}, precluding measurements of key parameters such as the intrinsic frequency bandwidth or frequency drift rate of the radio bursts, making the interpretation of these puzzling events difficult. It is only with the recent advent of radio spectrometers capable of supporting both a large bandwidth ratio and high time resolution simultaneously that progress in understanding the physics of radio bursts in the coronas of other stars becomes possible. AD~Leonis, a young disk star at a distance of 4.9 pc from the Sun, is one of the most active flare stars known, producing intense, quasi-steady chromospheric and coronal emissions \citep{hawleyetal2003, hunschetal1999, jackson1989} seen at UV, X-ray, and radio wavelengths. The star is also highly variable, producing flares from radio to X-ray wavelengths \citep[e.g.,][]{bastian1990, hawleypettersen1991, hawleyetal2003, hawleyetal1995, favataetal2000}. Its propensity for frequent and extreme radio bursts \citep[with intensities peaking at $>$ 500 times the quiescent radio luminosity of 5.5$\times$10$^{13}$ erg s$^{-1}$ Hz$^{-1}$;][]{jackson1989} makes it a frequent target for radio investigations of stellar flares. In a previous paper \citep[][hereafter, Paper I]{ob2006} we described the initiation of a pilot program to observe active M dwarfs with the Arecibo Observatory's Wideband Arecibo Pulsar Processor (WAPP), and first results from that program. Here we describe the next phase, which increased the time resolution by a factor of 10 to 1~ms. | We have described observations of a unique set of stellar radio bursts, which take advantage of the wide bandwidth and high time resolution capabilities of the Wideband Arecibo Pulsar Processor at the Arecibo Observatory. These ultra-high time resolution observations reveal phenomena that differ from those previously described using a similar observational setup, pointing out the complexity and diversity of processes likely occurring in stellar coronal plasmas. Whereas in Paper I we concluded that a plasma emission process appeared to be producing the two types of radio bursts observed in June 2003, in the current paper we prefer a different explanation, a cyclotron maser instability, for the fast-drift striae observed in April 2005. While all sets of phenomena show drifting structures of highly circularly polarized radiation, key discriminants between them are the durations and bandwidths of spectral features, as well as the magnitude and sign of the drift rates. In Paper I and here, we have demonstrated that the analysis of dynamic spectra of stellar radio bursts provide observational constraints which can be used as a measure to gauge the likelihood that a particular emission process is operative. Extensions of the current observational setup can look for dynamics at even higher time resolution, search for harmonic emissions over larger frequency bandwidths, expand the observational program to other dMe flare stars, and search for high time resolution behavior on other classes of active stars. Given the complexity of solar radio emissions at meter wavelengths compared with the already rich variety of decimetric phenomena, the observational results presented here for the dMe flare star AD Leo suggest that the next generation of radio instrumentation, particularly at metric wavelengths, promises to reveal a wealth of new phenomena which can diagnose plasma processes occurring in stellar coronae. As highly circularly polarized radio emission appears to be a common phenomenon on active stars, these spectacular radio bursts on M dwarf flare stars apparently represent the tip of the iceberg of stellar coronal plasma physics soon to be available for study. | 7 | 10 | 0710.5881 |
0710 | 0710.5286_arXiv.txt | We present new observations of the strongly-barred galaxy NGC~1365, including new photometric images and Fabry-Perot spectroscopy, as well as a detailed re-analysis of the neutral hydrogen observations from the VLA archive. We find the galaxy to be at once remarkably bi-symmetric in its I-band light distribution and strongly asymmetric in the distribution of dust and in the kinematics of the gas in the bar region. The velocity field mapped in the \ha\ line reveals bright HII regions with velocities that differ by 60 to $80\;$\kms\ from that of the surrounding gas, which may be due to remnants of infalling material. We have attempted hydrodynamic simulations of the bar flow to estimate the separate disk and halo masses, using two different dark matter halo models and covering a wide range of mass-to-light ratios ($\Upsilon$) and bar pattern speeds ($\Omega_p$). None of our models provides a compelling fit to the data, but they seem most nearly consistent with a fast bar, corotation at $\sim1.2r_B$, and $\Upsilon_ I \simeq 2.0 \pm 1.0$, implying a massive, but not fully maximal, disk. The fitted dark halos are unusually concentrated, a requirement driven by the declining outer rotation curve. | The centrifugal balance of the circular flow pattern in a near-axisymmetric spiral galaxy yields a direct estimate of the central gravitational attraction as a function of radius. However, the division of the mass giving rise to that central attraction into separate dark and luminous parts continues to prove challenging. The radial variation of the circular speed simply does not contain enough information to allow a unique decomposition between the baryonic mass, which has an uncertain mass-to-light ratio, $\Upsilon$, and the dark halo, whose density profile is generally described by some adopted parametric function \citep{albada3198, LF89, BSK04}. Predictions for $\Upsilon$ from stellar population synthesis models that match broad-band colors \citep[e.g.][]{Bell03} are useful, but not precise. Despite intense effort, they are still sufficiently uncertain to be consistent with both maximum and half-maximum disk, which is the range of disagreement \citep[e.g.][]{Sackett97, Bottema97, S99Rutgers}. \citet{McGaugh05} argues that the values can be refined by minimizing the scatter in the Tully-Fisher and/or mass discrepancy-acceleration relation. A number of dynamical methods have been employed to break the disk-halo degeneracy. \citet{Casertano83}, \citet{bosma98}, and others have suggested that the slight decrease in orbital speed near the edge of the optical disk of a bright galaxy -- the ``truncation signature'' -- could be used as an indicator of disk $\Upsilon$, but in practice it does not provide a tight constraint. \citet{ABP87} and \citet{Fuchs03} attempt to constrain the disk mass using spiral structure theory. \citet{Bottema97} and \citet{verheijen04} measure the vertical velocity dispersion of disk stars in a near face-on galaxy, which they assume has the same mean thickness of similar galaxies seen edge-on \citep{KvdKdG}, to constrain the disk mass. A similar approach is reported by \citet{CD04} using velocity measurements of individual planetary nebulae. One of the most powerful, although laborious, methods for barred galaxies was pioneered by Weiner, Sellwood \& Williams (2001), who made use of the additional information in the driven non-circular motions caused by the bar. By modeling the observed non-axisymmetric flow pattern of the gas in a 2-D velocity map, they were able to determine the mass-to-light ratio of the visible disk material. They found that the luminous disk and bar contributed almost all the central attraction in NGC~4123 inside $\sim 10\;$kpc, requiring the dark halo to have a very low central density. \citet{Ben3095} reports a similar result for a second case, NGC 3095. The method has also been applied by \citet{perez04} for several barred galaxies and by \citet{kranz03} who modeled motions caused by spiral arms. \citet{BEG03} present a similar study for the Milky Way. Earlier studies \citep[\eg][]{DA83} did not attempt to separate the disk from the dark matter halo \citep[see][for a review]{SW93}. Here, we apply the \citep{wein2} method to the more luminous barred galaxy NGC~1365 in the Fornax cluster. As one of the most apparently regular, nearby barred spiral galaxies in the Southern sky, NGC~1365 was selected by the Stockholm group for an in-depth study \citep[see \eg][]{Lindblad1365}. Hydrodynamic models of the bar flow pattern were already presented by \citep{Linlinatha}, based mainly on the velocities of emission-line measurements from many separate long-slit observations. J\"ors\"ater \& van Moorsel (1995, hereafter JvM95) present a kinematic study using the 21 cm line, which suggests that the galaxy is somewhat asymmetric in the outer parts, where the shape of the rotation curve is hard to determine. \citet{sandqvist} find substantial amounts of molecular gas, but only within 2~kpc of the nucleus, which is resolved in interferometric observations \citep{Sakamoto07} into a molecular ring in the center plus a number of CO hot spots. \citet{Galliano05} have found previously unknown MIR sources in the inner 10\arcsec\ around the AGN. They are able to correlate some of these MIR sources with radio sources, which they interpret in terms of embedded star clusters because of the lack of strong optical counterparts. \citet{JCA97} present H-band photometry of the bright inner disk, finding an elongated component in the central region suggesting that the NGC~1365 is a double-barred galaxy, although they note that the light in this component is not as smooth as in their other nuclear bar cases. \citet{LSKP} also classify it as a double barred galaxy. However, \citet{Emsellem01} and \citet{Erwin04} argue against a nuclear bar, citing an HST NICMOS image which resolves the feature into a nuclear spiral. \citet{Emsellem01} also present stellar kinematics from slit spectra using the $^{12}$CO bandhead. They propose a model for the inner 2.5 kpc of NGC 1365 consisting of a decoupled nuclear disk surrounded by spiral arms within the inner Lindblad resonance (ILR) of the primary bar. \citet{Beck05} observed NGC 1365 in radio continuum at 9\arcsec-25\arcsec\ resolution and find radio ridges roughly overlapping with the dust lanes in the bar region. They propose that magnetic forces can control the flow of gas at kiloparsec scales. Here we present new photometric images, a full 2-D velocity map of the \ha\ emission, and a reanalysis of the neutral hydrogen data from JvM95. We also compare many hydrodynamic models to these new data in an effort to determine the separate disk and dark halo masses in this galaxy. | We have presented a detailed study of the strongly-barred galaxy NGC~1365, including new photometric images and Fabry-Perot spectroscopy, as well as a detailed re-analysis of the neutral hydrogen observations by J\"ors\"ater \& van Moorsel (1995). We find the galaxy to be at once both remarkably bi-symmetric in its I-band light distribution and strongly asymmetric in the distribution of dust and gas, and in the kinematics of the gas. These asymmetries extend throughout the galaxy, affecting the bar region, the distribution of gas in the spiral arms and the neutral hydrogen beyond the edge of the bright disk. The velocity field mapped in the \ha\ line showed bright HII regions with velocities that differed by up to $\sim 80\;$\kms\ from that of the surrounding gas. Our sparsely-sampled line profiles in these anomalous velocity regions hint at unresolved substructure, suggesting a possible double line profile. The strong bar and spiral arms complicate the determination of the projection geometry of the disk, assuming it can be characterized as flat in the inner parts. The inclination of the plane we derive from the kinematic data is smaller by about 10$^\circ$ from that determined from the photometry. The strong spiral arms that cross the projected major axis far out in the disk seem likely to bias the photometric inclination and we therefore adopt, in common with other workers, the inclination derived from the gas kinematics. This preference is supported by the much poorer fits to the observed kinematics obtained when we adopt the photometric inclination (\S~\ref{comparsect}). Our attempts to derive the rotation curve of NGC~1365 were complicated by the fact that neither the \ha\ nor the HI velocity maps are consistent with a simple circular flow pattern over a significant radial range. The bar and spirals clearly distort the gas flow in the luminous disk. The neutral hydrogen extends somewhat beyond the visible disk but unfortunately has neither a uniform distribution nor regular kinematics. JvM95 attempted to fit a warp to the outer HI layer that extends into the visible disk, and derived a strongly declining rotation curve. We chose instead to assume a coplanar flow out to a deprojected radius of 255\arcsec\ and to neglect the asymmetric velocities in the neutral hydrogen beyond. The velocities derived separately from the \ha\ and HI data are in good agreement. Our resulting rotation curve shows a gentle decrease beyond a radius of $\sim 10\;$kpc, similar to those observed in other massive galaxies \citep{CvG91, noordermeer07}. We used our deprojected I-band image to estimate the gravitational field of the luminous matter, which can be scaled by a single mass-to-light ratio, $\Upsilon_I$. We also employed a gradual increase to $\Upsilon_I$ in the central few kpc to allow for an older bulge-like stellar population, although the light distribution does not appear to have a substantially greater thickness near the center. We combined the central attraction of the axially-symmetrized disk for various values of $\Upsilon_I$ with two different halo models to fit the observed rotation curve in the region outside the bar -- finding, as always, no significant preference for any $\Upsilon_I$. We attempt to fit hydrodynamic simulations of the gas flow pattern in the bar region, in order to constrain $\Upsilon_I$. For each type of halo adopted, we run a grid of simulations covering a range of both $\Upsilon_I$ and $\Omega_p$, the pattern speed of the bar. We then project each simulation to our adopted orientation of the galaxy and compare the gas flow velocities in the model with those observed. Since the light distribution in NGC~1365 is highly symmetric, our simulations were constrained to be bi-symmetric, yet the observed gas flow has strong asymmetries. None of our simulations is capable, therefore, of fitting both sides of the bar simultaneously. The anomalous position of the dust lane in the western part of the bar suggests that side is the more likely to be disturbed, and we therefore fit our models to the eastern half of the bar only. After smoothing the model to match the resolution of the kinematic data and masking out five blobs of gas with strongly anomalous velocities, we are able to obtain moderately satisfactory fits to the remaining velocities. The best fit pattern speed is $\Omega_p = 24\;$\kms$\rm{kpc}^{-1}$ for both types of halo which places corotation for the bar at $r_L \simeq 1.23r_B$, in excellent agreement with the value found by \citet{Linlinatha} and consistent with most determinations of this ratio for other galaxies \citep[\eg][]{ADC03}. Our estimated mass-to-light ratio values are $\Upsilon_I \simeq 2.50 \pm 1$ for the isothermal halo models and $\Upsilon_I=1.75 \pm 1$ for the NFW halo. While the constraints are disappointingly loose, the preferred mass-to-light ratio in both our halo models, $\Upsilon_I \simeq 2.0 \pm 1$, is consistent with that obtained by \citet{wein2} and \citet{Ben3095} in two other cases. For NGC~1365, however, this value implies a massive, but not fully maximal, disk, and we do not find support for the disk-only model with no halo that was suggested by JvM95. Although such a model can reproduce the declining rotation curve (see their Fig. 24), the simulated gas flow produced by such a model (which in the I band is $\Upsilon\sim 3.75$) is quite strongly excluded. The preferred value of $\Upsilon_I$ is nicely consistent with those obtained in the two previous studies using this method \citep{wein2, Ben3095}, but suggest somewhat more massive disks than predicted by population synthesis models \citep{BdJ,Bell03} for galaxies of these colors. The halos of our two models required to fit the declining rotation curve in the outer disk are distinctly non-standard, however. The circular speed in the pseudo-isothermal model declines steadily outside the large core, while the NFW halo has a very high concentration and small scale radius. Even allowing for compression of the halo as the massive disk forms within it, the original NFW halo has $c \simeq 22$, $V_{200} \simeq 123\;$kms. The total dark matter mass out to $r_{200}$ in this case is less than three times our estimated disk mass, and the halo is quite unlike those predicted by LCDM models for a galaxy of this mass. The disturbed distribution and kinematics of the gas in this galaxy clearly complicates our attempt to identify a preferred mass model. Its projected position near the center of the Fornax cluster, together with its velocity within 200~\kms\ of the cluster mean, suggest it is a cluster member. The disturbed nature of the outer HI distribution should not therefore be regarded as surprising. But the high central density of this massive galaxy should ensure that tidal forces have little influence on the inner part, where we indeed see that the bar and inner spirals are very pleasingly bi-symmetric in the I-band light. The existence of such strong asymmetries in the inner parts of the gas and dust is rather surprising, therefore. The asymmetry in the dust distribution and the kinematic map, combined with the existence of a number of patches of \ha\ emission with anomalous velocities all suggest that the agent that caused the disturbance was an infalling gas cloud. We cannot say whether the gas was an isolated intergalactic cloud not associated with a galaxy, or whether it could be a stream of debris from a gas-rich dwarf galaxy that had been tidally disrupted. The anomalous velocities clearly suggest that the infalling gas has yet to be assimilated in the disk of NGC~1365. | 7 | 10 | 0710.5286 |
0710 | 0710.5765_arXiv.txt | We examine halo gas cross sections and covering fractions, $f_c$, of intermediate redshift {\MgII} absorption selected galaxies. We computed statistical absorber halo radii, $R_{\rm x}$, using current values of $dN/dz$ and Schechter luminosity function parameters, and have compared these values to the distribution of impact parameters and luminosities from a sample of 37 galaxies. For equivalent widths $W_r(2796) \geq 0.3$~{\AA}, we find $43 \leq R_{\rm x} \leq 88$~kpc, depending on the lower luminosity cutoff and the slope, $\beta$, of the Holmberg--like luminosity scaling, $R \propto L^{\beta}$. The observed distribution of impact parameters, $D$, are such that several absorbing galaxies lie at $D>R_{\rm x}$ and several non--absorbing galaxies lie at $D < R_{\rm x}$. We deduced $f_c$ must be less than unity and obtain a mean of $\left< f_c \right> \sim 0.5$ for our sample. Moreover, the data suggest halo radii of {\MgII} absorbing galaxies do not follow a luminosity scaling with $\beta$ in the range of $0.2-0.28$, if $f_c= 1$ as previously reported. However, provided $f_c \sim 0.5$, we find that halo radii can remain consistent with a Holmberg--like luminosity relation with $\beta \simeq 0.2$ and $R_{\ast} = R_{\rm x}/\sqrt{f_c} \sim 110$~kpc. No luminosity scaling ($\beta=0$) is also consistent with the observed distribution of impact parameters if $f_c \leq 0.37$. The data support a scenario in which gaseous halos are patchy and likely have non--symmetric geometric distributions about the galaxies. We suggest halo gas distributions may not be govern primarily by galaxy mass/luminosity but also by stochastic processes local to the galaxy. | Understanding galaxy formation and evolution is one of the most important topics of modern astronomy. The extended distribution of baryonic gas surrounding galaxies holds great potential for constraining theories of their formation. However, the sizes of gaseous galaxy halos along with the distribution of gas within are not well understood. Numerical models have been able to synthesize the formation and evolution of large scale structures, however, there are unresolved issues regarding the evolution of individual galaxies and halos. The halo baryon--fraction problem \citep[e.g.,][]{mo02} and the rapid cooling of gas \citep[e.g.,][]{white78} result in galaxy halos which have little or no gas soon after they form. These effects are not seen in the observable universe since there is an abundance of galaxies where gas has been detected in halos via quasar absorption lines. From an observational standpoint, quasar absorption lines provides a unique means of probing the extent and abundance of halo gas. Although, quasar absorption line observations to date are sufficient the recognize the aforementioned problems, they are lacking the detail required to statistically constrain the distribution of the baryonic gas in the halos of simulated galaxies. Cross--correlations between absorbers and galaxies hold the promise to yield useful information on cloud sizes and halo gas covering fractions. First steps towards incorporating multi--phase gas in semi--analytical models and numerical simulations suggest that warm gas in halos extends out to galactocentric distances of $\sim 150$~kpc with cloud covering fractions of $\sim 0.25-0.6$ \citep{maller04,kaufmann06}. The association of {\MgIIdblt} doublet absorption in quasar spectra with normal, bright, field galaxies has been firmly established \citep[e.g.,][]{bb91,sdp94,cwc-china}. In an effort to understand halo sizes and gas distributions, \citet[][hereafter S95]{steidel95} searched for foreground galaxies associated with {\MgII} absorption within $\sim10''$ ($\sim65$~kpc for $z=0.5$) of quasars\footnote{Throughout we adopt a $h=0.70$, $\Omega_{\rm M}=0.3$, $\Omega_{\Lambda}=0.7$ cosmology. All quoted physical quantities from previously published works have been converted to this cosmology.}. The sample consisted of 53 absorbing and 14 non--absorbing galaxies with a {\MgII} $\lambda 2796$ equivalent width sensitivity limit of $W_{r}(2796) > 0.3$~{\AA}. S95 directly fitted the data by assuming a Holmberg--like luminosity scaling, \begin{equation} R(L) = R_{\ast} \left( \frac{L}{L^{\ast}} \right) ^{\beta} \quad {\rm kpc}, \label{eq:rl} \end{equation} and minimizing the number of non--absorbing and absorbing galaxies above and below the $R(L)$ relation. The best fit obtained clearly showed that absorbing and non--absorbing galaxies could be separated and that the halo radii $R(L_K)$ and $R(L_B)$ scale with luminosity with $\beta = 0.15$ and $\beta = 0.2$, respectively, where an $L_B^{\ast}$ galaxy has a gas halo cross section of $R_{\ast} = 55$~kpc. Furthermore, since almost none of the absorbing galaxies were observed above the $R(L)$ boundary and that almost none of the non--absorbing galaxies were observed below the $R(L)$ boundary, S95 inferred that {\it all\/} $L> 0.05L^{\ast}$ galaxies are hosts to {\MgII} absorbing gas halos characterized by a covering fraction of unity and a spherical geometry which truncates at $R(L)$. Examination of this now ``standard model'' has been the subject of several theoretical studies \citep[e.g.,][]{cc96,mo96,lin01}. \citet{gb97} determined a steeper value of $\beta = 0.28$ for the B--band luminosity obtained from a best fit to the upper envelope of the distribution of impact parameters of 26 absorbing galaxies. They found $R_{\ast} = 67$~kpc. Using a reverse approach of establishing foreground galaxy redshifts and then searching for {\MgII} absorption in the spectra of background quasars yields results inconsistent with a covering fraction of unity. For example, \citet{bowen95} identified 17 low--redshift galaxies with background quasar probing an impact parameter range between $3-162$~kpc. Galaxies that were probed at impact parameters greater than 13~kpc had no absorption in the halo ($W_{r}(2796) \geq 0.40-0.9$~{\AA}), however, four of the six galaxies within 13~kpc of the halo produced {\MgII} absorption. For intermediate redshift galaxies, \citet{bechtold92} reported a covering fraction $f_c \simeq 0.25$ for $W_{r}(2796) \geq 0.26$~{\AA} for eight galaxies with $D \leq 85$ kpc. Also, \citet{tripp-china} reported $f_c \sim 0.5$ for $W_{r}(2796) \geq 0.15$~{\AA} for $\sim 20$ galaxies with $D \leq 50$ kpc. These results are also consistent with the findings of \citet{cwc-china} who reported very weak {\MgII} absorption, $W_{r}(2796) < 0.3$~{\AA}, well inside the $R(L)$ boundary of bright galaxies; these galaxies would be classified as ``non--absorbers'' in previous surveys. They also report $W_r(2796) > 1$~{\AA} absorption out to $\simeq 2 R(L)$. All these results suggest that there are departures from the standard model, that the covering fraction of {\MgII} absorbing gas is less than unity, and that the halo sizes and the distribution of the gas appear to diverge from the $R(L)$ relation with spherical geometry. Another approach to understanding halo sizes and gas distributions is to determine the statistical properties of {\MgII} absorbing gas and then compute the statistical cross section from the redshift path density, $dN/dz$ \citep[see][]{lanzetta95}. The downfall of this method is that a galaxy luminosity function must be adopted in order to estimate $R_{\ast}$. \citet{nestor05} acquired a sample of over 1300 {\MgII} absorption systems, with $W_{r}(2796) \geq 0.3$~{\AA} from the Sloan Digital Sky Survey (SDSS). Using the $K$--band Holmberg--like luminosity scaling and luminosity function of MUNICS \citep{drory03}, Nestor {\etal} computed $R_{\ast} = 60-100$~kpc for adopted minimum luminosity cutoffs of $L_{min}=0.001-0.25L^{\ast}$. They found no redshift evolution of $R_{\ast}$ over the explored range of $0.3\leq z \leq 1.2$. \citet{zibetti06} studied the statistical photometric properties of $\sim2800$ {\MgII} absorbers in quasar fields imaged with SDSS. Using the method of image stacking, they detected low--level surface brightness (SB) azimuthally about the quasar. The SB profiles follow a decreasing power law with projected distance away from the quasar out to $100-200$~kpc. These results imply that absorption selected galaxies may reside out to projected distances of 200~kpc. However, it is worth noting that the extended light profiles may be an artifact of clustering of galaxies. Cluster companions of the {\MgII} absorbing galaxies could extend the observed light profile over hundreds of stacked images. Thus, one would infer that {\MgII} absorbing galaxies are present at a larger impact parameters than would be found in direct observation of individual galaxies. Motivated by recent expectations from simulations that halo gas is dynamically complex and sensitive to the physics of galaxy formation, we investigate the standard halo model of {\MgII} absorbers. We also aim to provide updated constraints on $f_c$ and $\beta$ for galaxy formation simulations. In this paper, we demonstrate that $f_c < 1$ and question the validity of the Holmberg--like luminosity scaling (Eq.~\ref{eq:rl}). Using high resolution quasar spectra, we explore {\MgII} absorption strengths to an order of magnitude more sensitive than previous surveys which allow us to re--identify non--absorbing galaxies as ``weak'' absorbing galaxies. In \S~\ref{sec:data} we describe our sample and analysis. In \S~\ref{sec:results}, we present new calculations of the statistical absorber radius computed using the statistically measured absorption path density $dN/dz$ and the Schechter luminosity function. We then compare these values to the empirical results of S95 and to a sample of known {\MgII} absorption selected galaxies with measured luminosities and impact parameters. We also examine how individual halos behave with respect to the statistical halo. In \S~\ref{sec:dis}, we discuss the properties and distribution of gas in halos. Our concluding remarks are in \S~\ref{sec:conclusion}. | \label{sec:conclusion} In conclusion, the gas covering fraction must be less than unity since the observed impact parameter distribution of absorbing galaxies does not fall exclusively within the statistical absorber halo radius in the range of $43 \leq R_{\rm x}\leq 88$~kpc. The fact that some absorbing galaxies are found at $D>R_{\rm x}$ and some non--absorbing galaxies are found at $D<R_{\rm x}$ implies $f_c < 1$ and that the standard halo model cannot describe halos on a case by case basis. This highlights the power of using the statistics of absorption line surveys to constrain the properties of halos in relation to the measured distributions in absorption selected galaxy surveys. By quantifying how individual galaxy halos deviate from a ``standard'' halo, we have obtained an average gas covering fraction of $\left< f_c \right> \sim 0.5$. It is possible that $f_c$ exhibits both a radial and an equivalent width dependence, though we cannot address this with our sample. Values of $f_c$ are likely to depend on galaxy star formation rates, and galaxy--galaxy mergers and harassment histories; processes that give rise to patchy and geometrically asymmetric gas distributions. Alternatively, the absorption properties of intermediate redshift halos may be governed by the dark matter over density, $\Delta \rho /\rho$, and redshifts at which the galaxies formed \citep{cwc06}. Our results also show that, if $f_c<1$, the sizes of {\MgII} absorbing halos can still follow a Holmberg--like luminosity relation with $\beta$ in the range of $0.2-0.28$ \citep[S95;][]{gb97}, which corresponds to $R_{\ast}\sim 110$~kpc. If $\beta=0$ is assumed, then $f_c \leq 0.37$ for our sample to be consistent with no luminosity scaling. In semi--analytical models in which {\MgII} absorbing gas is infalling and is pressure confined within the cooling radius of hot halos \citep[e.g.,][]{mo96,burkert00,lin00,maller04}, a Holmberg--like luminosity relation in quasar absorption line systems naturally arises \citep{mo96}. However, these models have great difficulty explaining {\MgII} absorption at impact parameters greater than $\sim 70$~kpc. If on the other hand halo gas spatial distributions are governed by stochastic mechanical processes, as suggested by \citet{kacprzak07}, then there is no {\it a priori} reason to expect a clean halo--size luminosity scaling. It is likely that some combination of these scenarios contribute to the statistical values of $f_c$ and $\beta$. Thus, it is reasonable to suggest that {\MgII} halos sizes may not be strictly coupled to the host galaxy luminosity. Further work on the cross--correlations between absorbers and galaxies would provide better estimates of $f_c$ and $\beta$, two quantities that provide direct constraints of galaxy formation simulations. Also needed are additional constrains on the relative kinematics of the absorbing halo gas and galaxies \citep[e.g.,][]{s02,ellison03,kacprzak07b}. What is required is the development of techniques to quantitatively compare observational data with mock quasar absorption line analysis of simulated galaxy halos \citep{cwcaas06}. | 7 | 10 | 0710.5765 |
0710 | 0710.5179.txt | We propose a scenario for the formation of the Main Belt in which asteroids incorporated icy particles formed in the outer Solar Nebula. We calculate the composition of icy planetesimals formed beyond a heliocentric distance of 5 AU in the nebula by assuming that the abundances of all elements, in particular that of oxygen, are solar. As a result, we show that ices formed in the outer Solar Nebula are composed of a mix of clathrate hydrates, hydrates formed above 50 K and pure condensates produced at lower temperatures. We then consider the inward migration of solids initially produced in the outer Solar Nebula and show that a significant fraction may have drifted to the current position of the Main Belt without encountering temperature and pressure conditions high enough to vaporize the ices they contain. We propose that, through the detection and identification of initially buried ices revealed by recent impacts on the surfaces of asteroids, it could be possible to infer the thermodynamic conditions that were present within the Solar Nebula during the accretion of these bodies, and during the inward migration of icy planetesimals. We also investigate the potential influence that the incorporation of ices in asteroids may have on their porosities and densities. In particular, we show how the presence of ices reduces the value of the bulk density of a given body, and consequently modifies its macro-porosity from that which would be expected from a given taxonomic type. | In recent years, some objects within the Main Belt of asteroids have been found to display cometary characteristics (Hsieh \& Jewitt 2006). Objects such as 133P/Elst-Pizarro, P/2005 U1 and 118401 (1999 RE$_{70}$) occupy orbits that are entirely decoupled from Jupiter within the Main Belt, and are probably bodies that have undergone a recent collision, revealing previously buried volatile material, and leading to the observed dusty outgassing. In addition, present-day surface water ice and possible water sublimation have been reported on Ceres (Lebofsky et al. 1981; A'Hearn \& Feldman 1992; Vernazza et al. 2005). This is consistent with recent Hubble Space Telescope (HST) observations which suggest that Ceres' shape is the result of the dwarf planet consisting of a rocky core surrounded by an ice-rich mantle (Thomas et al. 2005) - an idea in agreement with several thermal evolution scenarios (McCord \& Sotin 2005) that suggest that the ice content of the asteroid is between 17\% and 27\%, by mass. These observations are supported by the evidence of hydrated minerals in meteorites which provide samples of rock from asteroids in the Main Belt. Most of these minerals formed as a result of water ice accreting with the chondritic meteorite parent bodies, melting, and driving aqueous alteration reactions (Clayton \& Mayeda 1996; Jewitt et al. 2007). It seems likely, then, that some objects in the asteroid belt have incorporated significant amounts of water ice (and possibly other volatiles) during their formation in the early stages of the solar System. These bodies would have incorporated icy particles\footnote{By icy particles is meant planetesimals composed of a mix of ices and rocks.} coming from the outer nebula that survived their inward drift due to gas-drag through the disk (Mousis \& Alibert 2005 -- hereafter MA05). The volatile fraction incorporated in this manner could vary depending on the inward flux of icy planetesimals from the external region and the heliocentric location of the asteroid, together with the density of the proto-solar nebula. Given that the current asteroid belt lies closer to the Sun than the ``snow-line'', postulated to lie at around 5 AU in the Solar Nebula, these results are a little unexpected. In this context, understanding how volatiles were incorporated into the asteroids is therefore important, not only for the study of the asteroids themselves, but also for our understanding of the processes by which the solar system came into being. To this end, MA05 studied the possibility of determining the nature and composition of the ices which were incorporated into Ceres. They used a time dependant model of the Solar Nebula and showed that icy particles of sizes between 0.1 and 10 metres could drift from heliocentric distances greater than 5 AU to the present location of Ceres without encountering temperatures or pressures high enough to vaporise the ices within. The authors then suggested that ices produced in the outer Solar Nebula were transported inwards to become incorporated in the solids which accreted to form Ceres. The present work aims to improve upon the calculation detailed in MA05, along with expanding the results to involve the entire asteroid belt, rather than just its largest member. In particular, MA05 postulated that all volatiles, except CO$_2$\footnote{CO$_2$ is the only major volatile species which does not form a clathrate hydrate in the Solar Nebula because it condenses as a pure ice prior to being trapped by water.}, were trapped by water in the form of hydrates or clathrate hydrates in the outer solar nebula. This assumption was supported by the work of Hersant et al. (2001) who estimated that Jupiter was formed at temperatures higher than $\sim$40 -- 50 K. The accretion of ices in the form of hydrates and clathrate hydrates was thus required during the formation of the planet in order to explain the volatile enrichments observed in its atmosphere\footnote{The abundances of volatile species in Jupiter's atmosphere have been measured using the mass spectrometer on board the {\it Galileo} probe. These measurements reveal that the giant planet's atmosphere is enriched by a factor of $\sim$3 in Ar, Kr, Xe, C, N, and S compared to the solar abundances (Owen et al. 1999).} (Gautier et al. 2001a,b). Indeed, since these ices are usually formed at temperatures higher than that reached by the nebula at the time of Jupiter's completion, as defined by Hersant et al. (2001), they can be incorporated in the planetesimals accreted by the giant planet during its growth. However, the amount of water that would be required in the nebula to trap all these volatiles as hydrates and clathrate hydrates exceeds that derived from the solar oxygen abundance. Therefore, MA05 made the {\it ad hoc} hypothesis that oxygen was ``oversolar'' in the gas-phase in order to provide enough available water in the Solar Nebula\footnote{ The oxygen abundance required for to allow the trapping of all volatile species in the form of hydrates or clathrate hydrates is $\sim$1.9 times the solar abundance, with CO$_2$:CO:CH$_4$ = 1:1:1 and N$_2$:NH$_3$ = 1:1 (the nominal nebula gas phase ratios used in this work).}. Additionally, Hersant et al. (2001) only used an evolutionary Solar Nebula model to derive the disk's temperature at the time when the mass of Jupiter's feeding zone was equal to that of the gas in its current envelope. They thus neglected many important effects such as the influence of protoplanet formation on the structure of the disk (e.g. Fig. 2 of Alibert et al. 2004). However, recent giant planet core-accretion formation models that include migration, disk evolution, such as that proposed by Alibert et al. (2004), have shown that the disk's temperature can be as low as $\sim$10 -- 20 K at the end of Jovian formation. This implies that Jupiter itself can accrete ices during its formation that were produced at temperatures lower than those required for clathration. As a result, no extra water is required in the nebula to allow all the volatile species to be trapped in clathrate hydrates, and the oversolar oxygen abundance condition in the nebula can be relaxed. In Section 2, we calculate the composition of ices produced in the outer Solar Nebula under the assumption that the abundances of all elements, in particular that of oxygen, are solar. In Section 3, we consider the inward migration of particles produced at various locations in the nebula, and at different times. This allows us to examine whether some planetesimals formed in the outer Solar Nebula may have drifted to the current position of the Main Belt without encountering temperature and pressure conditions high enough to vaporize the ices they contain. In Section 4, we examine the uncertainties in the determination of the physical properties of asteroids. We also investigate the potential influence that the incorporation of ices in these objects may have on their porosities and densities. Section 5 is devoted to summary and discussion. | In order to explain the presence of hydratation and cometary features in the Main Belt, we have proposed that asteroids incorporated during their formation icy particles formed in the outer Solar Nebula. We have then calculated the composition of the ices trapped in these planetesimals formed beyond a heliocentric distance of 5 AU in the nebula, in a manner consistent with the formation of Jupiter, by assuming that the gas-phase abundances of all elements, in particular that of oxygen, are solar. As a result, we have found that the ices being formed in the outer Solar Nebula are composed of a mix of clathrate hydrates, hydrates formed above 50 K, and pure condensates produced at temperatures between $\sim$20 K and $\sim$50 K. We have noted that, whatever the input parameters adopted in the modelling of the disk, or the formation location considered for icy planetesimals at heliocentric distances beyond 5 AU, their composition remains almost constant, provided that the gas-phase abundances are homogeneous in the nebula. We have argued in this work that gas-drag is responsible for the inward drift of icy particles formed in the outer nebula towards the forming Main Belt. To support this hypothesis, we have showed that, at some epochs of the disk's evolution, some particles produced in the outer nebula may drift to the current position of the Main Belt without encountering temperature and pressure conditions high enough to vaporize the ices they contain. The current distribution of ices potentially existing in asteroids has probably been deeply altered after their formation. The effect of solar insolation may have vaporized the ice within nearer asteroids (semi-major axes of $\sim$2 AU), melted the ice of mid-range asteroids situated at $\sim$3 AU, but should not have affected the ice in asteroids located at greater heliocentric distances. Inner and outer asteroids would therefore display no detectable hydratation features, either because the ice was vaporized and dissipated, or because the ice never melted and thus did not react with the surface minerals to a sufficient extent as to allow detection (Cyr et al. 1998). In this context, we have proposed that, from the detection and identification of initially buried ices revealed by recent impacts on the surfaces of asteroids, it could be possible to infer the thermodynamic conditions that occurred within the Solar Nebula during the accretion of these bodies, as well as during the inwards migration of the icy planetesimals which they incorporated. However, this statement requires that either no parent body processing or modification took place during and after the formation of asteroids. For example, we have noted that subsequent alteration of the volatile phases in asteroids may occur due to catalytic reactions in their interiors. We have also investigated the potential influence that the incorporation of ices in asteroids may have on their porosities and densities. In particular, we have showed that the presence of ices can considerably reduce the value of the bulk density of the body, and consequently its macro-porosity, that would be expected from a given taxonomic type. That volatiles were delivered to areas within the ice line is clearly beyond doubt. In addition to the gas-drag mechanism described in this work, it is also likely that a significant amount of volatile material was dynamically driven inwards in the latter stages of planet formation. We still see the tail of this dynamical, chaotic volatile movement today -- the comets we observe passing through the inner solar system are the bearers of ices formed far beyond the snow line, and held in deep freeze since the early days. During the latter stages of planetary migration, the flux of such objects passing through the inner solar system, and hence encountering the asteroids, was significantly higher. Of particular interest, when one considers veneers of volatile material near the surface of the asteroids, is the Late Heavy Bombardment. In the Nice model, (see e.g. Gomes et al, 2005), vast amounts of volatile-rich material is flung inwards from the outer solar system approximately 700 Million years after its birth. This event, caused by the resonant destabilisation of the outer solar system, would have coincided with a simultaneous stirring of the asteroid belt, leading to an impact flux upon the Earth containing approximately even proportions of asteroidal and cometary material. It is clear, though, that the Earth would not be the only object to encounter volatiles injected in this way, and the possibility of a late-veneer of ice arriving in the asteroid belt is surely something which must be acknowledged in future work. In addition to this aggressive and chaotic injection of material, there is also aÞ gentler mechanism by which volatiles can be driven inwards as a result of planetary migration. As planets migrate, material can be trapped in the locations of mean-motion resonances (MMR), which sweep in front of the planet through it's motion. Evidence of material being swept outwards in the resonances of Neptune is clear for all to see -- the Plutino family of objects are locked in the 2:3 MMR with the planet, and have an inclination distribution which can tell us a great deal about the distance over which the planet migrated, sweeping them along. Inward migration can have the same effect -- the interior resonances of a planet can collect material as it moves inwards, and sweep it along -- giving a mechanism by which volatile material can be eased inwards, with the migration of a giant such as Jupiter. Work such as Fogg \& Nelson (2006) has shown that such resonant forcing can operate with a resonable efficiency, even for significantly faster migration than expected in our solar system, and so the effects of this behaviour should not be ignored in future work. In spite of the growing pool of evidence pointing towards the existence of water ice in the Main Belt, its detection on asteroids is a challenging observational problem. Large bodies such as Ceres are suspected to have retained a large amount of water since their formation, perhaps even including an internal liquid ocean, throughout the age of the solar system (McCord \& Sotin 2005). This could particularly be the case if this internal water was originally mixed with some ammonia, in agreement with our composition calculations in section \ref{icy}, which would have the effect of lowering the melting point of the water (ammonium bearing minerals have been suggested by King et al. (1992) as an alternate explanation for the origin of the 2.07 $\mu$m band seen in the spectrum of Ceres). Nevertheless, internal water can only be indirectly probed, either by measuring the hydrostatic shape of the object, as was done in the case of Ceres, or by inferring its density from its size and mass, when they are known or, more evidently, from outgasing activities. The case of Ceres is particularly interesting since, in spite of several possible pieces of evidence which support it being a highly hydrated body, the only report of water detection on the dwarf planet is the observation of OH escaping from its northern pole\footnote{An OH atmosphere was indeed observed around Ceres after perihelion by A'ÕHearn and Feldman (1992) by performing IUE long exposure spectra, with column densities of the order of $10^{11}$~cm$^{-2}$.}, is still not confirmed. Nevertheless, this detection could be explained in the context of the accumulation of ice during winter on the surface or within the subsurface layer, which would then dissipate during summer, when the surface temperature rises. Similar transient events have been suggested as possible mechanisms to trigger the geyser-like activity taking place near the south pole of Enceladus and reported by Cassini (Porco \& Team 2006). It is interesting to note that, considering the gravity and the day-side temperature of Ceres, any outgassed atmosphere would be rapidly lost. The mean thermal velocity $v_0$ of H$_2$O, for instance, would be close to the escape velocity ($v_{\infty}=516$~km~s$^{-1}$). Assuming a subsolar temperature of 215~K (Dotto et al. 2000), $v_0$ would vary between 450 to 350~km~s$^{-1}$ from the subsolar point to a zenith angle of 80$^{\circ}$. As a consequence, hydrodynamical escape would occur ($v_{esc}^{2}/v_{0}^{2} \le 2$). The photolysis of H$_2$O by solar EUV makes this atmospheric escape even more efficient by giving the photodissociation products OH and H some additional kinetic energy. Considering the short lifetime of H$_2$O at $\sim$3~AU ($<$9~days), and the fact that the mean thermal velocity of H atoms exceeds $v_{\infty}$, a tenuous atmosphere of OH is expected if water is outgassed by the asteroid at a sufficient rate. Due to the transcient nature of the atmosphere, the loss of water to space is limited by the flux of water from the interior to the surface. At low latitude, where ice is not stable, the continuous flux of water from the interior to space is too low to be detected. Only an accumulation of water ice at high latitude before perihelion, followed by an outgassing of H$_2$O associated with post-perihelion warming seems to result in an observable column density of OH. These results were found to be consistent with an earlier work done by Fanale and Salvail (1989), who estimated the mean loss rate of H$_2$O to be in the range 30-300~g~s$^{-1}$. Even if one assumes that the atmospheric loss observed by AÕ'Hearn and Feldman (1992) occurs continuously at the same rate and at all latitudes (which is obviously wrong as this maximum loss requires high latitudes and post-perihelion conditions), the water loss remains below 4 kg~s$^{-1}$, which, integrated over 4.5~Gyr, corresponds to only 0.07\% of the mass of Ceres. If the loss rate of H$_2$O in Ceres remained constant throughout its thermal history, the initial water reservoir is thus likely to be integraly preserved. Moreover, since the other volatile species are expected to be trapped as hydrates, clathrate hydrates and pure condensates in this reservoir, we can conclude that they have also been preserved from outgassing throughout the thermal history of the asteroid. Ceres being the largest and, due to its size, probably the wettest Main Belt asteroid, it is an ideal target for carrying out observations aiming at constraining its water regime. The experiment searching for water being vaporised within the polar regions of Ceres should be repeated with the state-of-the-art instrumentation available today on large telescopes. Such a detection would confirm unambiguously the presence of a large amount of water near the surface of Ceres. Direct observation of water ice, or of the effects of hydration, on the surface of Ceres can also be attempted for lower latitudes on the asteroid using a combination of high-angular resolution and spectroscopic instruments permitting the full resolution of its surface to the ~30-40 km level. Due to its low spectral resolution, imaging of the surface of Ceres, even when it is spatially resolved using HST or adaptive optics, is not sensitive to the presence of ice, while the detection of such is within the reach of low resolution spectroscopic observations (e.g. the detection of absorption features in the 1.0-3.5 $\mu$m region). A spatially resolved spectroscopic mapping of the surface of Ceres in the near-infrared can be done with today's ground-based telescopes and would permit the mapping of the strength of the 3 $\mu$m band, and allow the search for regions on the surface where interstitial water ice, or hydration features could be present, for instance at the location of cracks within the surface of Ceres, or the locations of deep impact craters. Indeed, recent HST (Thomas et al. 2005) and adaptive optics (Carry et al. 2007) imaging observations of Ceres revealed the presence of large impact craters across its surface which have likely disrupted the outer crust of the asteroid enough to directly expose the sub-surface mantle of wetter material. Finally, a spectroscopic study of the surface of Ceres, in order to search for the spectral signature of water and maybe those of other volatiles, should not be limited to one wavelength region (although the near-infrared range offers many diagnostic bands) but should, instead, encompass a wider range, from the near-UV to infrared wavelengths, in order to improve the identification of the chemicals species responsible for these spectral features. Finally, the NASA Discovery mission {\it Dawn}, which has been launched in September 2007 and whose arrival at Ceres is scheduled for 2015, will certainly bring new constraints on the presence of volatiles in the Main Belt. In particular, the {\it Dawn} mapping spectrometer (MS) covers the spectral range from the near UV (0.25 \micron) through the near IR (5 \micron) and has moderate to high spectral resolution and imaging capabilities (Russell et al. 2004). These characteristics make it an appropriate instrument for determining the asteroid's global surface composition. Near-infrared mapping of the surface of Ceres at small spatial scales will be very sensitive to volatile concentrations and may reveal ice spots on fresh impact-crater ridges. Moreover, the gravitation investigation of Ceres will allow the determination of its gravity field up to the 12th harmonic degree (Russell et al. 2004). Such a measurement will enable the shape and gravity models to characterize crustal and mantle density variations and, consequently, the amount of volatiles trapped therein. | 7 | 10 | 0710.5179 |
0710 | 0710.4280_arXiv.txt | Long Gamma Ray Bursts (GRBs) constitute an important tool to study the Universe near and beyond the epoch of reionization. We delineate here the characteristics of an 'ideal' instrument for the search of GRBs at $z\ge 6-10$. We find that the detection of these objects requires soft band detectors with a high sensitivity and moderately large FOV. In the light of these results, we compare available and planned GRB missions, deriving conservative predictions on the number of high-$z$ GRBs detectable by these instruments along with the maximum accessible redshift. We show that the {\it Swift} satellite will be able to detect various GRBs at $z\ge 6$, and likely at $z\ge 10$ if the trigger threshold is decreased by a factor of $\sim 2$. Furthermore, we find that INTEGRAL and GLAST are not the best tool to detect bursts at $z\ge 6$: the former being limited by the small FOV, and the latter by its hard energy band and relatively low sensitivity. Finally, future missions (SVOM, EDGE, but in particular EXIST) will provide a good sample of GRBs at $z\ge 6$ in a few years of operation. | The study of the Universe at the epoch of reionization is one of the main goal of available and future space missions. In the last few years, our knowledge of the early Universe has been enormously increased mainly owing to the observation of Quasars by the SDSS survey (Fan 2006). Long gamma ray bursts (GRB) may constitute a complementary way to study the cosmos and the early evolution of stars avoiding the proximity effects and possibly probing even larger redshifts up to $z\sim 10$. The five GRBs detected at $z\gsim 5$, over a sample of about 200 objects observed with the {\it Swift} satellite (Gehrels et al. 2004), show that a large percentage of GRBs is detected at high-$z$. The current record holder is $z = 6.29$ (Tagliaferri et al. 2005, Kawai et al. 2006). The identification of a large number of GRBs at $z\ge 6$ will open a new window in the study of the early Universe. Just to give some example, GRBs can be used to constrain the reionization history (Totani et al. 2006, Gallerani et al. 2007), to study the metallicity and dust content of normal galaxies at high-$z$ (Campana et al. 2007), to probe the small-scale power spectrum of density fluctuations (Mesinger, Perna \& Haiman 2005). Moreover, available and future GRB missions might be the first observatories to detect individual Population III stars, provided that massive metal-free stars were able to trigger GRBs (see Bromm \& Loeb 2007 for a review). Finally, the study of GRBs at high redshift is interesting by itself. In particular, thanks to cosmological time dilation, the study of the early phases of the afterglow is easier and can provide fundamental highlight on the central engine and burst physics. In this paper, we delineate the main characteristics of an 'ideal' instrument for the search of GRBs at $z\ge 6-10$. In particular, we explore different observational bands, deriving the best combination of sensitivity and field of view (FOV) in order to detect bursts at $z\gsim 10$. In the light of these results, we compare available and planned X-- and Gamma--ray missions, deriving conservative predictions on the number of high-$z$ GRBs detectable by these instruments along with the maximum accessible redshift. The paper is organized as follows. In Sect.~2 we briefly describe the different models here adopted. In Sect.~3, we derive the main characteristic of an 'ideal' instrument for exploring the high-$z$ GRB population, whereas predictions for available (planned) GRB missions are given in Sect.~4 (Sect.~5). Finally, we summarize our results in Sect.~6. | We have explored the characteristics of an ideal mission for the search of GRBs near and beyond the epoch of reionization, i.e. $z> 6-10$. In particular, we considered different observational bands, deriving the best combination of sensitivity and field of view in order to detect bursts at $z\gsim 10$. We found that such an experiment requires soft band detectors and high sensitivity, whereas large FOV or wide energy coverage are less important. Assuming 3 sr FOV, an observational band of 8-200 keV, and a sensitivity as low as 0.1 ph s$^{-1}$ cm$^{-2}$, this instrument would be able to detect $\sim 40$ (3) GRBs at $z\ge 6$ ($z\ge 10$) in one year of mission. In the light of these results, we compared available and planned GRB missions, deriving conservative predictions on the observable number of GRBs at $z\ge 6$ and $z\ge 10$ along with the maximum accessible redshift. We have shown that {\it Swift} is a viable tool to detect GRBs at $z\sim 6$. At the actual trigger threshold, $1.3-4$ GRBs per year should be identified above this redshift. We discuss also the possibility of increasing the number of high-$z$ detections by lowering the {\it Swift} trigger threshold. We found that the number of detectable GRBs doubles by lowering this by about 50\%. Assuming to be able to further lowering the trigger threshold down to 0.1 ph s$^{-1}$ cm$^{-2}$, {\it Swift} should detect $\sim 1$ GRB per year at $z\ge 10$. The INTEGRAL and GLAST satellites do not appear the best tools to search for GRBs at very high redshift. The former is limited by the very small FOV whereas the GLAST hard energy observational band and relatively low sensitivity do not allow to detect more than 1 GRB per year at $z\ge 6$. No GRB at $z\ge 10$ is expected during the entire mission of both instruments. Finally, we show that future missions, like SVOM, EDGE, and in particular EXIST, will be able to collect a good number of GRBs at $z\ge 6$ in a few years of operations. This sample can be use to study the early Universe, possible providing strong constrain on the reionization process (Gallerani et al. 2007), and deriving estimate on the star formation and metallicity/dust content in normal high-$z$ galaxies. | 7 | 10 | 0710.4280 |
0710 | 0710.5529_arXiv.txt | The strong spectral features near 2.2 $\mu$m in early-type galaxies remain relatively unexplored. Yet, they open a tightly focused window on the coolest giant stars in these galaxies -- a window that can be used to explore both age and metallicity effects. Here, new measurements of K-band spectral features are presented for eleven early-type galaxies in the nearby Fornax galaxy cluster. Based on these measurements, the following conclusions have been reached: (1) in galaxies with no signatures of a young stellar component, the K-band \naK\ index is highly correlated with both the optical metallicity indicator \mgfe\/ and the central velocity dispersion $\sigma$; (2) in the same galaxies, the K-band Fe features saturate in galaxies with $\sigma > 150$ \kms\/ while \naK\/ (and \mgfe) continues to increase; (3) [Si/Fe] (and possibly [Na/Fe]) is larger in all observed Fornax galaxies than in Galactic open clusters with near-solar metallicity; (4) in various near-IR diagnostic diagrams, galaxies with signatures of a young stellar component (strong \hb, weak \mgfe) are clearly separated from galaxies with purely old stellar populations; furthermore, this separation is consistent with the presence of an increased number of M-giant stars (most likely to be thermally pulsating AGB stars); (5) the near-IR \naK\/ vs.~$\sigma$ or \fe\/ vs.~$\sigma$ diagrams discussed here seem as efficient for detecting putatively young stellar components in early-type galaxies as the more commonly used age/metallicity diagnostic plots using optical indices (e.g H$\beta$ vs.~\mgfe). The combination of these spectral indices near 2.2 $\mu$m with high {\it spatial} resolution spectroscopy from ground-based or space-based observatories promises to provide new insights into the nature of stellar populations in the central regions of distant early-type galaxies. | \label{sec:intro} Understanding the stellar content of early-type galaxies is fundamental to understanding their star formation and chemical evolution history. Most early-type galaxies are too distant to resolve their individual stars with current technology, rendering the direct study of their stellar populations impossible. Thus, their stellar populations must be studied using indirect methods. In recent decades, significant effort has gone into trying to better constrain the stellar contents for early-type galaxies using optical spectroscopic data. The most commonly studied features have been Ca I H and K 0.38 $\mu$m, H$\beta$, Mgb 0.52 $\mu$m, Fe $\mu$m 0.53, Na 0.82 $\mu$m, and CaT 0.86 $\mu$m. Interpretation of all such spectral features is intrinsically complicated by their blended nature -- each feature is really the super-position of many spectral lines, usually from several different elements, blurred together by the line-of-sight velocity dispersion within each galaxy. There is no way to overcome this problem -- it must simply be taken into account during analysis. As population synthesis models have become more sophisticated and digital stellar libraries more complete, this problem has become more tractable over time. Another challenge arises from the composite nature of galaxies: each observed feature is the luminosity-weighted integrated sum of that feature from all stars in the observed line-of-sight. Naturally, luminosity-weighted does not imply mass-weighted. A relatively small fraction of the mass can dominate the observed luminosity and mask the underlying stellar population (e.g. as happens during a starburst event within a pre-existing galaxy). Even in relatively quiescent galaxies, light from stars at several important evolutionary stages contribute roughly equally to the observed spectral features between 0.4 -- 1 $\mu$m range. Hence, a feature depth change could be due to (e.g.) a change near the (mostly) age-driven main-sequence turnoff or the (mostly) metallicity-driven red giant branch. The details can become quite complicated, as illustrated by the long standing controversy about whether observed changes in Balmer line strength arise from the presence of younger main sequence stars, more metal-poor main sequence stars, or an extended horizontal giant branch (for recent discussions of this debate, see Maraston \& Thomas 2000 and Trager et al. 2005). A similar controversy surrounds Na 0.82 $\mu$m feature: is it driven by metallicity-driven red giant branch changes, initial mass function related differences in the relative number of cool dwarf and giant stars or both \citep[e.g.][]{car86, all89, del92}? However, the properties of the RGB component can be isolated by observing in the K-band (centered near 2.2 $\mu$m). At those wavelengths, cool giants near the tip of the first-ascent red giant branch (RGB) dominate the integrated light in old ($\geq$ 3 Gyr) stellar populations. In combination with optical observations, K-band observations should facilitate the separation of MSTO and RGB light contributions. There are two possible complications to this scenario. First, a very young stellar population containing red supergiants will contribute a significant fraction K-band light. Fortunately, such a population is obvious from the presence of H~II region emission lines at shorter wavelengths. Second, a somewhat older population (1 -- 2 Gyr, i.e. an intermediate-age population) may contain bolometrically bright carbon stars that can contribute a detectable amount of K-band light (see discussions in Silva \& Bothun 1998a,b). Such a population may or may not be connected to increased H$\beta$ strength. Initial development of these ideas can be found in Silva et al. (1994), Mobasher \& James (1996), James \& Mobasher (1999), Mobasher \& James (2000), all of whom focused on the CO 2.36 $\mu$m feature. Origlia et al. (1997) observed a Si dominated feature at 1.59 $\mu$m as well as CO dominated features. These observational studies were limited by small-format detectors to relatively low resolving powers and/or small wavelength ranges per observation. In the cases of Silva et al. and Origlia et al., only small, heterogeneous samples of galaxies were observed. A general conclusion of the James \& Mobasher studies was that changes in CO strength between early-type galaxies in high-density and low-density regions were statistically consistent with different fraction contributions of intermediate-age AGB light and hence galaxies in low-density regions had younger luminosity-weighted mean ages. Origilia et al. argued that [Si/Fe] was super-solar in the four elliptical galaxies they observed. To further develop these ideas and investigate the usefulness of other K-band spectral indices in the study of early-type galaxies, new data have been obtained for eleven E/S0 galaxies in the nearby Fornax cluster. Only measurements in the central regions of these galaxies are discussed here. In Section~\ref{sec:data}, the galaxy sample and its observations are discussed, while in Section~\ref{sec:proc} the data processing methodology is described. The measurement of spectral feature strength is explained in Section~\ref{sec:lines} while basic observation results are presented in Section~\ref{sec:results}. The broader astrophysical implications of our observational results are discussed in Section~\ref{sec:disc}. A summary is provided at the end. | \label{sec:disc} No self-consistent theoretical spectral synthesis models for the interpretation of integrated K-band spectra of early-type galaxies are widely available. Development of such models is on-going (e.g. Marmol Queralt, 2007, in preparation). Until such models exist, we must rely on basic (and perhaps imperfect) astrophysical intuition and available tools, such as the high-resolution near-IR spectral atlas of Wallace \& Hinkle (1996) (hereafter WK96). \subsection{Galaxies with young populations} Three galaxies in our sample (NGC 1316, NGC 1344, and NGC 1375) have stronger \hb\/ and weaker \mgfe\/ features than the rest of our sample (see Figure~\ref{fig:fornax_pops}). This suggests a simple, two component stellar population model. One stellar component is cold -- in the luminosity-weighted mean, it is presumably old ($>$ 8 Gyr) and metal-rich ([Fe/H] $>$ --0.3) with spectral properties consistent with the observed central velocity dispersion. In other words, it has properties similar to the purely old galaxies in our sample (see Figure~\ref{fig:fornax_pops}). The other component is warm -- it contains a significant number of A/F dwarf or sub-giant stars. This component could have several origins but only two are considered here. Each has different consequences for K-band spectral index behavior. First, the warm component could be associated with a young population with a warm ($\sim$2 \msun) main sequence turnoff (MSTO). The MSTO stars are tied to thermally pulsating asymptotic giant branch (TP-AGB) stars with bolometric magnitudes that place them above the tip of the first-ascent red giant branch (TRGB). The TP-AGB stars are cooler than stars on the first-ascent RGB and hence have an M spectral type. In the underlying luminosity function (spectral type vs. number of stars), the M-star bins are relatively more populated than in a purely old galaxy. The net effect is that integrated 2.2 $\mu$m spectra should become more M-like. Second, the warm component could be associated with a metal-poor population with a warm MSTO. These MSTO stars are tied to RGB stars with similar luminosity as the corresponding metal-rich RGB stars but with warmer \teff. In the underlying luminosity function, the K-star bins will be relatively more populated than in a purely old population. The net effect is to make the integrated 2.2 $\mu$m spectra more K-like. \begin{figure} \resizebox{\hsize}{!}{\includegraphics[angle=0]{f16.ps}} \caption{\label{fig:galModel} Optical-NIR corrections for young populations. See Figure~\ref{fig:galLineSigma} for a symbol explanation. The observed index strengths for the galaxies with young components are connected to the predicted index strengths ({\it solid squares}) after the effects of a young component has been removed. In the lower panel, the {\it slanted dashed line} indicates the unweighted linear regression fit for the purely old galaxies with $\sigma >$ 150 \kms. In the middle panels, the {\it horizontal dashed line} indicates the unweighted mean value for purely old galaxies with $\sigma >$ 150 \kms.} \end{figure} Can the observations presented here distinguish between these two scenarios? Consider Figure~\ref{fig:galModel} -- the galaxies with relatively strong \hb\/ and weak \mgfe\/ have relatively stronger (i.e. more M-like) near-IR features. Qualitatively, this is consistent with the first scenario. Can this conclusion be better quantified? What is the relative ratio of young to old stellar mass? Are the changes in optical and near-IR spectral features {\it quantitatively} consistent? To begin to answer these questions, the Thomas et al. (2003) models were used to create very simple two-component models that produced optical index values that matched the observed values in the three galaxies with strong \hb. One component was old (11 Gyr) with metallicity set to [Z/H] $=$ 0.35 for NGC 1316 and NGC 1344 and 0.0 for NGC 1375. The other component had the same metallicity but a young (1 Gyr) age. The relative mass fractions of these components were varied until the model \hb\/ strength matched the observed \hb\/ strength. The resultant young mass fractions (where total mass $=$ 1) were 0.135, 0.075, and 0.140 for NGC 1316, NGC 1344, and NGC 1375, respectively. The implied differential correction between the purely old model and the two-component model was then applied to the observed data (see Figure~\ref{fig:galModel}). Next, observed \fe\/ strength was adjusted manually until all three galaxies lay within the locus of purely old galaxies with similar $\sigma$ in the \fe\/ vs. \naK\/ panel. As part of this adjustment, $\Delta$(\fe)/$\Delta$(\naK) was forced to agree with the value (0.62) determined for the observed Galactic cluster giant stars (see Figure~\ref{fig:starLineLine}). This is a trend in effective temperature that is relatively insensitive to metallicity. Cooler giant stars (more late K and early M like) have strong K-band spectral features. In addition to the $\Delta$(\fe)/$\Delta$(\naK) measured from the Galactic open cluster stars, two other key numerical trends can be computed: \begin{eqnarray} \Delta(\fe)/\Delta(\naK) & = & 0.62 \\ \Delta(\naK)/\Delta(\mgfe) & = & 0.70 (1.10) \\ \Delta(\fe)/\Delta(\mgfe) & = & 0.28 (0.44) \end{eqnarray} NGC 1316 and NGC 1344 could be forced to have consistent trajectories. However, NGC 1375 (values shown in parenthesis) appears to follow somewhat different trajectories. Obviously, this kind of cartoon model is illustrative only and surely hides a plethora of details. Indeed, the real trajectories are unlikely to be linear. Nevertheless, a clear astrophysical conclusion emerges: for the \hb-strong galaxies, the observed optical and near-IR features all change in concert, consistent with a warm MSTO tied to an extended RGB, which is in turn consistent with the presence of a young stellar component, not a metal-poor stellar component. \subsection{Galaxies with bright K-band SBF} Liu, Graham, \& Charlot (2002) found that relative I-band and K-band surface brightness fluctuation strength (expressed as an SBF color, $\bar{I} - \bar{K}$) was essentially constant for the majority of the early-type galaxies they studied in Fornax. However, two galaxies in common with our sample (NGC 1419 and 1427) were found to have brighter (larger) K-band SBF relative to the mean relationship (see also Mei et al. 2001). A larger $\bar{K}$ is thought theoretically to arise from the presence of an extended giant branch, i.e. more cool, bright, M giants. In the spectroscopic data discussed here, there are no indications of such extended giant branches in these galaxies -- their central line indices are consistent with an integrated stellar populations dominated by an old, metal-rich population. Of course, the SBF measurements were made in the outer regions of these galaxies, as opposed to the optical and near-IR observations of the central regions discussed here. It may be that the integrated luminosity of such a component (if present at all) is not high enough in the central region for detection by our method. \subsection{Purely old galaxies} In purely old galaxies, we have seen that: \begin{itemize} \item{} \naK\ is stronger than in Galactic open clusters and is highly correlated with $\sigma$ and \mgfe. \item{} \fe\/ saturates for $\sigma \gtrsim 150$ \kms. \item{} \co\ is somewhat correlated with $\sigma$ and \mgfe, while \caK\/ is not (unless NGC 1399 is not included in the regression fits, see footnote 2 above). \end{itemize} The observed K-band spectral features are named for the dominant elemental species in the {\it solar} spectrum. However, at the effective temperatures of interest here, the on-band index definition and their companion off-band continuum bands contain lines from other elemental species as well (as Ramirez et al.~1997 discuss comprehensively). By referring to the high resolution spectra of WK96, the contribution from these other absorbers can be investigated and used to explain the overall spectral feature behavior. In turn, underlying astrophysical parameters related to galaxy formation and evolution are revealed. \subsubsection{\naK\/ index} In the current galaxy sample, \naK\/ is significantly stronger than observed in the Galactic open cluster stars. The off-bands defined for the \naK\/ feature are relatively line-free, so we can focus on the on-band spectral region (see the $\lambda$ Dra, M0 III spectrum in WK96, p. 352). In addition to sodium, Sc, Si, and (to a lesser extend) V are significant absorbers in this region. Origlia et al. (1997) argued that [Si/Fe] is super-solar in a small sample of early-type galaxies based in measurements of the Si I 1.59 $\mu$m feature. Trager et al. (2000) have argued that both [Na/Fe] and [Si/Fe] are super-solar in early-type galaxies based on models of optical spectral feature behavior. The K-band \naK\/ measurements presented here are consistent with enhanced silicon in the observed Fornax galaxies and hence indirectly confirm the conclusions of Origlia et al. and Trager et al. \subsubsection{\fe\/ index} The \fea\/ feature contains approximately equal absorption contributions from Fe, Sc and Ti while \feb\/ is dominated by Fe absorption features with some additional Sc absorption (in WK96, see the $\lambda$ Dra, M0 III spectra on p. 346 and 348). Hence, the \fe\/ index (a combination of \fea\/ and \feb) is dominated by absorption lines from Fe-peak elements). For purely old galaxies with $\sigma >$ 150 \kms, \fe\/ is observed to remain constant while \naK\/ (an index dominated by absoprtion lines from $\alpha$-elements) and \mgfe\/ (a total metallicity [Z/H] indicator) become stronger as $\sigma$ increases. It is tempting to conclude that above $\sigma >$ 150 \kms\/ luminosity-weighted mean total metallicity [Z/H] continues to increase as a function of central velocity dispersion, driven by a relative increase of [$\alpha$/H], while [Fe/H] remains constant. However, within the narrow [Fe/H] range of observed Galactic open cluster stars, \fe\/ is more strongly correlated with effective temperature than cluster [Fe/H] (see Figure~\ref{fig:starLineColor}). Increasing mean total metallicity [Z/H] in the galaxies should correspond to cooler mean RGB effective temperature and hence increased \fe. Recall that the relationship between \fe\ and mean RGB effective temperature has already been exploited above to explain the observed behavior of \fe\/ (and \naK) in galaxies with young stellar components. The dilemma is clear: how can a temperature-sensitive index like \fe\/ remain constant while mean RGB effective temperature decreases due to increasing [Z/H]? Without appropriate population synthesis models or stellar spectra, only a few speculative thoughts can be offered. In their detailed comparison of non-solar aboundance population synthesis models with measurements of spectral indices in the optical Lick system, Trager et al. (2000) concluded that their best-fit models included enhancements in C, N, Na, and Si (among others). As [$\alpha$/H] increases with $\sigma$ (as suggested by increasing \naK), stronger CN bands in the \fea\/ and \feb\/ on-band and off-band windows (see the WK96 spectra referenced above and the band definitions in Table~\ref{tab:features}) may have the effect of depressing the local continuum and hence decreasing the value of \fe\/ $=$ (\fea\/ $+$ \feb)/2. For example, Trager et al. noted a similar effect from C$_{\rm 2}$ bands in the optical Mg {\it b} index. Therefore, it may be that above some critical mean [$\alpha$/H] and below some critical mean effective temperature, \fe\/ stays within a narrow range and provides no useful information about mean [Z/H], [Fe/H], or effective temperature. A conclusive astrophysical explanation of \fe\/ in early-type galaxies with $\sigma >$ 150 \kms\/ awaits larger galaxy samples, observations of appropriate stars (e.g. in the Galactic bulge), and detailed population synthesis models. \subsubsection{\co\ index} The \co\/ feature is dominated by the $^{\rm 12}$CO(2,0) bandhead and has no significant absorption components from other elements. Like \fe\/ above, \co\/ is highly correlated with $T_{eff}$ in the observed Galactic open cluster stars and reaches larger values than observed in the galaxies. As a function of central velocity dispersion $\sigma$ in the observed galaxies, \co\/ does not appear to saturate (or at least not as definitively as \fe). This adds credence to the suspicion that \fe\ is not tracing mean RGB effective temperature in galaxies with $\sigma >$ 150 \kms. Based on the current measurements, no conclusions can be reached about relative carbon or oxygen abundance in these galaxies. \subsubsection{\caK\ index} No obvious explanation presents itself for the behavior of \caK\/ in this galaxy sample. The \caK\/ feature is intrinsically complex, consisting of contributions from many absorbers: Ca, S, Si, and Ti (all $\alpha$-elements) as well as Sc and Fe (see $\lambda$ Dra, MO III spectrum in WK96, p. 342). As a function of \teff\/ in the range of interest, some of the contributing absorption lines get stronger (Sc and Ti, the former faster than the latter), some features get weaker (Si), and some remain roughly constant (Ca, S, and Fe) (cf. Ramirez et al. 1997). Relative to solar abundance ratios, Trager et al. (2000) have argued that Si and S are over-abundant and Ca is under-abundant in early-type galaxies while [Ti/Fe] and [Sc/Fe] have their solar values. In the Galactic open cluster stars observed here, the \caK\/ feature is stronger at a given $J-K$ ($T_eff$) in the solar-metallicity stars than in the more metal-poor stars. Within the galaxies observed here, \caK\/ appears to have significant scatter at any given $\sigma$ or \mgfe. No obvious explanation for this scatter (or the global behavior of \caK\ with $\sigma$ and \mgfe) presents itself at this time. Using new, moderate resolution ($R \sim 2500$) K-band spectra, spectral indices have been measured in the central regions of eleven early-type galaxies in the nearby Fornax cluster. Based on these measurements, the following conclusions were reached: \begin{enumerate} \item{} The \naK\/ feature is much stronger in these Fornax early-type galaxies than observed in solar-metallicity Galactic open cluster stars. This is attributed to relative [Si/Fe] (and possible [Na/Fe]) differences between the open cluster stars and the Fornax galaxies, i.e. both are larger in the Fornax galaxies than in the cluster stars. \item{} In various near-IR diagnostic diagrams, galaxies with optical indices indicative of a warm stellar component are clearly separated from galaxies dominated by colder, presumably old ($\geq$ 8 Gyr) stellar populations. Changes in the near-IR spectra features are consistent with the presence of an cool component dominated by late K and/or early M giants stars. In combination, the optical and near-IR observations are consistent with the presence of a young stellar component with a warm MSTO and a significant extended giant branch consisting of TP-AGB stars. \item{} For detecting a young stellar component, the \naK\/ vs.~$\sigma$ or \fe\/ vs.~$\sigma$ diagnostic diagram seems as efficient as using H$\beta$ vs. \mgfe\/ (or other similar combinations of optical indices). The near-IR features have the additional advantage that no emission-line correction is needed (as it is necessary for the stronger optical Balmer lines). \item{} The \fe\/ index saturates in galaxies with central velocity dispersion $\sigma$ $>$ 150 \kms\/ dominated by old ($\geq$ 8 Gyr) stellar populations. For $\sigma >$ 150 \kms, these Fe features are unlikely to be useful for investigating stellar population differences between early-type galaxies. Above $\sigma >$ 150 \kms, the continued increase in \naK\ (and \mgfe) strength with $\sigma$ coupled with constant \fe\ presents an astrophysical challenge. Although it is tempting to conclude that [Fe/H] reaches a maximum value, while [$\alpha$/H] (and hence [Z/H]) continues to increase, more observational and population synthesis work is needed to understand conclusively the behavior of \fe\/ in these high-mass early-type galaxies. \end{enumerate} Adding these near-IR indices to the standard diagnostic toolkit for analyzing the integrated light of early-type galaxies clearly has great potential. To develop this potential, three obvious steps are needed: observe more galaxies over a larger range of central velocity dispersion (including field galaxies with existing optical data) and extend current population synthesis models to the near-IR. We know that various groups are working on both steps. It would also be useful to study the radial behavior of these indices within individual galaxies to compare and contrast index behavior between galaxies. For the foreseeable future, the study of the central populations in early-type galaxies will remain the study of integrated light. As distance increases, our ability to study the central regions of early-type galaxies within metric apertures equivalent to nearby galaxies relies on the high {\it spatial} resolution spectroscopy achievable from space or with adaptive optics on the ground. In the former case, the James Webb Space Telescope represents the frontier -- yet, no optical spectrograph that works below 0.8 $\mu$m is currently planned. In the latter case, the frontier will be shaped by further development of adaptive optics systems -- and these systems will achieve their best performance beyond 1 $\mu$m. Hence, understanding how to interpret near-IR spectral indices in nearby galaxies is key to facilitating the kind of investigation and characterization of more distant early-type galaxies that will not be possible at optical wavelengths. | 7 | 10 | 0710.5529 |
0710 | 0710.4059.txt | The turbulent magnetic diffusivity tensor is determined in the presence of rotation or shear. The question is addressed whether dynamo action from the shear--current effect can explain large-scale magnetic field generation found in simulations with shear. For this purpose a set of evolution equations for the response to imposed test fields is solved with turbulent and mean motions calculated from the momentum and continuity equations. The corresponding results for the electromotive force are used to calculate turbulent transport coefficients. The diagonal components of the turbulent magnetic diffusivity tensor are found to be very close together, but their values increase slightly with increasing shear and decrease with increasing rotation rate. In the presence of shear, the sign of the two off-diagonal components of the turbulent magnetic diffusion tensor is the same and opposite to the sign of the shear. This implies that dynamo action from the shear--current effect is impossible, except perhaps for high magnetic Reynolds numbers. However, even though there is no alpha effect on the average, the components of the $\alpha$ tensor display Gaussian fluctuations around zero. These fluctuations are strong enough to drive an incoherent alpha--shear dynamo. The incoherent shear--current effect, on the other hand, is found to be subdominant. | Many of the stellar and planetary magnetic fields are believed to be the result of a dynamo process that converts kinetic energy from turbulent motions and shear into magnetic energy. A particular challenge consists in explaining the field on length scales that exceed the scale of the turbulence. This topic has traditionally been addressed within the framework of mean--field electrodynamics (Krause \& R\"adler 1980). Over the decades the applicability of this theory has repeatedly been questioned (e.g., Piddington 1981, Vainshtein \& Cattaneo 1992). Meanwhile, direct simulations of hydromagnetic turbulence have begun to show dynamo action (Meneguzzi et al.\ 1981, Meneguzzi \& Pouquet 1989, Nordlund et al.\ 1992, Brandenburg et al.\ 1996, Cattaneo 1999). In some particular cases, large-scale fields are being generated (Glatzmaier \& Roberts 1995, Brandenburg et al.\ 1995, Brandenburg 2001) which raises the question about the mechanism responsible for this phenomenon. In cases where the flow is systematically non-mirror symmetric the association with an $\alpha$ effect is obvious. However, there are now also examples of nonhelical large-scale dynamos owing to turbulence under the influence of shear alone (Brandenburg 2005a, Yousef et al.\ 2007). Their interpretation is not straightforward, because several possible mechanisms have been proposed that might produce dynamo action from turbulence and shear alone, i.e.\ {\it without} rotation and stratification that otherwise would have been the main ingredients of an $\alpha$ effect. The most detailed investigations have been carried out in connection with the so-called shear--current effect (Rogachevskii \& Kleeorin 2003, 2004, R\"adler \& Stepanov 2006, R\"udiger \& Kitchatinov 2006). Another possibility is a magnetic $\alpha$ effect that is driven by a current helicity flux, as was suggested by Vishniac \& Cho (2001; see also Brandenburg \& Subramanian 2005c). A third possibility might be an incoherent (random) $\alpha$ effect with zero mean and finite variance, suggested by Vishniac \& Brandenburg (1997) in connection with accretion discs (see also Sokolov 1997, Silant'ev 2000, Fedotov et al.\ 2006, Proctor 2007). The only reliable way to determine what is the dominant effect is to calculate all relevant components of the $\alpha$ and turbulent magnetic diffusivity tensors in a general expansion of the electromotive force in terms of the mean magnetic field. The case considered in Brandenburg (2005a) is unnecessarily complicated because the shear employed there depends on two Cartesian coordinates. A simpler possibility is to consider a shear flow depending linearly on only one coordinate and we shall pursue this idea in the present paper. The shear--current effect and the incoherent $\alpha$ effect could then still operate. Because we will use periodic boundary conditions there can be no magnetic helicity flux, so the Vishniac \& Cho (2001) effect is then ruled out, even though it could still, at least in principle, explain the generation of a mean magnetic field in the simulations of Brandenburg (2005a), which do possess a helicity flux. In this paper we calculate all relevant components of $\alpha_{ij}$ and $\eta_{ijk}$ using the so-called test field method. This method was introduced by Schrinner et al.\ (2005, 2007) in connection with convection in a spherical shell and used later by Brandenburg (2005b), Sur et al.\ (2007) and Brandenburg et al.\ (2008) in connection with forced turbulence in Cartesian boxes. The essence of this method consists in solving evolution equations for the fluctuations of the magnetic field around suitably defined test fields such that all relevant coefficients can be computed. | The present work has demonstrated that the test field method provides a robust means of determining all components of the turbulent magnetic diffusivity tensor that are relevant for mean fields depending only on $z$ and $t$. Both rotating and weakly shearing turbulence are studied. In either case the diagonal components of the turbulent diffusivity tensor are about equal to each other. Shear slightly enhances the turbulent magnetic diffusivity while rotation quenches it. In the presence of rotation, the $\OO\times\meanJJ$ effect occurs, which is described by the off-diagonal components of the turbulent magnetic diffusivity tensor. Shear leads to the shear--current effect, again described by off-diagonal components of this tensor. In both cases the results are consistent with those found in the framework of the second-order correlation approximation. The possibility of the so-called shear-current dynamo has been scrutinized. It depends crucially on the sign of the component $\eta_{21}$ of the magnetic diffusivity tensor. It turns out that, within the ranges of parameters considered, its sign is in general not suited for driving a dynamo based on this effect, with a possible exception at large magnetic Reynolds numbers. In this way the analytic results found in the second--order correlation approximation for incompressible fluids (R\"udiger \& Kitchatinov 2006; R\"adler \& Stepanov 2006) are confirmed and generalized. Direct numerical simulations are presented which exhibit growing mean magnetic fields in shear flow turbulence. An interpretation as a (coherent) shear-current dynamo is hardly possible. Instead, it is argued that it can be explained by an incoherent alpha--shear dynamo. The incoherent shear--current effect has also been determined, but it is found to be less important. | 7 | 10 | 0710.4059 |
0710 | 0710.0760_arXiv.txt | { This series of papers is dedicated to a new technique to select galaxies that can act as standard rods and standard candles in order to perform geometrical tests on large samples of high redshift galaxies to constrain different cosmological parameter. The goals of this paper are (1) to compare different rotation indicators in order to understand the relation between rotation velocities extracted from observations of the H${\alpha} \lambda$6563\AA\ line and the [OII]$\lambda 3727$\AA\ line, and (2) to determine the scaling relations between physical size, surface brightness and magnitude of galaxies and their rotation velocity using the SFI++, a large catalog of nearby galaxies observed at I-band. A good correlation is observed between the rotation curve-derived velocities of the H${\alpha}$ and [OII] observations, as well as between those calculated from velocity histograms, justifying the direct comparison of velocities measured from H${\alpha}$ rotation curves in nearby galaxies and from [OII] line widths at higher redshifts. To provide calibration for the geometrical tests, we give expressions for the different scaling relations between properties of galaxies (size, surface brightness, magnitude) and their rotation speeds. Apart from the Tully-Fisher relation, we derive the size-rotation velocity and surface brightness-rotation velocity relations with unprecedentedly small scatters. We show how the best size-rotation velocity relation is derived when size is estimated not from disc scale lengths but from the isophotal diameter $r_{23.5}$, once these have been corrected for inclination and extinction effects.} \titlerunning{Geometrical Tests of Cosmological Models II.} \authorrunning{Saintonge et al.} | With the new generation of telescopes and instruments, such as multi-object spectrograph and large field of view cameras, it has become possible to perform large, deep redshift surveys. Using these data, the structure and geometry of the Universe can be studied through new channels, namely geometrical tests such as the angular diameter test (angular size - redshift relation), the Hubble diagram (magnitude - redshift relation) and the Hubble test (count - redshift relation). These tests are generally difficult, requiring differentiation between the natural evolution of galaxies and the effects of geometry. Large redshift surveys ease the task in that respect. To perform the geometrical tests mentioned above, a population of objects that can be tracked through redshift needs to be defined. More specifically, the angular diameter test requires a standard rod to be identified, the angular size of which is then being measured over the redshift range of interest. The objects taken to serve that purpose could be anything from galaxies \citep[e.g.][]{sandage72}, clusters of galaxies \citep[e.g.][]{sereno,cooray98}, or dark matter halos \citep[]{cooray01}. Regardless of its nature, a standard rod should satisfy simple criteria: its size should be measured reliably up to high redshifts, and it should be observable in the local Universe to provide calibration. In this study, the relation between the physical size of a spiral galaxy and its rotation velocity (or global profile width) is used. By virtue of this relation, selecting a population of objects with a given rotation velocity is equivalent to selecting them based on their physical size. In other papers of this series \citep[][thereafter Paper I and Paper III]{paperI,paperIII}, the angular diameter test is performed on a dataset of precursor Vimos/VLT Deep Survey (VVDS) \citep[]{vvds} galaxies, using the linewidth-diameter relation to select standard rods. The test is used to isolate the effects of disk evolution with redshift, and to constrain the values of cosmological parameters. Though the VVDS provides excellent data at high redshift, the angular diameter test highlights one of its shortcomings: limited volume coverage at low redshift and therefore the lack of a comparison sample in the local Universe. In this context, the goal of this paper is twofold. First, using a large sample of nearby galaxies \citep[SFI++;][]{spr07} the physical properties of the standard rods and standard candles used to perform the geometrical tests are established, free from any evolutionary biases. This is especially useful to calibrating the angular diameter relation and to separating the effects of galaxy evolution from those of geometrical variations (see Paper I). By establishing clearly the scaling relations between the size, surface brightness and magnitude of these nearby galaxies observed at I-band and their rotation velocity, we establish the calibration that will allow us to select standard rods/candles simply by tracing through redshift galaxies with a given rotation speed. Having a strong handle on the structural parameters of disc galaxies in the local universe gives the unique opportunity of tracing over time the evolution of these properties for disc galaxies hosted in dark matter halos of the same mass, if rotation velocity is used as a proxy for halo mass, and compare them with predictions made under the hierarchical scenario for the growth of structures \citep[e.g.][]{mo98}. The second goal of this paper is to compare the rotation velocity indicator used for the high redshift data to those used in the local Universe, cross-calibrating rotation indicators used at different redshifts. This last point is of relevance since rotation information for spiral galaxies can be obtained through the observation of various spectral lines, the choice of which varies with the redshift of the sample. For galaxies with $z \lesssim 0.1$, the HI 21 cm line is an excellent candidate. The H${\alpha} \lambda$6563\AA\ line is also frequently used for galaxies with low to moderate redshifts. However, the H${\alpha}$ line is quickly redshifted into the near-infrared and becomes unavailable to ground observers using optical telescoptes for galaxies with $z \gtrsim 0.4$, even though with new near-infrared spectrographs such as SINFONI it is now possible to obtain H$\alpha$ rotation curves for high redshift galaxies \citep[e.g.][]{forster06}. Even with the advent of such instruments, most large studies of galaxies at high redshift rely on the [OII]$\lambda 3727$ \AA\ line, including the VVDS and the DEEP2 Redshift Survey \citep[]{davis01}. In order to compare sets of local and distant galaxies, it is therefore necessary to understand how rotation velocities extracted from these different lines relate. It has already been shown that velocity widths derived from HI 21cm and H${\alpha}$ observations are in excellent agreement \citep[see for example][]{courteau97,vogt}. The correlation between the [OII] and HI line widths has also been investigated \citep[]{kobulnicky}. Using a sample of 22 nearby late-type spirals they find [OII] widths to be accurate to within $10 \%$ for galaxies with a roation width of 200 km s$^{-1}$, which is comparable to the overall scatter in the local Tully-Fisher relation. However, they conclude that the uncertainties go up to about 50\% for galaxies with widths $<150$ km s$^{-1}$. Here, a sample of 32 spiral galaxies with $0.155<z<0.25$ is used to compare velocity widths from the H${\alpha} \lambda$6563\AA\ and [OII]$\lambda 3727$ \AA\ lines (hereafter, H${\alpha}$ and [OII]). Even though these galaxies are more distant than in the \citet[]{kobulnicky} sample, the scatter in the rotation curves data points is significantly smaller, resulting in a better estimation of the rotation velocities. Our data also allows for a direct comparison between H${\alpha}$ and [OII], therefore bridging nicely the gap between low and high redshift samples. Contrary to the \citet[]{kobulnicky} study, the galaxy sample used here is restricted to late spiral types, therefore we will not address the question of possible biasses due to varying morphologies. In \S \ref{data} a description of the galaxies observed and of the data reduction process is given, while in \S \ref{widths} we describe how velocity widths were extracted from the data and how they compare for both sets of emission lines. In \S \ref{localtf} we presente a sample of nearby galaxies used as calibrators for the purpose of the angular diameter test (Paper III) and discussion of our results are given in \S \ref{discussion}. Calculations were done assuming $H_0=70$\ km s$^{-1}$ Mpc$^{-1}$, $\Omega_M=0.3$, $\Omega_{\Lambda}=0.7$. | } The main goal of this paper was to provide calibration in two different ways for the angular diameter test that we performed on data from the VVDS: (1) by directly comparing velocity indicators commonly used in the local and distant Universe, and (2) by using a large sample of nearby galaxies to provide zero points for the radius - rotation velocity, magnitude - rotation velocity and surface brightness - rotation velocity relations. Previous studies have focused on comparing the H${\alpha}$ or [OII] rotation velocities with HI data for nearby galaxies. \citet[]{kobulnicky} have used a sample of 22 nearby galaxies to compare [OII] and HI linewidths. Their galaxy sample was smaller and less homogeneous that the one used in this study. They found the two sets of velocity estimates to be consistent within $10\%$ overall, with the [OII] underestimating the HI rotation velocity by up to $50\%$ in a few extreme cases. In the local Universe, the consistency of the HI and H${\alpha}$ rotation velocities was also established \citep[see for example][]{courteau97,vogt}. Most recently, \citet[]{catinella} has shown that on average HI widths are larger than H${\alpha}$ ones by $\sim10\%$ for a galaxy with $w=100$ km s$^{-1}$ and that optical rotational widths measured from velocity histograms are affected by systematic biases and therefore are less reliable that those derived by using the full spatial information through the fitting of the rotation curves. For this study, we first analysed the rotation curves for all the galaxies in our sample and concluded that the rotation velocities derived from the [OII]$\lambda 3727$ \AA\ line translate directly into the velocities obtained from H${\alpha}$ rotation curves, given that the sampling of the [OII] rotation curve extends far enough to reach the region of constant (or slowly increasing) velocity. Even under this condition and as shown in Figure \ref{vrot_plot}, there is scatter in the relation of about 30 km s$^{-1}$. The next step was to compare the techniques used to determine rotation velocities in the local Universe (H${\alpha}$ rotation curves) and at high redshift ([OII] velocity histograms). Once again, there is a direct relation between the two but significant scatter on the order of 50 km s$^{-1}$ (see Figure \ref{sigma_plot}). We conclude that all the rotation velocity indicators studied here give comparable results, but before combining samples analysed with different methods one should be aware of the important scatter in the different relations. This scatter is due mostly to the poorer spatial extent of the [OII] rotation profiles compared to the H${\alpha}$ ones obtained with the same integration time, and can be as large as $25\%$ for a galaxy with $w=200$ km s$^{-1}$ or $100\%$ if $w=50$ km s$^{-1}$ in the case of the $\sigma_{rot}$([OII])-$v_{rot}$(H${\alpha}$) relation (Fig.\ref{sigma_plot}). Even considering this large scatter, the fact that there is no systematic deviation from a 1:1 relation allows us to use both methods in their respective context to divide galaxies in classes based on rotation velocity, as required to perform the geometrical tests with the high redshift sample (see Paper III). In Paper I it was showed that this is a viable way of selecting standard rods for the purpose of the angular diameter test. Since one cannot perform the test only with high redshift data, it was of great importance to establish that a set of local calibrators can be reliably compared to it, even if the rotation measure is different. Once it is established that the various rotation indicators can be reliably compared, the next critical step is to derive the scaling relations between luminosity, size, surface brightness and rotation velocity for a set of nearby galaxies, free from any evolution. We used the SFI++ catalog, a very large homogeneous sample of nearby spiral galaxies with I-band photometry and rotation information obtained either through optical long-slit spectroscopy or HI measurements. We have particularly focussed our attention on the $r-v_{rot}$ relation which is central to the cosmological tests performed in the other papers of this series. We have shown that by carefully correcting sizes for inclination and extinction effects as well as obtaining accurate distances to the galaxies, the $r_{23.5}-v_{rot}$ relation can be almost as tight as the Tully-Fisher relation, as judged by their Pearson correlation coefficients. The relation derived using $r_{23.5}$ as the size measurement is considerably tighter than the ones obtained using the disc scale length $r_d$ or a radius defined to contain a given fraction of the light of the galaxy (in our case $r_{83}$ or $r_{50}$). The next phase for this study on scaling relations will be to see what constraints our $r-v_{rot}$ relation imposes on models for disc galaxy formation, for example the range of spin parameters its tight scatter allows. These scaling relations also offer exciting possibilities for the study of disc galaxy evolution. Since they provide accurate values for luminosity, size and surface brightness as a function of rotation velocity, measurements of the properties for galaxies with the same rotation velocity (i.e. hosted in a dark halo of the same mass) over a large redshift range can provide in a new and unique way information about the evolution of galaxies and a mean of testing predictions from models under the hierarchical scenario for the growth of structures. A pilot study is presented in the next paper of this series. | 7 | 10 | 0710.0760 |
0710 | 0710.3763_arXiv.txt | {} {To measure the supernova (SN) rates at intermediate redshift we performed the Southern inTermediate Redshift ESO Supernova Search (STRESS). Unlike most of the current high redshift SN searches, this survey was specifically designed to estimate the rate for both type Ia and core collapse (CC) SNe.} {We counted the SNe discovered in a selected galaxy sample measuring SN rate per unit blue band luminosity. Our analysis is based on a sample of $\sim 43000$ galaxies and on 25 spectroscopically confirmed SNe plus 64 selected SN candidates. Our approach is aimed at obtaining a direct comparison of the high redshift and local rates and at investigating the dependence of the rates on specific galaxy properties, most notably their colour.} {The type Ia SN rate, at mean redshift $z=0.3$, amounts to 0.22$^{+0.10 +0.16}_{-0.08 -0.14}$ $h_{70}^2$ SNu, while the CC SN rate, at $z=0.21$, is 0.82$^{+0.31 +0.30}_{-0.24 -0.26}$ $h_{70}^2$ SNu. The quoted errors are the statistical and systematic uncertainties. } {With respect to local value, the CC SN rate at $z=0.2$ is higher by a factor of $\sim 2$ already at redshift , whereas the type Ia SN rate remains almost constant. This implies that a significant fraction of SN~Ia progenitors has a lifetime longer than $2-3$ Gyr. We also measured the SN rates in the red and blue galaxies and found that the SN~Ia rate seems to be constant in galaxies of different colour, whereas the CC SN rate seems to peak in blue galaxies, as in the local Universe. SN rates per unit volume were found to be consistent with other measurements showing a steeper evolution with redshift for CC SNe with respect to SNe~Ia. Finally we have exploited the link between SFH and SN rates to predict the evolutionary behaviour of the SN rates and compare it with the path indicated by observations. We conclude that in order to constrain the mass range of CC SN progenitors and SN Ia progenitor models it is necessary to reduce the uncertainties in the cosmic SFH. In addition it is important to apply a consistent dust extinction correction both to SF and to CC SN rate and to measure SN~Ia rate in star forming and in passive evolving galaxies in a wide redshift range. } | We now detail how all the ingredients described in the previous sections have been combined to compute the control time of our galaxy sample. First, for a given SN type and filter $F$, the control time of the $i$-observation of the $j$-galaxy was computed as: \begin{equation} CT^{\rm SN,F}_{j,i} = \int{\tau^{\rm SN,F}_{j}(m) \,\epsilon^{\rm F}_{i} (m)\,{\rm d}m} \end{equation} where $\tau^{\rm SN,F}_{j}(m)$ is the time spent by the SN in the magnitude range $m$ and $m+{\rm d}m$ and $\epsilon^{\rm F}_{i} (m)$ is the detection efficiency. More specifically, we convolved the distribution of the absolute magnitude at maximum ($M^{\rm SN}_{B}(0)$) and of the absorption due to host galaxy extinction ($A_B^h$), so as to determine the distribution of the quantity $M^{\rm SN}_{B}(0)+A_B^h$ appearing in Eq.~\ref{lc}, and computed $\tau^{\rm SN,F}_{j}(m)$ as a weighted average of the individual times over this combined distribution. Then, the total control time $CT^{\rm SN,F}_j$ of the $j$-galaxy was computed by summing the contribution of individual observations. If the temporal interval elapsed since the previous observation is longer than the control time, that contribution is equal to the control time of the observation, otherwise it is equal to the interval of time between the two observations \citep{Cappellaro99}. The total control time of the galaxy sample is obtained as the $B$ band luminosity weighted ($\overline {CT}^{\rm SN,F}_j$) average of the individual galaxies. Since we merged all CC subtypes, including type Ib/c IIP and IIL, the control time for CC SNe is computed as follows: \begin{equation}\label{eqctcc} \overline{CT}^{\rm CC}_j = f_{\rm Ib/c}\, \overline{CT}^{\rm Ib/c}_j + f_{\rm IIL}\, \overline{CT}^{\rm IIL}_j + f_{\rm IIP}\, \overline{CT}^{\rm IIP}_j \end{equation} where the relative fractions of the different CC subtypes is assumed to be constant with redshift and equal to that observed in the local Universe, namely 20$\%$ of Ib/c and 80$\%$ of II \citep{Cappellaro99}, out of which 35$\%$ are IIL and 65$\%$ are IIP events \citep{Richardson02}. In order to illustrate the role of the host galaxy extinction, we calculated the CC SN rate by adopting the control time both for a {\em standard} ($\tau(0)=1.0$), and for a {\em high} ($\tau(0)=5.0$) extinction scenario. | \label{Discussion} In this section we discuss the evolution of the SN rates and investigate on the link between SF and SN rates. First we verify the consistency of our estimate of the redshift evolution of SN rates with other measurements from the literature (Sec.~\ref{Complit}). Then we compare the observed evolution of the CC SN rate with those expected for different SFHs (Sec.~\ref{CompSFR}). Finally, we convolve the SFH of \citet{Hopkins06} with the delay time distribution (DTD) for different progenitor scenarios, and compare the results to the measurements of the SN~Ia rate evolution (Sec.~\ref{Iamod}). \subsection{Comparison with other estimates}\label{Complit} Published measurements of the SN rates at intermediate and high redshift are expressed in units of co-moving volume. To convert our rates from SNu to volumetric rates we assumed that the SN rates are proportional to the galaxy $B$ band luminosity. If this assumption is true, by multiplyng the rates in SNu by the total $B$ band luminosity density in a given volume we derived the total rates in that volume, even if we did not sample the faint end of the galaxy luminosity function. We accounted for the evolution of the $B$ band luminosity density with redshift. A compilation of recent measurements of the $B$ band luminosity density is plotted as function of redshift in Fig \ref{lumdens}, where we also show a linear least-square fit to the data in the range $z=0-1$: $j_B(z) = (1.03+1.76\times z)$ ~$10^8 L_{\odot}^{B} {\rm Mpc}^{-3}$. \begin{figure} \resizebox{\hsize}{!}{\includegraphics{lumden.eps}} \caption{Measurements of the galaxy luminosity density at different redshifts. The DEEP2 and COMBO-17 data are taken from Table 2 in \citet{Faber}. The line represents the linear least-square fit in the redshift interval $z=0-1$.} \label{lumdens} \end{figure} Multiplying our measurements by the value of $j_B$ at the average redshifts of the Ia and CC SN samples, the rates per unit of co-moving volume result: \begin{itemize} \item $r^{\rm Ia}(z=0.30) = 0.34^{+0.16 +0.21}_{-0.15 -0.22}$ $10^{-4} h_{70}^3 \mbox{yr}^{-1} \mbox{Mpc}^{-3}$\\ \item$r^{\rm CC}(z=0.21) = 1.15^{+0.43 +0.42}_{-0.33 -0.36}$ $10^{-4} h_{70}^3 \mbox{yr}^{-1} \mbox{Mpc}^{-3}$ \end{itemize} where both statistical and systematic errors are indicated. Also local rates are converted into volumetric rates: $r^{\rm Ia}(z=0.01) =0.18\pm0.05$ $10^{-4} h_{70}^3 \mbox{yr}^{-1} \mbox{Mpc}^{-3}$ and $r^{\rm CC}(z=0.01) = 0.43\pm0.17$ $10^{-4} h_{70}^3 \mbox{yr}^{-1} \mbox{Mpc}^{-3}$. With respect to the rates in SNu's, the volumetric rates evolve more rapidly with redshift, due to the increase of the $B$ band luminosity density. We find an increase of a factor of $\sim 2$ at $z=0.3$ for SNe~Ia, and a factor of $\sim 3$ at $z=0.21$ for CC SNe. Measurements of Ia and CC SN rate as function of redshift are shown in Fig.~\ref{obsrate}, where those originally given in SNu \citep{Hardin,Blanc,Cappellaro05} were converted into measurements per unit volume as above. \begin{figure*} \resizebox{\hsize}{!}{\includegraphics{obs.eps}} \caption{Observed SN rates as function of redshift from different authors as indicated in the legend. The black (gray) symbols indicate SN~Ia (SN~CC) rate measurements. The shaded area represents the 1 $\sigma$ confidence level of our rate evolution estimate as deduced from the MLE fit.} \label{obsrate} \end{figure*} As it can be seen, the few measurements of the CC SN rate appear to be fully consistent, while those of the SN~Ia rate show a significant dispersion which increases with redshift, in particular in the range $0.5<z<0.7$ where the values of \citet{Barris} and \citet{Dahlen04} are 2-3 times higher that those of \citet{Pain02} and \citet{Neill07}. Our estimate of the SN~Ia rate is consistent with all other measurements in the redshift range we explored; our result does not help to discriminate between the steep trend suggested by the \citet{Barris} and \citet{Dahlen04} measurements and the slow evolution indicated by the \citet{Neill07} measurement. The robust indication from the current data appears that the SN~Ia rate per unit volume at redshift 0.3 is a factor of $\sim 2$ higher that in the local Universe, while in the same redshift range the CC SN rate increases by a factor of $\sim 5$. \subsection{Comparison with the predicted evolution of the CC SN rate}\label{CompSFR} The stellar evolution theory predicts that all stars more massive than 8-10~M$_\odot$ complete the eso-energetic nuclear burnings, up to the development of an iron core that cannot be supported by any further nuclear fusion reactions or by electron degenerate pressure. The subsequent collapse of the iron core results into the formation of a compact object, a neutron star or a black hole, accompanied by the high-velocity ejection of a large fraction of the progenitor mass. TypeII SNe originate from the core collapse of stars that, at the time of explosion, still retain their H envelopes, whereas the progenitors of type Ib$/$Ic SNe are thought to be massive stars which have lost their H (and He) envelope \citep{Heger}. Given the short lifetime of their progenitors ($<30$~Myr), there is a simple, direct relation between the CC SN and the current SF rate: \begin{equation} r^{\rm CC}(z) = K^{\rm CC} \times \psi(z) \end{equation} where $\psi(z)$ is the SFR and $K^{CC}$ is the number of CC SN progenitors from a 1 $M_\odot$ stellar population: \begin{equation} K^{\rm CC} = \frac{\int_{m_l^{\rm CC}}^{m^{\rm CC}_u} \phi(m) dm}{\int_{m_L}^{m_U}m\phi(m) dm} \end{equation} where $\phi(m)$ is the IMF, $m_L-m_U$ is the total stellar mass range, and $m^{\rm CC}_l-m^{\rm CC}_u$ is the mass range of CC SN progenitors. \begin{figure} \resizebox{\hsize}{!}{\includegraphics{rate_cc.eps}} \caption{Comparison between the SN~CC and SF rate evolution. Symbols are as in Fig.~\ref{obsrate} with additional open symbols (measurements not corrected for extinction) and filled black symbols (estimates for the high extinction correction). Lines are selected SFR evolutions from the literature. All SFHs have been scaled to the SalpeterA IMF.} \label{ccrate} \end{figure} In principle, if accurate measurements of the CC SN and SF rates are available, it is possible to probe the possible evolution with redshift of either the CC SN progenitor scenarios or the IMF by determining the value of $K^{CC}$. Here however we assume that $K^{CC}$ does not evolve significantly in the redshift range of interest and compare the observed evolution of the CC SN rate with that predicted by the SFH. The estimate of the cosmic SFH is based on different SF indicators, depending on redshift range, and requires a suitable parameterization and accurate normalization. Since there is a large scatter between the measurements obtained by different SFR indicators (a factor 2-3 at $z=0.3-0.5$ and increasing with redshift) it is hard to obtain a consistent picture of the SFH \citep{Hopkins06}. Indeed observations made at different wavelengths, from X-rays to radio, sample different facets of the SF activity and are sensitive to different time scales over which the SFR is averaged. Thus, different assumptions are required to convert the observed luminosities at the various wavelengths to the SFR, and different systematic uncertainties affect the SFR estimates. In particular the significant difference between the SFR inferred from the UV and $H\alpha$ luminosity and that inferred from the far-IR (FIR) luminosity may be related to the effect of dust extinction (expecially for the UV) and/or to the contribution to the light from old stars and AGNs (for the FIR). To illustrate this point, we select three representative prescriptions for the SFH, namely: \begin{itemize} \item the piecewise linear fit of selected SFR measurements in the range $0<z<6$ by \citet{Hopkins06}, \item the fit to the SFR measurements from the $H\alpha$ emission line, by \citet{Hippelein}, with an exponential increase from $z=0$ to $z=1.2$, \item the prescriptions by \citet{Hernquist}, i.e. a double exponential function that peaks at redshift $z\sim 5.5$, obtained from an analytical model and hydro-dynamic simulations. \end{itemize} All these SFHs were converted to the same IMF, a modified Salpeter IMF (SalA) with a turnover below 0.5 $M_{\odot}$ and defined in the mass range $m_L=0.1$$M_{\odot}$ to $m_U=120$ $M_{\odot}$ \citep{Baldry}. We assumed a mass range of $8-50 M_{\odot}$ for CC SN progenitors, which gives a scale factor $K^{\rm CC}=0.009$. The measurements of the CC SN rate per unit volume and the predicted evolutionary behaviours are shown in Fig~\ref{ccrate}. The observations confirm the steep increase with redshift expected by the SFH from \citet{Hopkins06} and \citet{Hippelein}. For a look-back time of 3~Gyr ($z=0.25$) both the SFR and the CC SN rate increase by a factor of $\sim3$ compared with the local values. A flat evolution, as that proposed by \citet{Hernquist}, appears inconsistent with the observed CC SN rates in the overall range of redshift. With the adopted $K^{\rm CC}$, the level of the CC SN rate predicted by the SFH of \citet{Hippelein} and, in general, by the $UV$ and $H\alpha$ based SFHs fits well the data. Instead, for the SFH of \citet{Hopkins06} and in general the SFHs inferred through FIR luminosity, the predicted CC SN rate is higher than observed over the entire redshift range \citep[see also][]{Dahlen04,Hopkins06,Mannucci07}. If we correct the observed CC SN rate according to the high extinction scenario, we obtain an acceptable agreement between the data and the predictions with \citet{Hopkins06} SFH, as shown in Fig.~\ref{ccrate}. However, this correction would require an extremely high dust content in galaxies which is not favored by present measurements. Indeed, \citet{Mannucci07} derived an estimate of the fraction of SNe which are likely to be missed in optical SN searches because they occur in the nucleus of starburst galaxies or, in general, in regions of very high extinction and found that, at the average redshift of our search, the fraction of missing CC SNe is only $\sim 10\%$, far too small to fill the gap between observed and predicted rates. Alternatively we may consider the possibility of a narrower range for the CC SN progenitor masses: in particular, a lower limit of $10-12$ $M_{\odot}$ would bring the observed CC SN rates in agreement with those predicted from FIR based SFHs. In this respect we notice that, from a theoretical point of view, there is the possibility that a fraction of stars between 7-8 M$_\odot$ and 10-12 M$_\odot$ avoids the collapse of the core and ends up as ONeMg White Dwarfs \citep{Ritossa,Poelarends}. On the other hand, estimates of the progenitor mass from the detection in pre-explosion (HST) images has been possible for a few SNe IIP (e.g. SN 2003gd \citep{Hendry}, SN 2005cs \citep{Pastorello}): their absolute magnitudes and colours seem indicate a moderate mass ($8-12$ $M_{\odot}$ )\citep{VanDyk,Smartt,Maund,Li06}. Given these controversies, we conclude that in order to constrain the mass range of CC SN progenitors it is necessary to reduce the uncertainties in the cosmic SFH. In addition it is important to apply a consistent dust extinction correction both to SFH and to CC SN rate. \subsection{Comparison with the predicted evolution of SN Ia rate}\label{Iamod} According to the standard scenario SNe~Ia originate from the thermonuclear explosion of a Carbon and Oxygen White Dwarf (C-O WD) in a binary system. In the first phase of the evolution the primary component, a star less massive than $8 M_{\odot}$, evolves into a C-O WD. When the secondary expands and fills its Roche Lobe, two different paths are possible, depending on whether a common envelope forms around the two stars (double degenerate scenario, DD) or not (single degenerate scenario, SD). In the SD scenario, the WD remains confined within its Roche Lobe, grows in mass until it reaches the Chandrasekhar limit and explodes, while in the DD scenario the binary system evolves into a close double WD system, that merges after orbital shrinking due to the emission of gravitational wave radiation. In both scenarios two basic ingredients are required to model the evolution of the SN~Ia rate: the fraction of the binary systems that end up in a SN~Ia, and the distribution of the time elapsed from star formation to explosion (delay-time). In the SD scenario, the delay time is the evolutionary lifetime of the secondary; in the DD, the gravitational radiation timescale has to be added. In both cases, the distribution of the delay times depends on the distribution of the binary parameters \citep{Greggio05}. In principle, the SD and DD scenarios correspond to different realization probabilities and different shape of the delay time distribution (DTD) functions, hence rather different evolutionary behavior of the SN~Ia rate. As mentioned previously, the observations indicate that the distribution of the delay times of SN~Ia progenitors is rather wide. The cosmic SFH is the other critical ingredient that modulates the evolution of the SN~Ia rate. Following \citet{Greggio05} the SN~Ia rate is given by: \begin{equation}\label{sniaeq} r^{\rm Ia}(t) = k_{\alpha}A^{\rm Ia}\int_{\tau_{i}}^{min(t,\tau_{x})} f^{\rm Ia}(\tau) \psi(t- \tau)d{\tau} \end{equation} where $k_{\alpha}$ is the number of stars per unit mass of the stellar generation, $A^{\rm Ia}$ is the realization probability of the SN~Ia scenario (the number fraction of stars from each stellar generation that end up as SN~Ia), $f^{\rm Ia}(\tau)$ is the distribution function of the delay times and $\psi(t-\tau)$ is the star formation rate at the epoch $t-\tau$. The integration is extended over all values of the delay time $\tau$ in the range $\tau_{i}$ and $min(t,\tau_{x})$, with $\tau_{i}$ and $\tau_{x}$ being the minimum and maximum possible delay times for a given progenitor scenario. Here we assumed that both $k_{\alpha}$ and $A^{\rm Ia}$ do not vary with cosmic time. A detailed analysis of the predicted evolution of the SN~Ia rate for different SFHs and DTDs is presented elsewhere \citep{Forster,Blanc07}. Here we consider only one SFH, the piecewise interpolation of \citet{Hopkins06}, since this is conveniently defined also at high redshift, and limit our analysis to few DTDs representative of different approaches to model the SN~Ia rate evolution: three models from \citet{Greggio05}, and two different parametrizations \citep{Mannucci06,Strolger} designed to address some specific observational constraints, regardless the correspondence to a specific progenitor scenario. Specifically, among the \citet{Greggio05} models we select one SD and two DD-models, one of the ``close'' and the other of the ``wide'' variety, the latter being an example of a relatively flat DTD. This choice is meant to represent the full range of plausible DTDs. The minimum delay time for the SD model is the nuclear lifetime of the most massive secondary stars in the SN Ia progenitor's system, i.e. $8 M_{\odot}$. In principle, for the DD model the minimum delay time could be appreciably larger than this because of the additional gravitational waves radiation delay. In practice, also for the DDs the minimum delay is of a few $10^{7}$ yrs. The maximum delay time is quite sensitive to the model for the SN~Ia progenitors. The reader should refer to \citet{Greggio05} for a more detailed description of these models. The DTD parametrization proposed by \citet{Mannucci06} is the sum of two distinct functions: a Gaussian centered at 5$\times10^{7}$ yr, representative of a ``prompt'' progenitor population which traces the more recent SFR, and an exponentially declining function with characteristic time of 3 Gyr, a ``tardy'' progenitor population proportional to the total stellar mass. This DTD was introduced to explain, at the same time, the dependence of the SN~Ia rate per unit mass on the galaxy morphological type, the cosmic evolution of the SN~Ia rate and, in particular, the high SN~Ia rate observed in radio-loud galaxies. Finally, \citet{Strolger} showed that the best fit of the apparent decline of SN~Ia rate at $z> 1$ is achieved for a DTD with a Gaussian distribution centered at about 3~Gyr. Nevertheless this DTD fails to reproduce the dependence of the SN rate on galaxy colours which is observed in the local Universe \citep{Mannucci06}. \begin{figure} \resizebox{\hsize}{!}{\includegraphics{delay.eps}} \caption{Delay time distributions as derived by \citet{Greggio05} for SD and DD models, \citet{Mannucci06} and \citet{Strolger}.} \label{delay} \end{figure} The selected DTD functions are plotted in Fig.~\ref{delay} while the predicted evolutionary behaviours of the SN~Ia rate are compared with all published measurements in Fig.~\ref{rateia}. In all cases, the value of $k_{\alpha}A^{\rm Ia}$ was fixed to match the value of the local rate; depending on the model it ranges between 3.4-7.6$\times 10^{-4}$. This normalization implies that, for the adopted SalA IMF, and assuming a mass range for the progenitors of $3-8$ $M_\odot$, the probability that a star with suitable mass becomes a SN~Ia, is $\sim$ 0.01-0.03. \begin{figure} \resizebox{\hsize}{!}{\includegraphics{rate_ia.eps}} \caption{SN~Ia rate measurements fitted with different DTD functions and the SFH by \citet{Hopkins06}. Symbols for measurements are as in Fig.~\ref{obsrate}.} \label{rateia} \end{figure} The models obtained with the different DTDs are all consistent with the observations with the exception of the "wide" DD model, whose redshift evolution is definitely too flat. On the other hand none of the DTD functions, with the adopted SFH, is able to reproduce at the same time the very rapid increase from redshift 0 to 0.5 suggested by some measurements \citep{Barris,Dahlen04} and the decline at redshift $>1$ \citep{Dahlen04}. We note that a new measurement of \citet{Poznanski} suggests that the SN~Ia rate decline at high redshift may be not as steep as estimated by \citet{Dahlen04}. Given the current uncertainties of both SN~Ia rate and SFH it is difficult to discriminate between the different DTD functions and hence between the different SN~Ia progenitor models. To improve on this point, more measurements of SN~Ia rate at high redshifts are required to better trace the rate evolution. At the same time measurements in star forming and in passive evolving galaxies in a wide redshift range can provide important evidence about the SN~Ia progenitor models. In addition, it is essential to estimate the cosmic SFH more accurately because the position of the peak of the SFH was found to be the crucial parameter for the recovered delay time \citep{Forster}. | 7 | 10 | 0710.3763 |
0710 | 0710.1550_arXiv.txt | We present VLT/FORS2 spectroscopic observations of globular clusters (GCs) in five low surface brightness (LSB) dwarf galaxies: KK211 and KK221, which are both dwarf spheroidal satellites (dSph) of NGC~5128, dSph KK84 located close to the isolated S0 galaxy NGC~3115, and two isolated dwarf irregular (dIrr) galaxies UGC~3755 and ESO~490-17.~Our sample is selected from the Sharina et al. (2005) database of Hubble Space Telescope WFPC2 photometry of GC candidates in dwarf galaxies. For objects with accurate radial velocity measurements we confirm 26 as genuine GCs out of the 27 selected candidates from our WFPC2 survey.~One candidate appears to be a distant galaxy.~Our measurements of the Lick absorption line indices in the spectra of confirmed GCs and the subsequent comparison with SSP model predictions show that all confirmed GCs in dSphs are old, except GC KK211-3-149 ($6 \pm$2 Gyr), which we consider to be the nucleus of KK211. GCs in UGC~3755 and ESO~490-17 show a large spread in ages ranging from old objects ($t>10$ Gyr) to clusters with ages around 1 Gyr. Most of our sample GCs have low metallicities $\zh \le -1$. Two relatively metal-rich clusters with $\zh \approx -0.3$ are likely to be associated with NGC~3115. Our sample GCs show in general a complex distribution of $\alpha$-element enhancement with a mean $\langle$[$\alpha$/Fe]$\rangle=0.19\pm0.04$ derived with the $\chi^{2}$ minimization technique and $0.18\pm0.12$ dex computed with the iterative approach. These values are slightly lower than the mean $\langle$[$\alpha$/Fe]$\rangle=0.29\pm0.01$ for typical Milky Way GCs. We compare other abundance ratios with those of Local Group GCs and find indications for systematic differences in N and Ca abundance. The specific frequencies, $S_N$, of our sample galaxies are in line with the predictions of a simple mass-loss model for dwarf galaxies and compare well with $S_N$ values found for dwarf galaxies in nearby galaxy clusters. | \label{intro} The hierarchical structure formation scenario predicts that dwarf galaxies are the first systems to form in the Universe \citep{peebles68}, and that more massive galaxies form through dissipative processes from these smaller sub-units. The involved physical mechanisms of this sequence depend on the density and mass of the parent dark matter halo, in the sense that more massive halos initiate star formation at earlier epochs and form their stars at a faster rate \citep[e.g.][]{peebles02, renzini06, ellis07}. Because of this environmental gradient, we expect that dwarf galaxies in the field formed the first stellar population relatively late and at a lower pace compared to their counterparts in dense galaxy clusters. In other words, the difference in age and chemical composition between the oldest stellar populations in cluster and field dwarf galaxies should reflect the delay in the onset of structure formation in these two environments. The task of measuring the age and chemical composition of the oldest stellar populations in distant dwarf galaxies from their integrated light is very challenging. An alternative approach is to investigate the oldest globular clusters (GCs) that are found in dwarf galaxies. Several photometric surveys of extragalactic GCs in dwarf galaxies outside the Local Group have been performed in the past decade \citep[see review by][]{miller06}. However, only a handful of those were followed up with 8--10m-class telescopes to derive spectroscopic ages and chemical composition. Observations of galaxies in groups and clusters provide more and more evidence that environment is a major factor influencing the process of GC formation \citep[e.g.][]{west93, tully02, grebel03, miller98}.~Recent progress in modeling the assembly history of GC systems in massive elliptical galaxies suggests that a significant fraction of metal-poor GCs were accreted from dwarf satellites at later times compared to the number of GCs initially formed in the parent galaxy halo \citep{ppm07}. These results underline the ideas put forward in the work of \cite{forte82} and \cite{muzzio87} as well as the models of \cite{cote98, cote02, hilker99}, who suggested that the rich GC systems of massive galaxies may be the result of significant GC accretion through tidal stripping of less massive systems. Spectroscopic studies of a few GCs in cluster and field dwarf galaxies showed that most of these systems host at least some old GCs with ages $t\ga10$ Gyr \citep{puzia00, sharina03, strader03a, strader03b, beasley06, conselice06}.~Although today's accuracy of relative spectroscopic age determinations ($\Delta t/t \approx 0.2-0.3$) is not sufficient to resolve the expected delay of $\sim\!0.5-4$ Gyr in the onset of star-formation between cluster and field environment \citep[depending on cosmology, ionizing source population, ionization feedback efficiency, etc., see][]{kauffmann96, treu05, thomas05, delucia06, clemens06}, the old ages combined with information on abundance ratios can provide a powerful tool to decide whether stellar populations in field dwarf galaxies followed the same early enrichment history as their analogs in denser environments. Furthermore, any difference in GC chemical composition between dwarf and more massive galaxies opens an attractive way of chemically tagging accreted sub-populations in massive halos and, therefore, enables us to quantify the mass accretion history of galaxies, a task that for old galaxies is infeasible from studies of the diffuse galaxy light alone. Similar ideas have been formulated for stellar populations that make up the diffuse component of the most nearby galaxies, which are close enough for high-resolution spectroscopy of individual stars \citep{fh02, geisler07}. The obvious advantage of GC systems is that their spectra can be observed out to about 10 times greater distances. In this work we analyze spectroscopic observations of GCs in dwarf irregular (dIrr) and dwarf spheroidal galaxies (dSph) in the field/group environment. Our sample consists of systems that are representatives of the lowest-mass bin of the Local Volume (LV) galaxy population, limited to distances $D<10$ Mpc. We refer to \cite{karachentseva85} and \cite{grebel99} for a morphological type definition of our sample galaxies. The paper is organised as follows. In Section~\ref{observations} we describe our observations and data reduction as well as the methods of measuring radial velocities. Section~\ref{analysis} summarizes the measurement of Lick line indices, their calibration, and the determination of spectroscopic ages, metallicities, and abundance ratios. Section~\ref{discussion} is devoted to the discussion of our results. Conclusions are presented in section~\ref{conclusion}. \begin{deluxetable*}{lccccccclcr}[!ht] \tabletypesize{\scriptsize} \tablecaption{Properties of sample dwarf galaxies \label{dwgprop}} \tablewidth{0pt} \tablehead{ \colhead{Galaxy} & \colhead{RA (J2000)}& \colhead{DEC (J2000)} & \colhead{$ \mu_0$} & \colhead{$D_{\rm MD}$} & \colhead{$A_{B}$} & \colhead{$B$} & \colhead{$B-I$} & \colhead{$M_V$} & \colhead{$N_{\rm GC}$} & \colhead{$S_{N}$} } \startdata KK211, AM1339-445 & 13 42 06 & $-$45 13 18& 27.77 & 0.25 & 0.477 & 16.3$\pm$0.2 & 1.8$\pm$0.2 & $-$12.58& 2 & 18.6\\ KK221 & 13 48 46 & $-$46 59 49& 28.00 & 0.50 & 0.596 & 17.3$\pm$0.4 & 2.0$\pm$0.4 & $-$11.96& 6 & 95.1\\ KK084, UA200, KDG65 & 10 05 34 & $-$07 44 57& 29.93 & 0.03 & 0.205 & 16.4$\pm$0.2 & 1.7$\pm$0.2 & $-$14.40& 7 & 10.4\\ UGC3755, PGC020445 & 07 13 52 & $+$10 31 19& 29.35 &$\sim5$ & 0.384 & 14.1$\pm$0.2 & 1.1$\pm$0.2 & $-$16.12& 32 & 11.4\\\smallskip E490-017, PGC019337 & 06 37 57 & $-$25 59 59& 28.13 & 4.50 & 0.377 & 14.1$\pm$0.2 & 1.1$\pm$0.2 & $-$14.90& 5 & 5.4 \\ \enddata \tablecomments{Columns contain the following data: (1) galaxy name, (2) equatorial coordinates, (3) and (4) are the distance modulus and distance from the nearest bright galaxy in Mpc from \cite{kara04}, and from \cite{tully05} for UGC3755, (5) reddening from \cite{sch98} in the B-band, (6) is the integrated $B$ magnitude, and (7) the integrated $B-I$ color, derived from surface photometry on the VLT-FORS2 images in this work (see Appendix~\ref{sbprof}), (8) absolute $V$ magnitude from SPM05, (9) number of GCs according to SPM05 and this paper, (10) specific frequency, $S_N=N_{\rm GC}10^{0.4(M_V+15)}$ \citep{harris81}.} \end{deluxetable*} | \label{conclusion} Numerous photometric and spectroscopic studies of globular clusters in Virgo and Fornax cluster dwarf galaxies have been undertaken in the last years, which targeted bright dwarf galaxies down to $M_V\approx-15$ mag \citep[see][]{miller06}. Due to observational selection effects dwarf galaxies fainter than this are missed at distances of $D \approx 17$ Mpc. Faint LSB dwarf galaxies down to $M_V \approx -12$ mag have long been thought to be free of globular clusters, because they have insufficient mass. Our HST/WFPC2 survey of low-mass dwarf galaxies (SPM05), situated at distances $2-6$ Mpc in the Local Volume, revealed a rich population of globular cluster candidates (GCCs). In this work, we observed five of these galaxies with the VLT/FORS2 spectrograph in MXU mode and found that all targeted GCCs except one are genuine globular clusters. We could also confirm five additional globular clusters in our sample galaxies. Two clusters appear to be the nuclei of KK84 and KK211. The confirmed globular clusters are in general old and metal-poor, and show a range of \afe\ ratios. The mean $\langle$[$\alpha$/Fe]$\rangle=0.19\pm0.04$ that was determined with the $\chi^{2}$ minimization technique and $0.18\pm0.12$ dex which was computed using the iterative approach appears slightly lower than the mean $\langle$[$\alpha$/Fe]$\rangle=0.29\pm0.01$ for typical Milky Way clusters. Globular clusters in the two isolated, relatively bright dwarf galaxies UGC~3755 and ESO~490-17 show a wide range of ages from 1 to 9 Gyr, and imply extended star formation histories in these galaxies. This goes in hand with the measured low \afe\ ratios for the younger clusters and is consistent with low intensity star bursts. The oldest clusters with the highest \afe\ are found in KK84, a companion of NGC~3115. Other chemical abundances indicate potentially interesting differences between globular clusters in dwarf and more massive galaxies and, if confirmed, would facilitate the quantification of the accreted mass in rich GC systems of massive early-type galaxies. | 7 | 10 | 0710.1550 |
0710 | 0710.4143_arXiv.txt | The gravitational magnification and demagnification of Type Ia supernovae (SNe) modify their positions on the Hubble diagram, shifting the distance estimates from the underlying luminosity-distance relation. This can introduce a systematic uncertainty in the dark energy equation of state (EOS) estimated from SNe, although this systematic is expected to average away for sufficiently large data sets. Using mock SN samples over the redshift range $0 < z \leq 1.7$ we quantify the lensing bias. We find that the bias on the dark energy EOS is less than half a percent for large datasets ($\gtrsim$ 2,000 SNe). However, if highly magnified events (SNe deviating by more than 2.5$\sigma$) are systematically removed from the analysis, the bias increases to $\sim$ 0.8\%. Given that the EOS parameters measured from such a sample have a 1$\sigma$ uncertainty of 10\%, the systematic bias related to lensing in SN data out to $z \sim 1.7$ can be safely ignored in future cosmological measurements. | \label{sec:introduction} Since the discovery of the accelerating expansion of the universe~\citep{Riess:98, Perl:99, Knop:03, Riess:04}, the quest to understand the physics responsible for this acceleration has been one of the major challenges of cosmology. At present the dominant explanation entails an additional energy density to the universe called dark energy. The physics of dark energy is generally described in terms of its equation of state (EOS), the ratio of its pressure to density. In some models this quantity can vary with redshift. While there exist a variety of probes to explore the nature of dark energy, one of the most compelling entails the use of type Ia supernovae to map the Hubble diagram, and thereby directly determine the expansion history of the universe. With increasing sample sizes, SN distances can potentially provide multiple independent estimates of the EOS when binned in redshift \citep{Huterer:05,Sullivan:07a,Sullivan:07b}. Several present and future SN surveys, such as SNLS \citep{snls} and the Joint Dark Energy Mission (JDEM), are aimed at constraining the value of the dark energy EOS to better than 10\%. Although SNe have been shown to be good standardizable candles, the distance estimate to a given SN is degraded due to gravitational lensing of its flux \citep{Frieman:97, Wambsganss:97, HolzWald:98}. The lensing becomes more prominent as we observe SNe out to higher redshift, with the extra dispersion induced by lensing becoming comparable to the intrinsic dispersion (of $\sim0.1$ magnitudes) at $z \gtrsim 1.2$ \citep{Holz:05}. In addition to this dispersion, which leads to an increase in the error associated with distance estimate to each individual supernova, lensing also correlates distance estimates of nearby SNe on the sky, since the lines-of-sight pass through correlated foreground large-scale structure \citep{Cooray:05,Hui:06}. Although this correlation error cannot be statistically eliminated by increasing the number of SNe in the Hubble diagram, the errors can be controlled by conducting sufficiently wide-area ($> 5\mbox{ deg}^2$) searches for SNe (in lieu of small-area pencil-beam surveys). In addition to the statistical covariance of SN distance estimates, gravitational lensing also introduces systematic uncertainties in the Hubble diagram by introducing a non-Gaussian dispersion in the observed luminosities of distant SNe. Since lensing conserves the total number of photons, this systematic bias averages away if sufficiently large numbers of SNe per redshift bin are observed. In this case the average flux of the many magnified and demagnified SNe converges on the unlensed value \citep{Holz:05}. Nonetheless, even with thousands of SNe in the total sample it is possible that the averaging remains insufficient, given that one may need to bin the Hubble diagram at very small redshift intervals to improve sensitivity to the EOS. Furthermore, SNe at higher redshifts are more likely to be significantly lensed. If ``obvious'' outliers to the Hubble diagram are removed from the sample, this introduces an important bias in cosmological parameter determination, and can lead to systematic errors in the determination of the dark energy EOS. In this paper we quantify the bias introduced in the estimation of the dark energy EOS due to weak lensing of supernova flux. We consider the effects due to the non-Gaussian nature of the lensing magnification distributions~\citep{Wang:02}, performing Monte-Carlo simulations by creating mock datasets for future JDEM-like surveys. The paper is organized as follows: In $\S$2.1 we discuss our parameterization of the dark energy EOS, $\S$2.2 discusses gravitational lensing, and $\S$3 is an in-depth description of our methodology. We present our results in $\S$4. | \label{sec:discussion} We first present our results for the magnitude case, as described in \S\ref{sec:mag}. The left panel of Figure~\ref{fig:2} shows histograms of the best-fit values of $w_0$ from the likelihood analysis, after marginalizing over $w_a$ and $h$. The empty histogram, which peaks at -1.009 (marked with a vertical dot-dashed line), is for the model with 300 SNe. The shaded histogram, representing the 2,000 SN case, peaks at -1.003 (vertical dashed line), while the hatched histogram representing the 10,000 SN case has its peak at -1.002 (vertical solid line). These distributions have 1$\sigma$ widths of 0.016, 0.006, and 0.003, respectively. This scatter is primarily due to the intrinsic uncertainty associated with absolute calibration, and is not dominated by lensing. Without the inclusion of lensing, however, the distributions peak at exactly -1, and show no bias. The shifted mode gives us a rough idea of the bias to be expected, on average, due to lensing. We find that 68\% of the time a random sample of 300 SNe will have an estimated value for $w_0$ within 3\% of its fiducial value, and this drops to 0.5\% when a sample size of 10,000 SNe is considered. The right panel of Figure~\ref{fig:2} shows the same distributions as the left panel, but this time using the flux-averaging technique instead of averaging over magnitudes. The empty, shaded, and hatched histograms peak at -1.007, -1.003, and -1.001, respectively, showing the mean bias for the 300, 2,000 and 10,000 SN cases (marked with dot-dashed, dashed, and solid vertical lines). With flux-averaging, we expect that 68\% of the time a random sample of 300 SNe will yield a value of $w_0$ within 2.5\% of the fiducial value, and within 0.5\% for a sample of 10,000 SNe. The 1$\sigma$ parameter uncertainty on $w_0$ ranges from the 20\% level (for 300 SNe) to less than 5\% (for 10,000 SNe), dwarfing the bias due to lensing. Thus, we need not be concerned about lensing degradation of dark energy parameter estimation for future {\it JDEM}-like surveys. We note, however, that our estimated bias on the EOS is larger than the lensing bias of $w<0.001$ quoted in Table 7 of \citet{Wood:07}. This is not surprising, given their use of the simple Gaussian approximation to lensing from \citet{Holz:05}, which is less effective for low statistics. Nonetheless, we agree with their conclusion that lensing is negligible. A similar conclusion was also reached by \citet{Martel:07} who used a compilation of 230 Type Ia SNe \citep{Tonry:03} in the redshift range $0 < z < 1.8$ to show that the lensing errors are small compared to the intrinsic SNe errors. \begin{figure}[!tb] \begin{center} \includegraphics[scale=0.5]{f3.eps} \caption{Histograms showing the distribution of the values obtained for $w_0$ after marginalizing over $w_a$ and $h$ for the 2,000 SN case (using the flux-averaging technique). The shaded histogram assumes that the full sample of 2,000 SNe is used for parameter estimation. The hatched histogram shows the shift when outliers (SNe that are shifted above or below the Hubble diagram by more than 25\% on either side) are removed from the sample. The bias in the distribution is due to the removal of highly-magnified lensing events from the sample.} \label{fig:3} \end{center} \end{figure} We now discuss the bias which arises if anomalous SNe are removed from the sample. Gravitational lensing causes some SNe to be highly magnified, and it is conceivable that these ``obvious'' outliers are subsequently removed from the analysis. In this case the mean of the sample will be shifted away from the true underlying Hubble diagram, and a bias will be introduced in the best-fit parameters. To quantify this effect, we remove SNe which deviate from the expected mean luminosity-distance relation in the Hubble diagram by more than 25\% (corresponding roughly to a $2.5\sigma$ outlier). The SN scatter is a result of the convolution of the intrinsic error (Gaussian in flux of width 0.1) and the lensing PDF, and the outlier cutoff leads to a removal of $\sim$ 50 SNe out of the 2,000 SNe. These outliers are preferentially magnified, due to the strong lensing tail of the magnification distributions. The demagnification tail is cut off by the empty-beam lensing limit, and therefore isn't as prominent. The hatched histogram in Figure~\ref{fig:3} shows the distribution when events with convolved error greater than 2.5$\sigma$ are removed. The vertical dot-dashed line at -1.0075 shows the average value of $w_0$ obtained in this case, representing a bias in the estimate of $w_0$ roughly three times larger than when the full 2,000 SNe are analyzed (shown by shaded histogram). This bias is a result of cutting off the high magnification tail of the distribution, and thus shifting the data towards a net dimming of observed SNe, leading to a more negative value of $w_0$. We also apply a cutoff at $3\sigma$, in addition to the $2.5\sigma$ discussed above. This results in a removal of $\sim20$ SNe on average, for each 2,000 SN sample, and leads to a bias of $\sim0.6$\%. Any arbitrary cut on the (non-Gaussian) convolved (lensing + intrinsic) sample leads to a net bias in the distance relation, and even for large outliers and large SN samples, this can lead to percent-level bias in the best-fit values for $w_0$. To summarize, we have quantified the effect of weak gravitational lensing on the estimation of dark energy EOS from type Ia supernova observations. With generated mock samples of 2,000 SNe distributed uniformly in redshift up to z$\sim$1.7 (as expected in future surveys like {\it JDEM}), we have shown that the bias in parameter estimation due to lensing is less than 1$\%$ (which is well within the 1$\sigma$ uncertainty expected for these missions). Analyzing the data in flux or magnitude does not alter this result. If lensed supernovae that are highly magnified (such that the convolved error is more than 25\% from the underlying Hubble diagram) are systematically removed from the sample, we find that the bias increases by a factor of almost three. Thus, so long as all observed SNe are used in the Hubble diagram, including ones that are highly magnified, the bias due to lensing in the estimate of the dark energy EOS will be significantly less than the 1$\sigma$ uncertainty. Even for a post-{\it JDEM} program with 10,000 SNe, lensing bias can be safely ignored. | 7 | 10 | 0710.4143 |
0710 | 0710.1634_arXiv.txt | We present 107 new epochs of optical monitoring data for the four brightest images of the gravitational lens SDSS J1004+4112 observed between October 2006 and June 2007. Combining this data with the previously obtained light curves, we determine the time delays between images A, B and C. We confirm our previous measurement finding that A leads B by $\Delta t_{BA}=40.6\pm1.8$~days, and find that image C leads image A by $\Delta\tau_{CA}=821.6\pm2.1$ days. The lower limit on the remaining delay is that image D lags image A by $\Delta\tau_{AD}>1250$ days. Based on the microlensing of images A and B we estimate that the accretion disk size at a rest wavelength of 2300\AA\ is $10^{14.8\pm0.3}$~cm for a disk inclination of $\cos{i}=1/2$, which is consistent with the microlensing disk size-black hole mass correlation function given our estimate of the black hole mass from the MgII line width of $\log M_{BH}/M_\odot=8.44\pm0.14$. The long delays allow us to fill in the seasonal gaps and assemble a continuous, densely sampled light curve spanning 5.7 years whose variability implies a structure function with a logarithmic slope of $\gamma = 0.35\pm0.02$. As C is the leading image, sharp features in the C light curve can be intensively studied 2.3 years later in the A/B pair, potentially allowing detailed reverberation mapping studies of a quasar at minimal cost. | The quasar SDSS~J1004+4112 at $z_s=1.734$ is split into five images by an intervening galaxy cluster at $z_l=0.68$ \cite{inada,inada2,oguri}. With a maximum image separation of $14\farcs62$, it is a rare example of a quasar gravitationally lensed by a cluster \cite{wambsganss,inada3}. One of the most interesting applications of this system is to use the time delays between the lensed images to study the structure of the cluster. If we assume the Hubble constant is known, then the delays break the primary model degeneracy of lensing studies (the ``mass sheet degeneracy''), and the delay ratios constrain the structure even if the Hubble constant is unknown. After its discovery, several groups modeled the expected time delays in SDSS J1004+4112 and their dependence on the mean mass profile of the cluster \cite{kawano,oguri,williams}. When we measured the shortest delay in the system, between images A and B, we found a longer delay than predicted by the models (Fohlmeister et al. 2007, hereafter Paper I) where the discrepancy probably arose because the models included the cD galaxy and the cluster halo but neglected the significant perturbations from the member galaxies. As we measure the longer delays, where the cluster potential should be relatively more important than for the merging A/B image pair, we would not expect cluster substructures to play as important a role. We also expect this lens to have a fairly short time scale for microlensing variability created by stars either in the intracluster medium or in galaxies near the images. The internal velocities of a cluster are much higher than in a galaxy (700~km/s versus 200~km/s), and SDSS~J1004+4112's position on the sky is almost orthogonal to the CMB dipole (Kogut et al. 1993), giving the observer a projected motion on the lens plane of almost 300~km/s. In Paper I, we detected microlensing of the continuum emission of the A/B images in Paper I and there is also evidence for microlensing of the CIV broad line \cite{richards,lamer,gomez}. Once we have measured the time delays we can remove the intrinsic quasar variability and use the microlensing variability to estimate the mean stellar mass and stellar surface density, the transverse velocities, and the structure of the quasar source \cite{gilmerino,mortonson,poin,morgan}. Finally, we note that SDSS~J1004+4112 could be an ideal laboratory for studying correlations in the intrinsic variability of quasars. With, image C leading images A and B by 2.3 years, sharp variations in image C can be used to plan intensive monitoring of images A and B to measure the response times as a function of wavelength (e.g. Kaspi et al. 2007), with the additional advantage that the delay between A and B provides redundancies that protect against weather, the Moon and the Sun. The long delays between the images also mean that seasonal gaps are completely filled, and we can examine the structure function of the variability with a densely-sampled, gap-free light curve (modulo corrections for microlensing). Such data generally do not exist, since most time variability data for quasars (other than nearby reverberation mapping targets, e.g. Peterson et al. 2004) have very sparse sampling (e.g. Hawkins 2007 on long time scales for a small number of objects or Vanden Berk et al. 2004 on shorter time scales for many objects). In Paper I \cite{fohli} we presented three years of optical monitoring data for the four brightest images of SDSS J1004+4112 spanning 1000 days from December 2003 to June 2006. The fifth quasar image, E, is too faint to be detected in our observations. We measured the time delay between the A and B image pair to be $\Delta\tau_{BA}=38.4\pm2.0$ days. While larger separation lenses tend to have longer time delays, for these two images the propagation time difference is small, because they form a close image pair ($3\farcs8$) from the source lying close to a fold caustic. For the more widely separated C and D images we could only estimate lower limits on the delays of 560 and 800 days relative to image B and A. In this paper we present the 107 new optical monitoring epochs for the 2006/2007 season in \S2. When combined with our previous data we have light curves spanning 1250 days that allow us to measure the AC delay in \S3. In \S4 we use the microlensing variability of the A/B images to measure the size of the quasar accretion disk, and in \S5 we measure the structure function of the intrinsic variability. We discuss the future prospects for exploiting this system in \S6. \begin{figure*}[t] \centering \includegraphics[bb= 30 100 520 700, width=10cm,angle=0,clip]{f1.ps} \caption{ Light curves of the A, B, C and D images of the quasar SDSS J1004+4112 from December 2003 to June 2007. Images C and D have been offset by 0.3 and 1.0 mag, respectively, in order to avoid overlap. We present a running average of one point every 5 days averaged over $\pm7$ days to emphasize trends and to avoid confusion by noise. \label{lcurve}} \end{figure*} | We present a fourth season of monitoring data for the four bright images of the five image gravitational lens system SDSS J1004+4112. We confirm our previous estimate for the time delay between the merging A/B pair, finding that B leads A by $40.6\pm1.8$ days. We measure the delay for image C for the first time, finding that it leads image A by $821.6\pm2.1$~days. We note that this is nearly twice the longest previously measured delay (the 417 day delay in Q0957+561 \cite{schild, kundic}). We find a lower bound that D lags A by more than approximately 1250~days. Our current mass model predicts that D lags A by approximately 2000 days, which is consistent with the present limit. The fractional uncertainties in the AB delay are still dominated by sampling and microlensing, while the fractional uncertainties in the AC delay are dominated by cosmic variance due to density fluctuations along the line of sight rather than our measurement uncertainties of 0.3\% (e.g. Barkana 1996). A detailed model of this system, including the constraints from the multiply imaged, higher redshift arcs (Sharon et al. 2005), the X-ray measurements \cite{ota,lamer} and a detailed understanding of the uncertainties will be a challenge. We lack a completely satisfactory model for the system at present, in the sense that the modeling process is extraordinarily slow due to the ability of the gravitational potentials associated with the cluster member galaxies to generate additional but undetected images of the quasar, making it impossible to carry out a reliable model survey. The record of models for this system is discouraging. As we noted in Paper I, all three model studies (Oguri et al. 2004; Williams \& Saha 2004; Kawano \& Oguri 2006) generically predicted shorter AB delays than the observed 40 days, and that this could be plausibly explained by the absence of substructure (i.e. galaxies) in the potential models. The longer AB-C and AB-D delays should be less sensitive to substructure. Oguri et al. (2004) do not include an estimate of the AB-C delays and have A-D delays consistent with our present limits. The range of B-C delays in Williams \& Saha (2004) is consistent with our measurement of 820 days, but they predict AD delays shorter than our current lower bound of 1250~days. Kawano \& Oguri (2006) predict a range for the longer delays over a broad range of mass distributions, none of which match our delays in detail. However, models with sufficiently long C-B delays generally have C-D delays long enough to agree with our present limits. Based on our present mass model we used the microlensing between the A and B images to make an estimate of the size of the quasar accretion disk at 2300\AA\ in the quasar rest frame. If we convert this to the expected size at 2500\AA\ assuming the $R_\lambda \propto \lambda^{4/3}$ scaling for a thin disk and assume the mean disk inclination $\cos (i)=1/2$ the scale on which the disk temperature matches the photon energy is $R_{2500\AA}=10^{15.0\pm0.3}$~cm. Comparisons to other disk models should use the half-light radius which is $2.44$ times larger. Based on the quasar MgII emission line width we estimate that the black hole mass is $10^{8.4\pm0.2} M_\odot$. For this mass, the microlensing accretion disk size-black hole mass correlation found by Morgan et al. (2007) predicts that $R_{2500\AA}=10^{15.3}$~cm, which is in broad agreement with the measurement. Further observations, the inclusion of additional images, and monitoring in multiple bands should improve these measurements and potentially allow us to determine the mean surface density in stars near the images $\kappa_*$ and their average mass $\langle M\rangle$. Similarly, the ability to construct continuous light curves of the intrinsic variability and to use image C to provide early warning of sharp flux changes that can then be intensively monitored in images A and B may make this system a good candidate for applying reverberation mapping techniques to a massive, luminous quasar. At present, we already see that the system has a structure function typical of quasars. | 7 | 10 | 0710.1634 |
0710 | 0710.3631_arXiv.txt | We use high-resolution, three-dimensional hydrodynamic simulations to study the hydrodynamic and gravitational interaction between stellar companions embedded within a differentially rotating common envelope. We evaluate the contributions of the nonaxisymmetric gravitational tides and ram pressure forces to the drag force and, hence, to the dissipation rate and the mass accumulated onto the stellar companion. We find that the gravitational drag dominates the hydrodynamic drag during the inspiral phase, implying that a simple prescription based on a gravitational capture radius significantly underestimates the dissipation rate and overestimates the inspiral decay timescale. Although the mass accretion rate fluctuates significantly, we observe a secular trend leading to an effective rate that is significantly less than the rate based on a gravitational capture radius. We discuss the implications of these results within the context of accretion by compact objects in the common-envelope phase. | To understand the evolution of binary star systems, it is essential to analyze the interactions between their stellar components. Examples of such influences include the spin-orbit tidal interaction and mass transfer, as well as interactions that result in the loss of mass and angular momentum. Equally important are the interactions of stars orbiting about their common center of mass within a differentially rotating common envelope. It is generally accepted that such an evolutionary stage is essential for the formation of short-period binary systems containing compact objects (see, e.g., Iben \& Livio 1993 and Taam \& Sandquist 2000). In this case, the interaction determines the orbital evolution of the system and the conditions under which the common envelope is ejected, leading to the survival of a remnant binary system or to a merger that forms a rapidly rotating single star. The amount of mass and angular momentum accreted by the inspiralling components during this phase also has direct implications for the properties of the compact object population in binary systems. Lacking multidimensional hydrodynamical simulations of the common-envelope phase, the initial numerical and semi-analytical studies of the problem used simple prescriptions for the stellar interactions based on the pioneering work by Hoyle \& Lyttleton (1939) and Bondi \& Hoyle (1944), as generalized by Bondi (1952). These seminal studies focused on the idealized problem of the capture of matter by a gravitating point object moving supersonically with respect to a uniform medium. In this framework, a gravitational capture radius, $\rcap$, plays an important role in determining the rates of mass accretion and energy dissipation. $\rcap$ is given by \begin{equation} \label{Eqn:rcap} \rcap = {2 G M \over \vrel^2 + \cs^2}\ , \end{equation} where $M$ is the mass of the gravitating object, $\vrel$ is the velocity of the object with respect to the medium, and $\cs$ is the local speed of sound. When a density gradient with scale height $H$ is present, the effective accretion radius $\racc$ is (Dodd \& McCrea 1952) \begin{equation} \label{Eqn:racc} \racc = {\rcap \over 1 + (\rcap/2H)^2}\ . \end{equation} The energy dissipation rate is then $L_{\rm d} \approx \pi \racc^2 \rho \vrel^3$, where $\rho$ is the upstream density. To improve on these estimates, hydrodynamic effects were approximated analytically by Ruderman \& Spiegel (1971), Wolfson (1977), and Bisnovatyi-Kogan et al.\ (1979) as well as numerically by Hunt (1971, 1979), Shara \& Shaviv (1980), and Shima et al.\ (1985). These early multidimensional simulations considered axisymmetric flow, and their results have been used to calibrate the energy loss rate. In particular, the drag coefficients obtained from such simulations (see, e.g., Shima et al.\ 1985) have been used to estimate the rate of energy dissipation in the common envelope. However, many of the simplifying assumptions underlying these studies are inadequate for direct application to common-envelope interactions. The flow is nonaxisymmetric and distinctly nonuniform, reflecting the existence of velocity or density gradients (the density may span several scale heights within $\racc$). The effect of relaxing these assumptions has been studied in two dimensions by Fryxell \& Taam (1988) and Taam \& Fryxell (1989) and in three dimensions by Sawada et al.\ (1989) and Ruffert (1999). These studies could not encompass the full complexity of common-envelope interactions, since the envelope's self-gravity was ignored. Furthermore, because the companions move in elliptical orbits, their cores interact with matter that has already been affected in previous orbital phases. Thus, the state of the gas and its environment in these calculations must be regarded as highly idealized. Within the past decade, three-dimensional numerical studies of the common-envelope phase that have relaxed earlier geometrical assumptions have been carried out by Sandquist et al.\ (1998, 2000), DeMarco et al.\ (2003a,b), and Taam \& Ricker (2006). Recently, we have carried out high-resolution adaptive mesh refinement (AMR) simulations of common-envelope evolution with effective resolutions of $2048^3$ (R. E. Taam \& P. M. Ricker, in preparation), allowing the interaction of the stars within the common envelope to be examined and quantified. In this Letter we report on some results of our numerical studies. We focus on analyzing a single high-resolution simulation to determine the hydrodynamic and gravitational contributions to the drag forces affecting the orbital motion of the stellar components during the early inspiral phase. We also analyze the accumulation of mass by the stellar components within the common envelope to compare their magnitudes to estimates based on an accretion radius formalism. In \S~2, we briefly describe the numerical method and our assumed model for the binary system. Descriptions of the method of analysis and the numerical results are presented in \S~3. Finally, we summarize our results and comment on their possible implications for applications involving compact objects in short-period binary systems. | We have quantitatively described the interaction of stars within a common envelope on the basis of the analysis of the early inspiral stage of a $1.05 \msun$ red giant and a $0.6 \msun$ binary companion. The orbital decay is dominated by the nonaxisymmetric gravitational drag associated with the self-gravitating matter in the common envelope. On the basis of high-resolution three-dimensional hydrodynamic simulations, this drag is 1-2 orders of magnitude greater than the hydrodynamic drag. As a consequence, the orbital decay timescale is much shorter than that derived from analyses based on the Hoyle-Lyttleton-Bondi picture of accretion from a uniform medium by a gravitating point mass moving at supersonic speeds. In this latter description the gravitational drag associated with an accretion wake is not significantly larger than the hydrodynamic drag. The effect of long-range gravitational interactions is critical for reliable estimates of the orbital decay timescale and energy dissipation rate. In this picture the drag and the mass accretion rate are not as directly related as they are in the Hoyle-Lyttleton-Bondi-type description, because the dominant drag term is gravitational, rather than hydrodynamic, in origin. This decoupling is reflected in the much larger ratio $L_{\rm d} / \dot{M} \sim 10^{15}-10^{16}$~cm$^2$~s$^{-2}$ measured in the simulation than expected from the gravitational capture radius formalism ($\sim 10^{14}$~cm$^2$~s$^{-2}$). Although the mass accretion rates estimated from our simulation should be regarded as only indicative due to the lack of a detailed inner boundary treatment for the companion, the structure of the flow (dominated as it is by tidal effects) strongly suggests that the true mass accretion rate should be much smaller than the rate expected in the gravitational capture radius formalism. The effective capture radius, on the basis of the observed ambient density and relative velocity, is almost an order of magnitude smaller than the expected value. In any case, the expected value would lead to an unrealistic level of accretion over the common-envelope period. Furthermore, the inspiral time suggested by the early evolution of our three-dimensional simulation is much shorter than that found in earlier one- and two-dimensional calculations that are based on the Bondi-Hoyle-Lyttleton picture. Assuming that the discrepancy in mass accretion rate continues into the deep inspiral phase, the total accumulation of matter onto the companion should be much smaller than previously expected. These results would have little effect on the mass of an embedded main-sequence star because of its tendency to expand as a result of the high entropy within the common envelope (see Hjellming \& Taam 1991); however, it can significantly affect the outcome for neutron stars within a common envelope. In particular, the estimated accretion rates exceed $10^{-3} \mpy$, for which steady state accretion flows with neutrino losses are possible (Chevalier 1989; Houck \& Chevalier 1991). At these hypercritical mass accretion rates, photons are trapped in the flow and the Eddington limit is not applicable. On the basis of this hypercritical accretion flow regime, Chevalier (1993) and Brown (1995) suggest that neutron stars embedded in the common envelope would accrete sufficient mass to form low-mass black holes (although see Chevalier 1996), and, hence, the formation of binary radio pulsars would require an evolutionary scenario involving progenitor stars of nearly equal mass (Brown 1995). However, the population synthesis of binary black holes and neutron stars by Belczynski et al.\ (2002), including hypercritical accretion, resulted in an average accretion of $0.4 \msun$. Such a high rate of mass accretion is inconsistent with the observed masses of binary radio pulsars ($\sim 1.35 \msun$; Thorsett \& Chakrabarty 1999) and indicates the need for a reduction in accreted matter during the common-envelope phase (see also Belczynski et al.\ 2007). Our calculations show that the necessary reduction may arise naturally as a result of a more realistic treatment of the common-envelope phase. Consequently, this reduction could also lead to a reduction in the number of low-mass black holes, depending on the maximum mass of neutron stars, resulting from the accretion induced collapse of massive accreting neutron stars in the common-envelope phase. Similarly, the mass accretion, which was found to be as large as several solar masses for black hole accretors, would also be reduced, thereby affecting the masses and spins of double black holes emerging from the common-envelope phase. Further investigations are planned to examine the generality of these results regarding mass accretion and to quantify the importance of these processes for determining the properties (mass and spin) and ultimate fate of the compact components in short-period binary system populations. Such studies are not only important for determining the masses of binary neutron star and black hole systems resulting from the common-envelope phase (Belczynski et al.\ 2007), but also their orbital periods, which directly influence the expected merger rates of such binary populations as sources for gravitational wave detection in the advanced LIGO experiment. | 7 | 10 | 0710.3631 |
0710 | 0710.3588_arXiv.txt | We assess the relative merits of weak lensing surveys, using overlapping imaging data from the ground-based Subaru telescope and the Hubble Space Telescope (HST). Our tests complement similar studies undertaken with simulated data. From observations of 230,000 matched objects in the 2 square degree COSMOS field, we identify the limit at which faint galaxy shapes can be reliably measured from the ground. Our ground-based shear catalog achieves sub-percent calibration bias compared to high resolution space-based data, for galaxies brighter than $i^{\prime}\simeq$24.5 and with half-light radii larger than $1.8\arcsec$. This selection corresponds to a surface density of 15 galaxies arcmin$^{-2}$ compared to $\sim 71$ arcmin$^{-2}$ from space. On the other hand the survey speed of current ground-based facilities is much faster than that of HST, although this gain is mitigated by the increased depth of space-based imaging desirable for tomographic (3D) analyses. As an independent experiment, we also reconstruct the projected mass distribution in the COSMOS field using both data sets, and compare the derived cluster catalogs with those from $X$-ray observations. The ground-based catalog achieves a reasonable degree of completeness, with minimal contamination and no detected bias, for massive clusters at redshifts $0.2<z<0.5$. The space-based data provide improved precision and a greater sensitivity to clusters of lower mass or at higher redshift. | \label{sec:Introduction} Dark matter dominates the gravitational component of the cosmic energy density and thus provides the framework for structure formation in the Universe. However, by its nature, the distribution and cosmic growth are challenging to observe. The most promising probe is weak gravitational lensing: analysis of the distorted shapes of ordinary galaxies behind foreground mass concentrations. Several numerical techniques are now available to recover the projected mass distribution from these distortions, and tests on simulated datasets are underway to verify their precision \citep{step1,step2}. There is great optimism in the weak lensing community that such methods will enable both the tomographic mapping of dark matter structures in time and space. This will also provide a robust statistical measure of the nature of dark energy over redshifts 0$<z<$1 \citep{mellier99, refregier03} Observational progress has been particularly dramatic. The first detections of statistical ``cosmic shear'' were only published in 2000 \citep{bacon00, kaiser00, wittman00, vanwaerbeke00}. In the subsequent \com{seven} years, weak lensing surveys have measured the dark matter power spectrum \citep{brown03, heymans05, hoekstra06, sembolini06}, traced the evolution of structure \citep{bacon05, kitching06, massey07a}, enabled the construction of lensing-selected cluster catalogs \citep{miyazaki02a, wittman06, schirmer07, miyazaki07}, and non-parametrically reconstructed the total mass distribution both in clusters \citep{kneib03, clowe06, jee07} and on larger scales \citep{massey07b}. As a result, weak lensing has been identified as the most promising route to understanding the nature of dark energy by the ESA-ESO Working Group on Fundamental Cosmology\footnote{\tt http://www.stecf.org/coordination/esa\_eso/cosmology.php}, joint NSF-NASA-DOE Astronomy and Astrophysics Advisory Committee\footnote{\tt http://www.nsf.gov/mps/ast/aaac.jsp}, and NSF-DOE High Energy Physics Advisory Panel Dark Energy Task Force\footnote{\tt http://www.nsf.gov/mps/ast/detf.jsp}. The primary signal of any weak lensing analysis is the statistically coherent distortion of background galaxies along adjacent lines of sight. The main sources of statistical noise are the finite density of galaxies that can be sufficiently well-detected and resolved for accurate shape measurement, plus their intrinsic morphologies. The density of resolved galaxies also governs the angular resolution and fidelity of a reconstructed mass map which, in turn, determines the limiting halo mass that can be detected. On the other hand, statistical analyses of the dark matter power spectrum are less concerned with individual halos but require panoramic fields to counter the effects of cosmic (sample) variance. Minimizing statistical errors in such an analysis, within a finite survey lifetime, requires an optimal balance between area and depth. A key debate in the development of future weak lensing experiments concerns the relative merits of ground- versus space-based platforms. Ambitious surveys now being planned with dedicated, ground-based facilities (eg VST-KIDS, DES, Pan-STARRS, LSST). These are driven by technological progress including panoramic cameras with small optical distortions, highly sensitive imaging detectors, and (in the case of Pan-STARRS) on-chip active correction to reduce the width of the point spread function (PSF). Future surveys spanning significant fractions of the celestial sphere are envisaged, promising tight constraints on the cosmological parameters. However, measurements with current ground-based facilities are limited by the size and temporal variations of the PSF. There is concern in many quarters that wide-field facilities operating in space (e.g.\ DUNE, SNAP, JDEM) will ultimately be required to achieve the precision required (particularly) to distinguish between various models of dark energy. Space-based facilities will be more costly but will likely offer increased depth, better photometric performance and a stable PSF. The key issue in gauging their merits is not statistical error, but the extent to which potential biases in ground-based data may act as a ``systematic floor'' to prevent complete exploitation. Some valuable answers can be obtained by comparing simulated ground and space-based images, \citep{wittman05,lampton06} and the Shear TEsting Programme \citep[STEP:][]{step1,step2}. However, the input parameters used to generate the simulated data may not be realistic or address all the instrumental idiosyncrasies. Of particular concern are the stability and vagaries of the PSF. No simulations have yet adequately addressed this point -- which may, ultimately, be the limiting problem for ground-based data. It is often argued that future facilities will be carefully designed to mitigate any limitations realized with current observational facilities. While progress can no doubt be expected, both on the ground and in space, we believe many lessons can be learned from extant data and hardware with proven engineering pedigree. In this paper, we present the first direct comparison of weak lensing analysis {\it for the same sky field} using ground and space-based data. Deep, panoramic imaging has been obtained for the 1.64 deg$^2$ COSMOS field \citep{scoville07a} by both the Advanced Camera for Surveys ({\it ACS}) on board the Hubble Space Telescope (HST) \citep{scoville07b} and the {\it Suprime-Cam} imager at the prime focus of the Subaru 8.2m telescope \citep{taniguchi07}. In both cases, the entire field was covered by mosaicing many independent exposures. The SuPrimeCam instrument was constructed with weak lensing analysis particularly in mind, and currently provides the best image performance available from any ground-based telescope, in terms of optical distortions over a large field. A comparison of these datasets should therefore provide a realistic and valuable assessment of the relative performance of state-of-the-art imagers on the ground and in space. The paper is organized as follows. In \S\ref{sec:theory}, we briefly review the relevant theory. In \S\ref{sec:data}, we describe the two data sets, data reduction pipelines and weak lensing analyses. We then present the results. In \S\ref{sec:analyses}, we compare shear measures on a galaxy-by-galaxy basis to determine the optimum depth at which the ground-based data matches the performance of the (deeper) space based data. This permits us to determine the relative survey speeds of Subaru and HST for high precision cosmic shear experiments. In \S\ref{sec:Maps}, we construct maps of the mass distribution, treating the Subaru and HST maps as independent probes of the same field, and contrast these against X-ray data. This permits us to evaluate the completeness and reliability of a lensing-selected halo catalog, and evaluate the precision of their inferred masses as a function of redshift. In \S\ref{sec:conc}, we summarize our results and discuss their wider implications for future missions. | \label{sec:conc} We have performed parallel weak lensing analyses of Subaru and Hubble Space Telescope imaging in the COSMOS field. Our comparisons of the observed shear and convergence signals have revealed a number of issues, and suggest that such a study with real data usefully complements the independent approach based on blind analyses of simulated data \citep{step1, step2}. For statistical ``cosmic shear'' analyses, shear measurement with an existing ground based telescope, using existing measurement techniques, can be achieved with less than 1\% bias relative to higher resolution space-based data, for a galaxy surface density of 15 arcmin$^{-2}$. One limitation of our approach is that we cannot check the performance of our space-based analysis on the additional, small galaxies. At first sight the factor of $\sim$3 shortfall in surface density seems inconsequential given the lower cost and improved areal mapping speed of existing ground-based cameras such as SuPrime-Cam. However, accompanying the brighter Subaru limit is a reduction in survey depth and hence the redshift distribution of background sources. More distant sources contain a larger signal, and a narrower range of redshifts also hinders tomographical tests \citep{bacon05,massey07b}, which tighten cosmological parameter constraints significantly. A key issue is whether this limiting depth is a fundamental one for all future ground-based cameras. PanSTARRS, VST, and even LSST each have significantly smaller primary mirrors than Subaru, so achieving even the $S/N$ discussed here would require formidable exposure times. Most importantly, the deep infrared imaging that is required for photometric redshifts to enable tomographic analyses is likely to be difficult over large survey fields from the ground, because of increased sky background. Recent weak lensing analyses are limited at roughly the same level by uncertainty in galaxy shape measurement and photometric redshift estimation. Statistical measurements from the ground are also hindered by variable atmospheric seeing. Past experience has taught the authors that data collected in seeing worse than $0.8\arcsec$ is of little use for weak lensing analysis. The apparently rapid speed of data collection for our Subaru data belies the time spent waiting for better seeing, even with the excellent atmospheric conditions above Mauna Kea and the well-controlled dome seeing of Subaru. For this small-scale survey, we obtained exceptional quality imaging during a fortuitous observing window. The relevant quantity for larger-scale surveys in the future will be the time-averaged seeing quality, and the fraction of time spent with seeing better than $0.8\arcsec$. This is particularly true for surveys like Pan-STARRS and LSST, that plan to adopt a strategy of co-adding many shorter exposures. Their advantage is that the stacked images will achieve a near-uniform image quality, by virtue of the independent PSFs in each short exposure. This can then be tuned to the required image quality by rejecting a certain fraction of exposures. Variable seeing conditions is also of concern for the reconstruction of mass maps (c.f.\ Green et al.\ {\it in prep.}). Difficulties in the analysis of one pointing in the SW corner of the Subaru map result in a patchy recovery of large-scale structure; with \com{more noise and a lower range of probed redshift} in certain regions. Most importantly, \com{however, the four most massive clusters out of eight detected from space are also detected from the ground -- with one intriguing additional signal and two more confirmed clusters just ouside the field of view observed from space. Reassuringly, the three clusters conservatively deemed ``secure'' by the independent analysis of \citep{miyazaki07} have now been confirmed via space-based weak lensing and $X$-ray observations. The physical properties of the four massive halos in common (A, B and C) are remarkably consistent whether derived from from ground- or space-based weak lensing. The measured masses and radial profiles of these clusters are consistent and, for the lower redshift clusters, the error bars are comparable.} A Dark Energy Task Force ``Stage 3'' survey from the ground appears eminently feasible. A wide-field space-based platform would open up many new applications. Very important for statistical applications is the increased redshift range of resolved background galaxies. These not only contain a larger shear signal, but also more readily split into redshift bins for tomographic analysis. Three-dimensional analysis techniques will tighten constraints on cosmological parameters by factors of $3-5$, and directly measure quantities that depend upon the properties of dark energy, like the growth of structure over cosmic time and the redshift-distance relation. They will also eliminate sources of error due to the intrinsic correlations of galaxy shapes. Sufficiently good photometric redshifts require deep, wide-field near-IR imaging, and these are also realistically possible over large surveys only from space. Recent weak lensing analyses with relatively shallow near-IR coverage like \citep{massey07a} are limited to roughly the same degree by uncertainty in galaxy shape measurement and photometric redshift estimation. Full implementations of cross-correlation cosmography will almost certainly require deep near-IR imaging from space. Such advanced techniques will become particularly important as ground-based surveys expand to encompass the entire observable sky. The increased surface density of galaxies resolved from space also improves maps of the mass distribution. \com{As shown in figures~\ref{emode} and \ref{bmode}, the statistical noise and systematic contamination in the $B$-mode are significantly reduced. Eight clusters are detected in the COSMOS $E$-mode signal without any contamination from spurious peaks. With the increased mass and spatial resolution of mass reconstructions from space, it becomes possible to detect halos the size of galaxy groups, as well as clusters over a wide range of redshifts -- thus tracing their formation,} which is governed by the properties of dark matter and the nature of gravity. Space-based data also crosses the threshold to mapping even filamentary large-scale structure in three dimensions \citep{massey07b}. Obtaining the detailed, 3D distribution of mass will be particularly important near regions of interest like the Bullet cluster \citep{clowe06}, where the small differences between the location of mass and baryons in a small patch of sky may yield the best possible information about the properties of dark matter. In this and other astrophysical phenomena, knowledge of the local mass environment and nearby large-scale structure is critical. Overall, we conclude that ground-based weak lensing surveys can perform several tasks remarkably well, with sufficiently small amount of systematic bias to easily justify the next generation of dedicated ground-based surveys. Two dimensional statistical analyses will be able to produce order-of-magnitude improvements in weak lensing constraints, using proven hardware technology and software pipelines. On the other hand, a wide-field space-based imager would provide important control over some systematic effects, and open up many new applications that are, at least currently, unachievable from the ground. For several of the most exciting techniques that will directly probe the nature of dark matter and dark energy, eventual space-based imaging is likely to be essential. | 7 | 10 | 0710.3588 |
0710 | 0710.3541_arXiv.txt | The massive OB-type binary $\sigma$~Ori~AB is in the centre of the very young $\sigma$~Orionis cluster. I~have computed the most probable distances and masses of the binary for several ages using a dynamical parallax-like method. It incorporates the $BVRIH$-band apparent magnitudes of both components, precise orbital parameters, interstellar extinction and a widely used grid of stellar models from the literature, the Kepler's third law and a $\chi^2$ minimisation. The derived distance is 334$^{+25}_{-22}$\,pc for an age of 3$\pm$2\,Ma; larger ages and distances are unlikely. The masses of the primary and the secondary lie on the approximate intervals 16--20 and 10--12\,$M_\odot$, respectively. I~also discuss the possibility of $\sigma$~Ori~AB being a triple system at $\sim$385\,pc. These results will help to constrain the properties of young stars and substellar objects in the $\sigma$~Orionis~cluster. | \label{introduction} The Trapezium-like system {$\sigma$~Ori}, that illuminates the encolure of the {Horsehead Nebula}, is the fourth brightest star in the young Ori~OB~1~b association. The multiple system is composed of at least five early-type stars (Burnham 1892; Greenstein \& Wallerstein 1958; van Loon \& Oliveira 2003; Caballero 2007b). The two hottest components, $\sigma$~Ori~A and B (O9.5V and B0.5V), are separated by only $\sim$0.25\,arcsec and were for a long time ``the most massive visual binary known'' (${\mathcal M}_{\rm A} + {\mathcal M}_{\rm B}\sim 25+15~M_\odot$; Heintz 1974). Although the binary has not yet completed a whole revolution, the orbital parameters are relatively well determined (Hartkopf, Mason \& McCalister 1996; Heintz 1997; Horch et~al. 2002). It has been suggested that $\sigma$~Ori~AB is a hyerarchical triple, being the primary a short-period, double-line spectroscopic binary (Frost \& Adams 1904; Henroteau 1921; Miczaika 1950; Bolton 1974; Morrell \& Levato 1991). However, a large amount of accurate, comprehensive spectroscopic investigations have failed to confirm this hypothesis (Heard 1949; Conti \& Leep 1974; Humphreys 1978; Bohannan \& Garmany 1978; Garmany, Conti \& Massey 1980; Sim\'on-D\'{\i}az \& Lennon, priv.~comm.). The $\sigma$~Ori system is located in the centre of the well-known $\sigma$~Orionis open cluster. The proper motions, radial velocities and spacial distribution of stars in this cluster strongly suggests a physical association between $\sigma$~Ori itself and the young cluster (Zapatero Osorio et~al. 2002a; Caballero 2007a, 2007c -- see also Jeffries et al. [2006], who discovered a second older and kinematically and spacially distinct population). Because of its youth, comparative nearness and low extinction, the cluster has become the richest hunting ground for brown dwarfs and planetary-mass objects in the whole sky (B\'ejar et~al. 1999; Zapatero Osorio et~al. 2000, 2002b, 2007). There is plentiful material in the literature about this cluster, covering topics like the initial mass function down to a few Jupiter masses (B\'ejar et~al. 2001; Gonz\'alez-Garc\'{\i}a et~al. 2006; Caballero et~al. 2007), jets and Herbig-Haro objects (Reipurth et~al. 1998; Andrews et~al. 2004), the frequency of accretors and discs (Zapatero Osorio et~al. 2002a; Oliveira, Jeffries \& van Loon 2004; Kenyon et~al. 2005; Oliveira et~al. 2006; Hern\'andez et~al. 2007; Caballero et~al. 2007), the X-ray emission from young objects (Walter et~al. 1997; Sanz-Forcada et~al. 2004; Franciosini, Pallavicini \& Sanz-Forcada 2006) or their photometric variability (Caballero et~al. 2004; Scholz \& Eisl\"offel 2004). The most used values of heliocentric distance and age of the $\sigma$~Orionis cluster are $d \sim$ 360\,pc and $\sim$3\,Ma (Brown, de Geus \& de Zeeuw 1994; Perryman et~al. 1997; Zapatero Osorio et~al. 2002a; Oliveira et~al. 2002). There is a consensus in the literature that the cluster is younger than 8\,Ma and older than 1\,Ma. There is, however, a strong divergence of opinion on the heliocentric distance. Caballero (2007a) compiled determinations in the literature of the distance to the $\sigma$~Orionis cluster from the 352$^{+166}_{-168}$\,pc from {\em Hipparcos} parallax to almost 500\,pc from colour-magnitude diagrams. Apart from the uncertainties of theoretical isochrones at very young ages, the derivation of the initial mass function of the cluster is strongly affected by the uncertainty in the actual age and heliocentric distance (Jeffries et~al. 2006; Caballero et~al. 2007). There are other investigations that require a precise age determination, such as the evolution of the angular momentum due to discs and stellar winds (Eisl\"offel \& Scholz 2007), disc dissipation (Hern\'andez et~al. 2007) and evolution of hot massive stars. % In this Letter, I~revisit a well known method for distance determination: the dynamical parallax (e.g. Russell 1928). I~apply it to the $\sigma$~Ori~AB binary using state-of-the-art data and tools to determine its mass, age and heliocentric~distance. | The theoretical effective temperatures that correspond to the optimal fits lie on the intervals $T_{\rm eff}$ = 30.4--34.6\,kK for the primary and $T_{\rm eff}$ = 25.2--27.5\,kK for the secondary. The hottest temperatures are for the youngest ages. These values are consistent with the expected $T_{\rm eff}$ for O9.5V and B0.5V stars, respectively (e.g. O9.5V: 30--35\,kK -- Popper 1980; Gulati, Malagnini \& Morossi 1989; Castelli 1991; Vacca, Garmany \& Shull 1996; Martins, Schaerer \& Hillier 2005), and with previous measurements of the $T_{\rm eff}$ of $\sigma$~Ori~A (30.0--33.0\,kK -- Morrison 1975; Underhill et~al. 1979; Morossi \& Crivellari 1980; Repolust et~al. 2005). The corresponding theoretical gravities ($\log{g} \sim$ 4.00--4.22) are also normal for class V at such temperatures. The minima of $\chi^2$ in Table~\ref{bestfits} are very sensitive to the variations of heliocentric distance and of mass: on the one hand, at fixed mass and age, a fluctuation of $d$ of barely 30\,pc results in a change of three orders of magnitude in $\chi^2$; on the other hand, at fixed distance and age, a fluctuation of ${\mathcal M}_{\rm A}$ of barely 5\,$M_\odot$ results in a change of almost two orders of magnitude in $\chi^2$. The minima of $\chi^2$ are, however, quite unresponsive to the variations of the age between 1.0 and 4.9\,Ma (see last column in Table~\ref{bestfits}). A younger age gives a slightly better fit results and that favours a slightly larger distance ($d \sim$ 350\,pc). The results do not strongly suggest a younger age of 1\,Ma, but they are useful in excluding an older age. The absence of a solution at 10\,Ma agrees with previous upper limits on the ages of the {Ori~OB~1~b} association from the presence of very early-type stars in the main sequence (Blaauw 1964) and of the $\sigma$~Orionis cluster from spectral synthesis surrounding the Li~{\sc i} $\lambda$670.78\,nm line (Zapatero Osorio et~al.~2002a). The derived distance interval for 3$\pm$2\,Ma, $d$ = 334$^{+25}_{-22}$\,pc, is consistent with the canonical distance to the $\sigma$~Orionis cluster of $d \sim$ 360\,pc, but is difficult to conciliate with the distance of 440\,pc for 2.5\,Ma that Sherry, Walter \& Wolk (2004) used. The derived distance interval also deviates from very recent determinations of the distance to some elements in the {Ori~OB~1} complex. In particular, Terrell, Munari \& Siviero (2007), using the eclipsing spectroscopic binary {VV~Ori} close to {Alnilam} ($\epsilon$~Ori) in Ori~OB~1~b, and Sandstrom et~al. (2007), employing the Very Large Baseline Array in the {Orion Nebula Cluster} in {Ori~OB~1~a}, have determined very accurate heliocentric distances of 388--389\,pc (see also Menten et~al. 2007). These values are also lower than the classical distance to the {Ori~OB~1} complex of $\sim$440\,pc from average {\em Hipparcos} parallax (Brown, Walter \& Blaauw 1999; de Zeeuw et~al. 1999). Because of projection effects and the large physical size of Ori~OB~1 (Reynolds \& Ogden 1979), $\sigma$~Ori could be easily contained within the complex. Given their kinematic and spacial association, the $\sigma$~Ori system and the young $\sigma$~Orionis cluster are likely at the same heliocentric distance and also age, if one assumes that massive and low mass star formation in a cluster is coeval (Prosser et~al. 1994; Massey, Johnson \& Degioia-Eastwood 1995; Stauffer et~al. 1997; see, however, Sacco et~al. 2007). Recently, it has been suggested that the $\sigma$~Orionis cluster is actually kinematically distinct from the {Ori~OB~1~b} association (Jeffries et~al. 2006), just as {25~Ori} is distinct from Ori~OB~1~a (Brice\~no et~al.~2007). If $\sigma$~Ori~AB were a hyerarchical triple, as described in Section~\ref{introduction}, it would be located at a larger heliocentric distance. The hypothetical companion to $\sigma$~Ori~A, to which I~tentatively call {$\sigma$~Ori~F}, would be 0.5\,mag fainter than the primary in the 370--493\,nm interval according to Bolton (1974). This wavelength interval corresponds to the $U,~B$ Johnson bands. Taking into account $m_{\rm A} = m_{\rm A+F} + 2.5 \log{\left( 1 + 10^{-\frac{m_{\rm A}-m_{\rm F}}{2.5}} \right)}$ and the difference $B_{\rm A+F}-B_{\rm B}$ in Table~\ref{photometry}, then the apparent magnitudes in the blue band of the three components would be related to the combined magnitude $B_{\rm A+F}$ through: $B_{\rm A} \approx B_{\rm A+F} + 0.53$\,mag, $B_{\rm B} \approx B_{\rm A+F} + 1.33$\,mag and $B_{\rm F} \approx B_{\rm A+F} + 1.03$\,mag. I~have looked for the distances and theoretical masses whose corresponding apparent magnitudes match the $B_{\rm [A,B,F]}$ relations and the Kepler's third law, $({\mathcal M}_{\rm A} + {\mathcal M}_{\rm F}) + {\mathcal M}_{\rm B} = 7.45~10^{-7} d^3$. There are only a few solutions that simultaneously verify $T_{\rm eff,A} \approx$ 30.0--33.0\,kK and $T_{\rm eff,B} \approx$ 26.0--30.0\,kK, as expected from the spectral types of $\sigma$~Ori~A+F and $\sigma$~Ori~B. In the triple scenario, the F component would have an intermediate temperature between the A and B stars (roughly B0.0V) and would orbit very close to $\sigma$~Ori~A. The valid solutions are found for the narrow distance interval 370--400\,pc and only for ages between 1.0 and 4.9\,Ma. Although distances less than 290\,pc and larger than 450\,pc are ruled out again, the most probable distance to $\sigma$~Ori under the triple hypothesis, $d \sim$ 385\,pc, agrees very well with those of VV~Ori and the Orion Nebula Cluster. Further high-resolution spectroscopic studies are needed to ascertain the existence and characteristics of $\sigma$~Ori~F. | 7 | 10 | 0710.3541 |
0710 | 0710.1896_arXiv.txt | We present an analysis of five X-ray Multi-Mirror Mission (\xmm) observations of the anomalous X-ray pulsar (AXP) 1E 2259+586 taken in 2004 and 2005 during its relaxation following its 2002 outburst. We compare these data with those of five previous \xmm observations taken in 2002 and 2003, and find the observed flux decay is well described by a power law of index $-0.69\pm0.03$. As of mid-2005, the source may still have been brighter than preoutburst, and was certainly hotter. We find a strong correlation between hardness and flux, as seen in other AXPs. We discuss the implications of these results for the magnetar model. | \label{sec:intr} It is now commonly believed that soft gamma-ray repeaters (SGRs) and anomalous X-ray pulsars (AXPs) are neutron stars with ultra-strong magnetic fields, i.e. magnetars \citep{dt92a}. Their common nature was conclusively demonstrated when AXP \tfe was observed to emit SGR-like bursts in 2001 \citep{gkw02} and \ttfn, in the supernova remnant (SNR) CTB 109, was seen to undergo a major SGR-like outburst in 2002 \citep{kgw+03,wkt+04}. Subsequently, a variety of different types of activity in AXPs have been seen, including short- and long-term flux variations \citep{gk04,dkg07} and slow and rapid pulse profile changes \citep{ikh92,kgw+03,wkt+04,icd+07,dkg07,dkg07b}, in addition to bursts and outbursts (\citealp{gkw06,wkg+05,dkg07}; see \citealt{kas07} for a recent review). During \ttfn's 2002 outburst, the pulsed and persistent fluxes rose suddenly by a factor of $\ge$20 and decayed on a timescale of months. Coincident with the X-ray brightening, the pulsar suffered a large glitch of fractional frequency change $4\times10^{-6}$ \citep{kgw+03,wkt+04}. In the first few hours of the outburst, the pulsar's pulse profile changed significantly, its pulsed fraction decreased, and its spectrum hardened dramatically. Over 80 short SGR-like bursts from the pulsar were observed at the same time \citep{gkw04}. A near-infrared (K$_s$) enhancement was also observed during the epoch of the outburst \citep{kgw+03}. Combining {\it Rossi X-Ray Timing Explorer } (\xte) observations and \xmm observations of \ttfn\, taken during and after the outburst, \cite{wkt+04} found that the decay of \ttfn's unabsorbed flux (mostly inferred from \xte pulsed fluxes) after the outburst was well characterized by two power law components: a rapid steep decay visible only during the first several hours ($<1$ day) of the outburst, and a slower decay of index $-0.22$ for the next several months. \cite{tkvd04} found that the near-infrared enhancement at late times decayed at the same rate as the slow X-ray decay, although there were no IR observations during the first few hours of the outburst. Other AXPs have also exhibited transient behavior that could be explained by SGR-like outbursts. AXP \etten is called transient because it was only discovered in 2003 when it suddenly became brighter by a factor of 100 \citep{ims+04,ghbb04}. \cite{gh07} found that the flux of \etten after 2003 followed an exponential decay of timescale 233.5 days. Similarly, the AXP CXOU~J164710.2$-$455216 was found to have brightened by a factor of $\sim$300 between two \xmm observations taken 5 days apart in 2006 September \citep{icd+07}. Candidate AXP \etff, was discovered in an observation made in 1993 by {\it ASCA} \citep{gv98,tkk+98}. Follow-up observations in 1999 showed that the source's flux was smaller by a factor of $\sim$10 \citep{vgtg00}. \cite{tkgg06} found that \etff\, remains undetected in \chandra observations taken in 2003, with its flux $\sim $260-430 times fainter than observed in 1993. The transient AXP phenomena summarized above are qualitatively similar to the 1998 August 27 flare of \ninet, in which the X-ray flux decayed with a power law of index $\sim -0.9$ \citep{fmw+03}, and the flux decay of SGR 1627$-$41 since 1998, which followed a power law of index $\sim -0.47 $ and lasted for $\sim$800 days \citep{kew+03}. However, thus far, the AXP outbursts have been much less energetic than most SGR outbursts. Also, most of the burst energy was released during the afterglows of the AXP outbursts, while for SGR outbursts, the X-ray afterglows have less integrated energy than the burst itself. With now a handful of AXP and SGR outbursts and subsequent relaxations observed, we can begin to look for correlations between different outburst and relaxation properties in the hope of constraining magnetar physics. For example, SGR outburst recoveries have been modeled as crustal cooling following impulsive heat injection, and in principle can yield constraints on the nature of the crustal matter \citep{let02}. Alternatively, the AXP events have been interpreted in terms of magnetospheric twisting \citep{tlk02,bt07}, whose recovery depends on electrodynamics in the region of the magnetosphere immediately outside the stellar surface. On the other hand, \citet{gogk07} suggest that AXP recoveries can be modeled with a stationary magnetosphere, with only the surface temperature changing. They argue that their model, which includes the stellar atmosphere, can be used to quantitatively determine the source's magnetic field. In this paper we present a spectral and pulsed flux analysis of 10 \xmm\, observations of AXP 1E~2259+586\ taken between 2002 and 2005, as the source relaxed back toward quiescence following its 2002 outburst. We compare the X-ray flux and spectral evolution of \ttfn\, with those of other magnetars, and interpret these results in terms of the magnetar model. | \label{sec:disc} In this paper, we have presented a comprehensive study of the X-ray recovery of AXP 1E~2259+586 following its 2002 outburst. Here we discuss the properties of this recovery, compare them with those of other magnetar outbursts, and consider how they constrain the magnetar model. \subsection{Return to ``Quiescence''} In our 2004 and 2005 {\it XMM} observations, the source's temperature and unabsorbed fluxes were still higher than the preoutburst value (Fig. \ref{fig:all}). This suggests that the source was not fully back to the preoutburst flux level. Our power law fit to the flux decay shows that the after-outburst quiescent flux level is $(1.75 \pm 0.02)\times 10^{-11}\, \rm ergs,s^{-1} cm^{-2}$, which is significantly higher than the preoutburst value [$(1.59\pm0.01) \times 10^{-11}\,\rm ergs\,s^{-1} cm^{-2} $; Table \ref{tab:fit}]. Either the 2005 flux had still not returned to its quiescent level, or perhaps it had returned to quiescence but the flux just before the event was unusually low. Also possible is that this (and other) AXPs do not have well-defined constant quiescent fluxes, but have long-term flux variations. Indeed, there is evidence for some X-ray flux variability in \ttfn\, over the years since its discovery in 1981 \citep{bs96}. Other AXPs also show variability on a variety of timescales (see \citealp{kas07} for a review). \subsection{Comparison with other Magnetar Recoveries} It is useful to compare the behavior observed from \ttfn\, with that of other magnetars. \ninet's flux was found to follow a power law of index $-0.713\pm0.025 $ after its 1998 August 27 flare \citep{wkg+01}.\footnote{Later the afterglow of the \ninet\, August 27 flare was fitted with a power law plus constant model instead of the single power law model used by \cite{wkg+01}, and a decay index of $\sim0.9$ was obtained \citep{fmw+03}.} This has been interpreted as the cooling of the magnetar outer crust following a sudden release of magnetic energy \citep{let02}. This model predicts a power law decay of index $\sim -2/3$. The flux of SGR 1627$-$41 was found to decay following a power law of index $\sim-0.47$ since its 1998 source activation. Approximately 800 days after the source activation, SGR 1627$-$41's flux suddenly declined by a factor of 10. This behavior is also well fitted by the crust cooling model, although with some fine tuning \citep{kew+03}. We fit the \xmm 2--10 keV unabsorbed fluxes of \ttfn\, with a power law plus constant model, and found the best-fit power law index to be $-0.69 \pm 0.03 $, close to that of \ninet, and that predicted by the model. This suggests that the \ttfn\, outburst afterglow may also be explained by the diffusion of heat in the outer crust. The transient AXP \etten exhibited an outburst in 2003. \cite{ims+04} found that the afterglow of the \etten\, outburst as observed by \xte could be described by a power law decay model ($F\propto t^{-\beta} $) with $\beta=0.45-0.73 $. This is similar to the behavior of \ttfn\, and other SGRs. However, \cite{gh07} found that, with more observations taken by \chandra\, from 2003 to 2006, the afterglow of the \etten\, outburst actually followed an exponential decay of timescale 233.5 days. As we have shown in this paper, the pulsed and unabsorbed X-ray flux decay of \ttfn\, favors the power law decay model over the exponential decay. Perhaps the physical processes involved in the 2003 outburst of \etten were different from those in 2002 outburst of \ttfn. \subsection{Twisted Magnetosphere Model} \cite{tlk02} reported that, if there exists a global twist of the magnetosphere, the decay timescale $\tau$ of this twist would be \begin{equation} \label{eq:twist} \tau = 40\Delta \phi^2 (\frac{L_X}{10^{35}{\rm ergs\,s^{-1}}})^{-1}(\frac{B_{\rm pole}}{10^{14} \rm G})^2(\frac{R_{\rm NS}}{10 \rm km})^3 \rm yr. \end{equation} \cite{wkt+04} argued that, for \ttfn, the twist angle $\Delta \phi$ should be $\sim 10^{-2}$ rad. Thus, the predicted twist relaxation timescale of \ttfn\, is several hours, which is coincidently the timescale of the steeper flux decay observed at the beginning of the afterglow. However, \citet{bt07} have shown more recently that this decay timescale is actually expected to be much larger than equation~(\ref{eq:twist}) suggests. This is because, in their model, the self-induction of the twisted portion of the magnetosphere accelerates particles from the stellar surface and initiates avalanches of pair creation which forms the corona. This corona persists in dynamic equilibrium, maintaining the electric current, as long as dissipation permits. The relevant timescale in this picture for the decay of a sudden twist is given by \begin{equation} \tau \simeq 0.3 (\frac{L_X}{10^{35}{\rm ergs \, s^{-1}}}) (\frac{e\Phi_e}{\rm GeV})^{-2}(\frac{R_{\rm NS}}{10 \rm km}) \,\, {\rm yr}, \end{equation} where $L_X$ is the peak X-ray luminosity and $e\Phi_e$ is the voltage along the twisted magnetic field lines and should nearly universally be $\sim$1~GeV \citep[see][]{bt07}. For 1E~2259+586, we find $\tau \simeq 1.2$~yr, given the peak luminosity $L_X \sim 4 \times 10^{35}(d/3 \; {\rm kpc})$~ergs~s$^{-1}$. Thus, the longer observed decay after the initial steep decline may indeed correspond to the untwisting of a coronal flux tube in the \citet{bt07} picture, although the predicted timescale is somewhat smaller than the observed time to return to quiescence. We note that the \citet{bt07} model predicts a linear flux decline, in contrast to what we have observed for 1E~2259+586 and what has been observed for XTE~J1810$-$197 \citep{gh07}. Moreover, in the $\sim$5 yr of {\it RXTE} monitoring of 1E~2259+586 prior to its 2002 outburst \citep{gk02}, its pulsed X-ray luminosity in the 2--10~keV band was roughly constant at $\sim 2 \times 10^{34}$~ergs~s$^{-1}$. This also is puzzling given the \citet{bt07} prediction that if the time between large-scale events is longer than the decay time from the previous event, the magnetar should enter a quiescent state in which the observed luminosity is dominated by the surface blackbody emission. Why should the ``quiescent'' blackbody emission from 1E~2259+586 be a full order of magnitude larger than that from XTE~J1810$-$197, especially given the latter's much larger inferred magnetic field ($1.7 \times 10^{14}$ versus $6 \times 10^{13}$~G)? This disparity in ``quiescent,'' steady luminosities is even larger when considering AXP 1E~1841$-$045, which has an apparently steady 2--10~keV luminosity of $1.4 \times 10^{35}\rm ~ergs~s^{-1}$, and comparing with probable AXP AX~1845$-$0258, which has quiescent luminosity approximately 2 orders of magnitude smaller \citep{tkgg06}. Distance uncertainties may contribute but not on a scale that can significantly alleviate this problem. This remains an interesting puzzle in magnetar physics. The twisted magnetosphere or flux tube models generically predict that the flux and spectral hardness of magnetars in outburst should be roughly correlated due to increased scattering optical depth when the twist is larger. However, a similar prediction for a flux/hardness correlation was made by \cite{og07} in their thermally emitting magnetar model, using a simple prescription for the magnetosphere and scattering geometry, with the latter stationary, i.e. invoking no variable magnetospheric twists. \citet{gogk07} found that their model could reproduce the existing data for XTE~J1810$-$197. We note that hardness-intensity correlations have now been observed for \seven\, \citep{cri+07}, \tfe\, \citep{tgd+07}, and as we report, in our \ttfn\, \xmm observations. It would be interesting to apply analysis of \cite{og07} to these data, but it is outside the scope of this paper. \subsection{Other Observed Recovery Properties} The fact that the rms and area pulsed fractions remained largely constant while the blackbody radius (in the blackbody plus power law model) changed by a factor of $\sim$2 (Fig. \ref{fig:all}) is worth considering, if the empirical blackbody plus power law spectrum model somehow resembles the real radiation mechanism. % Pulsed fraction should generally decrease when the thermally radiating region on the star grows, provided that this region is not very small compared to the entire surface. Any realistic spectral model which takes radiative transfer in the atmosphere and scattering through the magnetosphere into account should be able to reproduce the observation in this regard as well. A clear anti-correlation between \tfe's pulsed fraction and unabsorbed flux has been observed \citep{tmt+05,gkw06,tgd+07}. However, we found no such correlation in the 2--10 keV band for \ttfn. On the contrary, its 0.1--2 keV area pulsed fractions seem to be correlated with both 0.1--2 and 2--10 keV unabsorbed fluxes (see Fig. \ref{fig:flxpf}). \cite{gh07} found that \etten's pulsed fraction measured between 2003 and 2006 after its outburst decreased with the decay of its flux, i.e. \etten's pulsed fraction is also correlated with flux. Thus, the striking anti-correlation between pulsed fraction and flux observed from \tfe\, is clearly not universal. Finally, we note that the near-infrared flux decay of \ttfn\, was found to follow a power law of index $-0.75^{+0.22}_{-0.33}$ when fitted to a power law plus constant model \citep{tkvd04}. This decay index is close to what we found for the X-ray flux decay, thus confirming the reported correlation between near-IR and X-ray fluxes postoutburst.\footnote{The $-$0.22 X-ray decay index reported by \cite{wkt+04} and the $-$0.21 near-infrared flux decay index reported by \cite{tkvd04} were obtained from a simple power law fitting, i.e. with no quiescent level included in the fit.} \citet{tgd+08} and \citet{wbk+08} showed that the near-IR flux of 1E~1048.1$-$5937 do show correlation with X-rays at times of outbursts. However, \citet{crp+07} show that the near-IR flux variation of XTE~J1810$-$197 is not simply correlated with X-ray flux nor even monotonic postoutburst. Thus, the AXP picture with regard to near-IR variability is not yet fully clear. | 7 | 10 | 0710.1896 |
0710 | 0710.3893_arXiv.txt | {Neutrinos heavier than $M_Z/2\sim 45$ GeV are not excluded by particle physics data. Stable neutrinos heavier than this might contribute to the cosmic gamma ray background through annihilation in distant galaxies as well as to the dark matter content of the universe.} { We calculate the evolution of the heavy neutrino density in the universe as a function of its mass, $M_N$, and then the subsequent gamma ray spectrum from annihilation of distant $N\bar{N}$ (from $0<z<5$). } { The evolution of the heavy neutrino density in the universe is calculated numerically. In order to obtain the enhancement due to structure formation in the universe, we approximate the distribution of $N$ to be proportional to that of dark matter in the GalICS model. The calculated gamma ray spectrum is compared to the measured EGRET data.} { A conservative exclusion region for the heavy neutrino mass is 100 to 200~GeV, both from EGRET data and our re-evalutation of the Kamiokande data. The heavy neutrino contribution to dark matter is found to be at most 15\%. } {} | The motivation for a fourth generation neutrino comes from the standard model of particle physics. In fact, there is nothing in the standard model stating that there should be exactly three generations of leptons (or of quarks for that matter). The present limits on the mass of a fourth generation of neutrinos are only conclusive for $M_N\lesssim M_Z/2 \approx 46$ GeV \citep[p.~35]{2006JPhG...33....1Y}. This limit is obtained from the measurement of the invisible width of the $Z^0$-peak in LEP, which gives the number of light neutrino species, as $N_\nu = 2.9841 \pm 0.0083$ \citep{2001hep.ex...12021T}. In \cite{2000PhLB..476..107M}, a fourth generation of fermions is found to be possible for $M_{N}\sim 50$ GeV, while heavier fermions are shown to be unlikely. However, this constraint is only valid when there is a mixing between the generations \citep{Novikov:2001md} and since this is not necessarily true, we will not take it for certain. In the context of cosmology and astrophysics there are other contraints. Light neutrinos, with $M_N\lesssim 1$ MeV, are relativistic when they decouple, whereas heavier neutrinos are not. The light neutrinos must have $\sum m_\nu\lesssim 46$ eV in order for $\Omega_\nu h^2<1$ to be valid \citep{2006PrPNP..57..309H}. For the dark matter (DM) content calculated by \cite{2003ApJS..148..175S}, the bound is $\sum m_\nu\lesssim 12$~eV. The number of light neutrino species are also constrained to $N_\nu = 4.2^{+1.2}_{-1.7}$ by the cosmic microwave background (CMB), large scale structure (LSS), and type Ia supernova (SNI-a) observations at 95\% confidence \citep{2006JCAP...01..001H}. Neutrinos heavier than about 1 MeV, however, leave thermal equilibirum before decoupling and therefore their number density drops dramatically, see for example \cite{RevModPhys.53.1}. This will be discussed in more detail in Sect.~\ref{se:Evolution}. The most important astrophysical bound on heavy neutrinos comes from Kamiokande \citep{1992PhLB..289..463M} and this will be considered separately in the end. In \cite{PhysRevD.52.1828}, it is found that the mass range $60\lesssim M_N \lesssim 115$ GeV is excluded by heavy neutrino annihilation in the galactic halo. However, according to \citet[p.~57]{2002PhR...370..333D} this constraint is based on an exaggerated value of the density enhancement in our galaxy. Other works constraining the heavy neutrino mass include \cite{1998JETPL..68..685F,1999astro.ph..2327F} and \cite{2004hep.ph...11093B}. There has also been a study of the gamma ray spectrum of dark matter (DM) in general \citep{2007PhRvD..75f3519A}. For an exhaustive review of modern neutrino cosmology, including current constraints on heavy neutrinos, see \citet{2002PhR...370..333D}. It is concluded that there are no convincing limits on neutrinos in the mass range $50\lesssim M_N \lesssim 1000$ GeV. A review of some cosmological implications of neutrino masses and mixing angles can be found in \cite{2003nema.conf...53K}. In this paper we consider a stable fourth generation heavy neatrino with mass $M_N \gtrsim 50$ GeV possessing the standard weak interaction. We assume that other particles of a fourth generation are heavier and thus do not influence the calculations. We assume a $\Lambda$CDM universe with $\Omega_{tot} = \Omega_m + \Omega_\Lambda = 1$, where $\Omega_m = \Omega_b + \Omega_{DM} = 0.135/h^2$, $\Omega_b = 0.0226/h^2$ and $h = 0.71$ \citep{2003ApJS..148..175S}, using WMAP data in combination with other CMB datasets and large-scale structure observations % (2dFGRS + Lyman $\alpha$). % Throughout the article we use natural units, such that the speed of light, Planck's reduced constant and Boltzmann's constant equal unity, $c = \hbar = k_B = 1$. If heavy neutrinos ($M_N\gtrsim 50$ GeV) exist, they were created in the early universe. They were in thermal equilibrium in the early stages of the hot big bang, but froze out relatively early. After freeze-out, the annihilation of $N\bar N$ continued at an ever decreasing rate until today. Since those photons that were produced before the decoupling of photons are lost in the CMB, only the subsequent $N\bar N$ annihilations contribute to the photon background as measured on earth. The intensity of the photons from $N\bar N$-annihilation is affected by the number density of heavy neutrinos, $n_N$, whose mean density decreases as $R^{-3}$, where $R$ is the expansion factor of the universe. However, in structures such as galaxies the mean density will not change dramatically, and since the number of such structures are growing with time, this will compensate for the lower mean density. Note that the photons are also redshifted with a factor $R$ due to their passage through space-time. This also means that the closer annihilations will give photons with higher energy than the farther ones. | \label{sec:Discussion} The numerical calculation of the evolution of the heavy neutrino number density indicates that in the mass region $100\lesssim M_N \lesssim 200$, the cosmological neutrinos would give a cosmic ray signal that exceeds the measurements by the EGRET telescope \citep{1998ApJ...494..523S}. Note that the clumping factor for these limits is rather conservative. In \citet{2002PhRvD..66l3502U}, this factor is much larger, which would also produce a stronger limit on the heavy neutrino mass. We can also compare our neutrino density with the results from the Kamiokande collaboration \citep{1992PhLB..289..463M}. We scale the neutrino signal in their Fig.~2 to $\Omega_N/\Omega_{DM}$, where we use $h_0=0.71$, $\Omega_{m}=0.2678$ and $\Omega_b=0.044$. This is shown in Fig.~\ref{fig:compKamiokande}, where we compare our numerical results for the relic neutrino density to the observed muon flux in the Kamiokande detector. \begin{figure}[here!] \resizebox{\hsize}{!}{\includegraphics{compKamiokande.pdf}} \caption{ Predicted signal from enhanced $N\bar N$ annihilation in the earth and the sun compared to the measured signal in the Kamiokande. On the y-axis: the number of muons (per 100~m$^2$year) produced by muon neutrinos resulting from heavy neutrino collisions in the sun and the earth, as evaluated by \cite{1992PhLB..289..463M}, but scaled to our $\Omega_N(M_N)$. On the x-axis: the heavy neutrino mass in GeV. } \label{fig:compKamiokande} \end{figure} This gives an exclusion region of $80\lesssim M_N \lesssim 400$ GeV. Our analytical results, which are comparable to the traditional relic neutrino densities, is about a factor two lower, giving an exclusion region of $90\lesssim M_N \lesssim 300$~GeV. The model that gives these limits \citep{1987ApJ...321..571G} is rather complicated and not verified experimentally, so these results cannot be taken strictly. Note also that in the three-year WMAP analysis \citep{2007ApJS..170..377S}, the value of $\Omega_{DM}$ depends on which other data the WMAP data are combined with. For WMAP+CFHTLS $\Omega_{DM}$ can be as high as 0.279 and for WMAP+CBI+VSA it can be as low as 0.155. The higher of these possibilities would give an exclusion region of $85\lesssim M_N \lesssim 350$ GeV. The lower boundary value would give an exclusion region of $75\lesssim M_N \lesssim 500$ GeV. A conservative limit based on the Kamiokande data gives the exclusion region $100\lesssim M_N \lesssim 200$ GeV. If a heavy neutrino exists with a mass $M_N\sim100$ GeV or $M_N\sim200$ GeV it would give a small bump in the data at $E_\gamma\sim1$ GeV. Currently the data points are too far apart and the error bars too large to neither exclude nor confirm the eventual existence of such a heavy neutrino. Most of this part of the gamma ray spectrum is usually attributed to blazars, which have the right spectral index, $\sim 2$ \citep{1997ApJ...490..116M}. We note that there could be an enhancement in the signal due to the higher DM densities within galaxies compared to the mean density in the halos. On the other hand, from within galaxies there will also be an attenuation due to neutral hydrogen, thus reducing the enhancement. There will also be a certain degree of extinction of the signal due to neutral hydrogen along the line of sight, but even if we assume complete extinction above $z=4$ the resulting spectrum decreases with only about 20\%. We are also aware of the ongoing debate concerning the antiprotons -- whether or not the DM interpretation of the EGRET gamma excess is compatible with antiproton measurements \citep{2006JCAP...05..006B, 2006astro.ph.12462D}. We note the argument by de Boer that antiprotons are sensitive to electromagnetic fields, and hence their flux need not be directly related to that of the photons, even if they too were produced by $N\bar N$ annihilation. In the advent of the Large Hadron Collider, we also point out that there may be a possibility to detect the existence of a heavy neutrino indirectly through the invisible Higgs boson decay into heavy neutrinos \citep{2003PhRvD..68e4027B}. It will of course be interesting to see the results of the gamma ray large area space telescope (GLAST). It has a field of view about twice as wide (more than 2.5 steradians), and sensitivity about 50 times that of EGRET at 100~MeV and even more at higher energies. Its two-year limit for source detection in an all-sky survey is $1.6 \times 10^{-9}$ photons cm$^{-2}$~s$^{-1}$ (at energies $>$ 100~MeV). It will be able to locate sources to positional accuracies of 30 arc seconds to 5 arc minutes. The precision of this instrument could well be enough to detect a heavy neutrino signal in the form of a small bump at $E\sim 1$ GeV in the gamma spectrum, if a heavy neutrino with mass $\sim$100 or 200~GeV would exist. There are also some other possible consequences of heavy neutrinos that may be worth investigating. The DM simulations could be used to estimate the spatial correlations that the gamma rays would have and to calculate a power spectrum for the heavy neutrinos. This could be interesting at least for masses $M_N\sim 100$~GeV and $M_N\sim 200$~GeV. The annihilation of the heavy neutrinos could also help to explain the reionization of the universe. Another possible interesting application of heavy neutrinos would be the large look-back time they provide \citep{2006PhLB..639...14S}, with a decoupling temperature of $\gtrsim 10^{13}$ K \citep{1989NuPhB.317..647E}. | 7 | 10 | 0710.3893 |
0710 | 0710.4966_arXiv.txt | { The EGRET excess of diffuse Galactic gamma rays shows all the features expected from dark matter annihilation (DMA): a spectral shape given by the fragmentation of mono-energetic quarks, which is the same in all sky directions and an intensity distribution of the excess expected from a standard dark matter halo, predicted by the rotation curve. From the EGRET excess one can predict the flux of antiprotons from DMA. However, how many antiprotons arrive at the detector strongly depends on the pro\-pagation model. The conventional isotropic propagation models trap the antiprotons in the Galaxy leading to a local antiproton flux far above the observed flux. According to Bergstr\"{o}m et. al. this excludes the DMA interpretation of the EGRET excess. Here it is shown that more realistic anisotropic propagation models, in which most antiprotons escape by fast transport in the z-direction, are consistent with the B/C ratio, the antiproton flux and the EGRET excess from DMA. } % | \label{intro} The interpretation of the observed EGRET excess of diffuse Galactic gamma rays as Dark Matter annihilation (DMA) (see \cite{us} or contributions by W. de Boer, C. Sander and M. Weber, this volume) could be a first hint at the nature of dark matter. The excess was observed in all sky directions. From the spectral shape of the excess the WIMP mass was constrained to be between 50 and 100 GeV and from the distribution of the excess in the sky the Dark Matter (DM) halo profile was obtained. One of the most important criticisms of this analysis was a paper by Bergstr\"{o}m et. al. \cite{bergstrom1} claiming that the antiproton flux from DMA would be an order of magnitude higher than the observed antiproton flux. They used a conventional propagation model assuming the propagation of charged particles to be the same in the halo and the disk. However, the propagation in the halo (perpendicular to the disk) can be much faster than the propagation in the disk \cite{breit}. In this paper we show that the local antiproton flux from DMA can be strongly reduced in an anisotropic propagation model and that the DMA interpretation of the EGRET excess can by no means be excluded by Galactic antiprotons. In section \ref{CM} we discuss the pro\-blems of the isotropic model for cosmic ray transport leading to the fact that our galaxy can work as a large storage box for antiprotons. An anisotropic pro\-pagation model, which simultaneously describes the \\ EGRET excess and and the observed local fluxes of charged cosmic rays is introduced in section \ref{CP}. Section \ref{conclusion} summarizes the results. | Tracing of charged particles in realistic models of the regular Galactic magnetic fields with a turbulent (small-scale) component has shown that CRs remember the regular field lines, even if the irregular component is of the same order of magnitude as the regular, thus leading to enhanced diffusion in $\phi$ and $z$ (see Fig. A1 in \cite{blasi}). With such an anisotropic propagation model the amount of antiprotons expected from DMA can be reduced by one to two orders of magnitude. Therefore the claim by \cite{bergstrom1} that the DMA interpretation of the EGRET excess of diffuse Galactic gamma rays is excluded is strongly pro\-pagation model dependent. It only applies to a pro\-pagation model with isotropic diffusion. An anisotropic pro\-pagation model with different pro\-pagation in the halo and the disk can reconcile the EGRET excess with the antiproton flux and the ratios of secondary/primary and unstable/stable nuclei. Clearly the DMA search for light DM particles is propagation model dependend. Taking these uncertainties into account shows that DMA is a viable explanation of the EGRET excess of diffuse Galactic gamma rays, as shown in \cite{us} and and can by no means be excluded by the antiproton flux predicted by a specific model. \vspace*{5mm} | 7 | 10 | 0710.4966 |
0710 | 0710.0774_arXiv.txt | In the third part of our photometric study of the star-forming region NGC~346/N~66 and its surrounding field in the Small Magellanic Cloud (SMC), we focus on the large number of low-mass pre-main sequence (PMS) stars revealed by the Hubble Space Telescope Observations with the Advanced Camera for Surveys. We investigate the origin of the observed broadening of the pre-main sequence population in the $V-I$, $V$ CMD. The most likely explanations are either the presence of differential reddening or an age spread among the young stars. Assuming the latter, simulations indicate that we cannot exclude the possibility that stars in NGC 346 might have formed in two distinct events occurring about 10 and 5 Myr ago, respectively. We find that the PMS stars are not homogeneously distributed across NGC 346, but instead are grouped in at least five different clusters. On spatial scales from 0.8$''$ to 8$''$ (0.24 to 2.4\,pc at the distance of the SMC) the clustering of the PMS stars as computed by a two-point angular correlation function is self-similar with a power law slope $\gamma \approx -0.3$. The clustering properties are quite similar to Milky Way star forming regions like Orion OB or $\rho$\,Oph. Thus molecular cloud fragmentation in the SMC seems to proceed on the same spatial scales as in the Milky Way. This is remarkable given the differences in metallicity and hence dust content between SMC and Milky Way star forming regions. | It is well known that Galactic OB associations also host large numbers of fainter, low-mass pre-main sequence (PMS) stars (e.g.\ Preibisch \& Zinnecker 1999; Sherry et al.\ 2004; Brice\~{n}o et al.\ 2005). Low-mass PMS stars in stellar associations provide a longer-lasting record of the most recent star formation events than the short-lived high-mass stars. Large-scale surveys have identified hundreds of PMS stars in nearby OB associations (Brice\~{n}o et al. 2007). In order to understand the star formation triggering and history, or the Initial Mass Function (IMF), one has to study both high- and low-mass stars in star forming regions. In many cases the low-mass populations of galactic OB associations cannot be easily distinguished from foreground main-sequence stars (e.g. Brice\~{n}o et al. 2001). Recently PMS stars have been discovered for the first time in an extragalactic stellar association, which suffers much less from contamination by the galactic disk. {\em Hubble Space Telescope} (HST) observations of the association LH\,52 in the Large Magellanic Cloud (LMC) revealed $\approx$500 PMS stars with masses down to 0.3\,M$_\odot$ (Gouliermis et al. 2006a). The investigation of individual extragalactic pre-main sequence stars opens a new field of study. As both high angular resolution and wide field-of-view are required, HST is the ideal observatory to carry out these studies. While HST observations of extragalactic associations cannot reach the detection limits achieved for local associations, they provide a unique opportunity for studying low-mass star formation in other galaxies. The OB association NGC~346 in the Small Magellanic Cloud (SMC) is located in the central part of the brightest {\sc H ii} region in the SMC, named LHA~115-N~66 or in short N~66 (Henize 1956). With 33 spectroscopically confirmed O and B stars, NGC~346 hosts the largest sample of young, massive stars in the SMC (Walborn 1978; Walborn \& Blades 1986; Niemela et al. 1986; Massey et al. 1989; Walborn et al. 2000; Evans et al. 2006). Photoionization models by Rela\~{n}o et al. (2002) imply that these stars are a major source of ionizing flux for the surrounding diffuse interstellar medium (ISM). Recent imaging from the Wide-Field Channel (WFC) of the {\em Advanced Camera for Surveys} (ACS) on-board HST (GO Program 10248; PI: A. Nota) revealed the PMS stellar content of the general region of NGC~346/N~66 down to the sub-solar mass regime (Nota et al.\ 2006; Gouliermis et al.\ 2006b -- hereafter Paper I -- ; Sabbi et al.\ 2007). Nota et al.\ (2006) suggest that all PMS stars in the association are the product of a single star formation event, taking place 3 to 5 Myr ago. The PMS distribution in the $V-I$, $V$ color-magnitude diagram (CMD), however, shows a prominent widening, which may be explained by an age spread of $\approx$10 Myr. This raises the question {\em if the PMS stars in NGC\,346 are indeed the result of a single star formation event.} Sabbi et al.\ suggest that the PMS population is mainly concentrated in a number of subclusters (three of them at the central part, where the association is located), which formed at the same time from the turbulence-driven density variations, and not following a sequential process. Star counts, however, reveal only a few compact PMS clusters with a significant number of stars in the area around the association. The main body of the association cannot be divided into separate subclusters (Paper I). {\em Could the remote clusters be the product of a star formation event occurring before or after the event which triggered the formation of the association?} In the first part of our study of this extraordinary star-forming region (Paper I), we compiled our ACS photometry of almost 100,000 stars, and presented preliminary results on stellar types and their distinct spatial distributions. We confirmed the co-existence of massive OB stars and low-mass PMS stars in NGC 346. In the present paper, we explore in greater detail the properties of the low-mass PMS stars with a focus on clustered star formation in NGC~346/N~66. The paper is organized as follows: In \S~2 we describe the datasets from HST/ACS, present the color-magnitude diagram, carry out a comparison of our photometry of the brightest stars with previous photometric studies, and discuss the reddening in the region. The broadening of the PMS distribution in the CMD, a thorough discussion of possible causes and explanations of this phenomenon as well as the spatial variations in the CMD are presented in \S~3. In \S~4 we present the clustering properties of the PMS stars and discuss them in terms of hierarchical star formation. Our analysis of H\alp\ observations with HST/ACS as well as previous {\em Spitzer} Space Telescope results on Young Stellar Objects (YSOs) are presented in \S~5 and \S~6, respectively. In \S~7 we summarize the results. | In this paper we present the results from our photometric study on the recent star formation history of N~66 the brightest {\sc H~ii} region in the SMC, related to the OB association NGC~346, as it is recorded in the observed PMS population of the region. Our deep photometry revealed an extremely large number of low-mass PMS stars in the association and the surrounding region, easily distinguishable in the $V-I$, $V$ CMD. We show that factors such as reddening, binarity and variability can cause a broadening of the positions of these stars in the CMD of the association, wide enough so that they can be misinterpreted as the result of multi-epoch star formation, and not as the product of a single star formation event. We found that a modest reddening like the one we found for the general area of NGC~346 ($E(B-V)\simeq 0.08$ mag) can make the PMS stars appear younger than what they actually are, and therefore an age of 10 Myr better fits the observed sequence of PMS stars. However, our results do not exclude the possibility of multi-epoch star formation in the area of the association if the reddening is even lower, as suggested from the OB stars of this region ($E(B-V)\simeq 0.04$ mag). In this case, two star formation events 10 and 5 Myr ago can explain the observed broadening of the PMS stars due to age spread and factors such as reddening and binarity. No specific dependence of the estimated ages of the PMS stars to their loci within the association, as a signature of sequential star formation, was found. It has been previously suggested that three different generations of stars occurred through sequential star formation in the region of NGC~346/N~66 (Rubio et al. 2000) within the last 3 Myr (based on the age of the youngest OB stars) and that the PMS stars of NGC~346 represent a star formation event that took place 3 - 5 Myr ago (Nota et al. 2006). However, the study by Massey et al. (1989) based on the presence of evolved 15~M{\solar} stars suggests that there might have been earlier star formation events in the region making NGC~346 as old as $\sim$~12 Myr. In addition, there are recent indications that NGC~346 might host classical Be-type stars (Evans et al. 2006), and if so, the age of the association should be at least 10~Myr, a threshold given by Fabregat \& Torrej{\'o}n (2000) for classical Be-type stars to form. These results fit very well to the hypothesis that star formation in NGC~346 did already occur about 10~Myr ago, as our observations and simulations of the PMS stars suggest. From star counts based on our ACS photometry we identify at least five PMS clusters across the region, covering a range of ages. On spatial scales from 0.8$''$ to 8$''$ (0.24 to 2.4\,pc at the distance of the SMC) the clustering of the PMS stars as computed by a two-point angular correlation function is self-similar with a power law slope $\gamma \approx -0.3$. The clustering properties are quite similar to Milky Way star-forming regions like the Orion OB association or $\rho$\,Oph. Thus molecular cloud fragmentation in the SMC seems to proceed on the same spatial scales as in the Milky Way. This is remarkable given the differences in metallicity and hence dust content between SMC and Milky Way star-forming regions. The youngest PMS stars are located mostly to the north of the bar of N~66, where three PMS clusters are identified. This area is also characterized by a high concentration of candidate YSOs (Simon et al. 2007), H\alp-excess stars (found with our photometry), and IR-emission peaks (Rubio et al. 2000). This indicates that star formation probably still takes place in an arc-like feature, as it is outlined by the spatial distribution of these sources. In an accompanying letter, we combine these results with previous multi-wavelength studies of the region to investigate the star formation history, which helped to shape NGC~346/N~66 (Gouliermis et al. 2007b). | 7 | 10 | 0710.0774 |
0710 | 0710.0890_arXiv.txt | A next-generation lunar laser ranging apparatus using the 3.5~m telescope at the Apache Point Observatory in southern New Mexico has begun science operation. APOLLO (the Apache Point Observatory Lunar Laser-ranging Operation) has achieved \emph{one-millimeter} range precision to the moon which should lead to approximately one-order-of-magnitude improvements in several tests of fundamental properties of gravity. We briefly motivate the scientific goals, and then give a detailed discussion of the APOLLO instrumentation. | \subsection{Scientific Motivation\label{sub:Scientific-Motivation}} A variety of observations and theoretical explorations---including the apparent acceleration of the expansion of the universe \citep{sn1,sn2}, the possible existence of extra dimensions \citep{extra_dim}, and attempts to reconcile quantum mechanics and gravity---provide motivation for improved tests of the fundamental aspects of gravity. Lunar Laser Ranging (LLR) currently provides the best tests of a number of gravitational phenomena \citep{jgw-96,jgw-latest} such as: \begin{itemize} \item the strong equivalence principle (SEP): $\eta\approx5\times10^{-4}$ sensitivity \item time-rate-of-change of the gravitational constant: $\dot{G}/G<10^{-12}$ yr$^{-1}$ \item geodetic precession: 0.6\% precision confirmation \item deviations from the $1/r^{2}$ force law: $\sim10^{-10}$ times the strength of gravity at $10^{8}$ meter scales \end{itemize} LLR also tests other gravitational and mechanical phenomena, including for example gravitomagnetism \citep{gravmag}, preferred frame effects \citep{alpha-1,alpha-2}, and Newton's third law \citep{newtonsthird}. LLR may also provide a window into the possible existence of extra-dimensions via cosmological dilution of gravity \citep{lue,dvalimoon}. Besides the SEP, LLR tests the weak equivalence principle (WEP) at the level of $\Delta a/a<1.3\times10^{-13}$, but the LLR constraint is not competitive with laboratory tests. In addition, LLR is used to define coordinate systems, probe the lunar interior, and study geodynamics \citep{dickey}. These constraints on gravity are based on about 35 years of LLR data, although the precision is dominated by the last $\sim$15 years of data at 1--3~cm precision. APOLLO aims to improve tests of fundamental gravity by approximately an order-of-magnitude by producing range points accurate at the one-millimeter level. \subsection{A Brief History of LLR\label{sub:A-Brief-History}} The first accurate laser ranges to the moon followed the landing of the first retroreflector array on the Apollo 11 mission by less than two weeks (August 1, 1969). These were performed on the 3.0~meter telescope at the Lick Observatory. One month later, a second station using the 2.7~meter telescope at the McDonald Observatory began ranging to the moon \citep{bender}. The operation at the Lick Observatory was designed for demonstration of initial acquisition, so that the scientifically relevant observations over the next decade came from the McDonald station, which used a ruby laser with 4~ns pulse width, firing at a repetition rate of about 0.3 Hz and $\sim3$ J/pulse. This station routinely achieved 20~cm range precision, with a photon return rate as high as 0.2 photons per pulse, or 0.06 photons per second. A typical ``normal point''---a representative measurement for a run typically lasting tens of minutes---was constructed from approximately 20 photon returns. In the mid 1980's, the McDonald operation was transferred to a dedicated 0.76~m telescope (also used for satellite laser ranging) with a 200~ps Nd:YAG laser operating at 10~Hz and 150 mJ/pulse. This station is referred to as the McDonald Laser Ranging System: MLRS \citep{mlrs}. At about the same time, a new station began operating in France at the Observatoire de la C\^ote d'Azur (OCA) \citep{oca}. Using a 1.5~meter telescope, a 70~ps Nd:YAG laser firing at 10~Hz and 75 mJ/pulse, this became the premier lunar ranging station in the world. In recent years, the MLRS and OCA stations have been the only contributors to lunar range data with typical return rates of 0.002 and 0.01 photons per pulse, respectively. Typical normal points from the two stations consist of 15 and 40 photons, respectively. Other efforts in LLR are described in \citet{ep-llr}, and more detailed histories may be found in the preceding reference as well as in \citet{bender,dickey}. \subsection{Millimeter Requirements\label{sub:Millimeter-Requirements}} The dominant source of random uncertainty in modern laser ranging systems has little to do with the system components, but rather comes from the varying orientation of the lunar retroreflector arrays. Although the arrays are nominally pointed within a degree of the mean earth position, variations in the lunar orientation---called libration---produce misalignments as large as 10 degrees, and typically around 7 degrees. This means the ranges between the earth and the individual array elements typically have a root-mean-square (RMS) spread of 15--36~mm, corresponding to about 100--240~ps of round-trip travel time. This dominates over uncertainties associated with the laser pulse width, and with jitter in the detector and timing electronics. A typical normal point containing 16 photons will therefore be limited to 4--9~mm range precision by the array orientation alone, though range residuals reported by analysis at the Jet Propulsion Laboratory tend to be larger than this. Reaching the one-millimeter precision goal demands at a minimum the collection of enough photons to achieve the appropriate statistical reduction. Assuming an ability to identify the centroid of $N$ measurements---each with uncertainty $\sigma$---to a level of $\sigma_{\mathrm{net}}=\sigma/\sqrt{N}$, the uncertainty stemming from the retroreflector array orientation typically demands 225--1300 photons in the normal point to reach the one millimeter mark. Worst-case orientations push the individual photon uncertainty to 50~mm, demanding 2500 photons. This is far outside of the capabilities of the aforementioned LLR stations. We point out that any constant range bias is accommodated in the analysis, so that only \emph{variations} in the range are important to the experiment. While adequate photon number is sufficient to reduce statistical uncertainty to the one-millimeter level, other sources of error could potentially limit the ultimate scientific capacity of LLR. Most importantly, the gravitational physics is sensitive to the center-of-mass separations of Earth and Moon, while one measures the distance between a telescope and reflectors that are confined to the body surfaces. The earth's surface in particular has a rich dynamic---experiencing diurnal solid-earth tides of 350~mm peak-to-peak amplitude, plus crustal loading from oceans, atmosphere, and ground water that can be several millimeters in amplitude. Moreover, the earth atmosphere imposes a propagation delay on the laser pulse, amounting to $\sim$1.5~m of zenith delay at high-altitude sites. Satellite laser ranging, very long baseline interferometry, and other geodetic efforts must collectively contend with these same issues, for which accurate models have been produced. A good summary of these models is published by the International Earth Rotation and Reference Systems Service \citep[IERS:][]{iers}. As an example of the state of these models, the long-standing atmospheric model by \citet{marini-murray} has recently been replaced by a more accurate model \citep{atmo1,atmo2}. The model differences for a high-altitude site are no more than 2~mm for sky elevation angles greater than 40 degrees---providing an indicative scale for the model accuracy. The primary input for this model is the atmospheric pressure at the site, as this represents a vertical integration of atmospheric density, which in turn is proportional to the deviation of the refractive index, $n$, from unity. Thus the zenith path delay, being an integration of $n-1$ along the path, is proportional to surface pressure under conditions of hydrostatic equilibrium. A mapping function translates zenith delay to delays for other sky angles. Measuring pressure to a part in 2000 (0.5~mbar) should therefore be sufficient to characterize the 1.5~m zenith delay at the one-millimeter level. Our experiment records atmospheric pressure to an accuracy of 0.1~mbar. The principal science signals from LLR appear at well-defined frequencies. For example, the equivalence principle signal is at the synodic period of 29.53 days, and even secular effects ($\dot{G}$, precession) are seen via the comparative phases between periodic (monthly) components in the lunar orbit. Because many of the effects discussed in the preceding paragraphs are aperiodic, they will not mimic new physics. To the extent that these effects are not adequately modeled, they contribute either broadband noise or discrete ``signals" at separable frequencies. The science output from APOLLO may be initially limited by model deficiencies. But APOLLO's substantial improvement in LLR precision, together with a high data rate that facilitates deliberate tests of the models, is likely to expose the nature of these deficiencies and therefore propel model development---as has been historically true for the LLR enterprise. Ultimately, we plan to supplement our LLR measurement with site displacement measurements from a superconducting gravimeter (not yet installed), in conjunction with a precision global positioning system installation as part of the EarthScope Plate Boundary Observatory (installed February 2007 as station P027). \subsection{The APOLLO Contribution} APOLLO---operating at the Apache Point Observatory (APO)---provides a major improvement in lunar ranging capability. The combination of a 3.5~meter aperture and 1.1~arcsecond median image quality near zenith translates to a high photon return rate. Using a 90~ps FWHM (full-width at half-maximum) Nd:YAG laser operating at 20~Hz and 115~mJ/pulse, APOLLO obtains photon return rates approaching one photon per pulse, so that the requisite number of photons for one-millimeter normal points may be collected on few-minute timescales. To date, the best performance has been approximately 2500 return photons from the Apollo 15 array in a period of 8 minutes. The average photon return rate for this period is about 0.25 photons per shot, with peak rates of 0.6 photons per pulse. Approximately half of these photons arrived in multi-photon bundles, the largest containing eight photons. APOLLO brings LLR solidly into the multi-photon regime for the first time. This paper describes the physical implementation of the APOLLO apparatus, including descriptions of the optical and mechanical design, the electronics implementation, and system-level design. For early reports on APOLLO, see \citet{iwlr12,spacepart,iwlr13,iwlr14}. For an analysis of our expected photon return rate, see \citet{weak_LLR}. A list of acronyms commonly-used in this paper appear in Appendix~\ref{app:acronyms}. | 7 | 10 | 0710.0890 |
|
0710 | 0710.0632_arXiv.txt | We present environmental dependence of the build-up of the colour-magnitude relation (CMR) at $z \sim 0.8$. It is well established that massive early-type galaxies exhibit a tight CMR in clusters up to at least $z \sim 1$. The faint end of the relation, however, has been much less explored especially at high redshifts primarily due to limited depths of the data. Some recent papers have reported a deficit of the faint red galaxies on the CMR at $0.8 \lsim z \lsim 1$, but this has not been well confirmed yet and is still controversial. Using a deep, multi-colour, panoramic imaging data set of the distant cluster RXJ1716.4+6708 at $z=0.81$, newly taken with the Prime Focus Camera (Suprime-Cam) on the Subaru Telescope, we carry out an analysis of faint red galaxies with a care for incompleteness. We find that there is a sharp decline in the number of red galaxies toward the faint end of the CMR below $M^*+2$. We compare our result with those for other clusters at $z \sim 0.8$ taken from the literature, which show or do not show the deficit. We suggest that the "deficit" of faint red galaxies is dependent on the richness or mass of the clusters, in the sense that poorer systems show stronger deficits. This indicates that the evolutionary stage of less massive galaxies depends critically on environment. | \label{sec:intro} It is well known that red early-type galaxies exhibit a tight sequence on colour-magnitude diagrams, which is called the colour-magnitude relation (CMR) (e.g., Visvanathan \& Sandage 1977; \citealp{bow92}). In nearby clusters, the CMRs extend down to at least $5-6$ magnitude fainter than the brightest cluster galaxies (e.g., Terlevich, Caldwell \& Bower 2001). The small colour scatter around the CMR is indicative of the homogeneity of early-type galaxies in clusters (e.g., Bower et al.\ 1992, 1998). At high redshifts, the CMR has already been well established in clusters at least out to $z\sim1$ as far as the bright end is concerned (e.g., \citealt{ell97}; \citealt{kod98}; \citealt{sta98}; \citealp{van98}; 2001; Blakeslee et al.\ 2003; Stanford et al.\ 2006; Mei et al.\ 2006a,b). The faint end of the CMR, however, has been much less explored and still highly uncertain. Some recent deep studies of distant galaxy clusters have shown a relatively small number of galaxies at the faint end of the CMR compared to local clusters. De Lucia et al.\ (2004, 2007) showed such a deficit of faint red galaxies in $z=0.6-0.8$ clusters observed by the ESO Distant Cluster Survey (EDisCS; \citealt{whi05}). A similar result was shown in Stott et al.\ (2007). They compared the faint end of the luminosity functions of red galaxies in $z\sim 0.5$ clusters from Massive Cluster Survey (MACS; Ebeling et al. 2001) with those of $z \sim 0.1$ clusters from Las Campanas/AAT Rich Cluster Survey (LARCS; Pimbblet et al. 2006). \cite{tan05} analysed the RXJ0152.7--1357 cluster (hereafter RXJ0152) at $z=0.83$ based on wide field data taken with the Subaru Prime Focus Camera on the Subaru Telescope (Suprime-Cam; Miyazaki et al.\ 2002), and they also showed a deficit of faint red galaxies on the CMR. Based on these results, De Lucia et al. (2004, 2007) and \cite{tan05} discussed that the faint end of the CMR well visible in the present-day universe was established at relatively later epochs as faint blue galaxies stopped their star formation after $z\sim0.8$ in contrast to much earlier ($z\gg1$) termination of star formation in massive galaxies. \cite{tan05} classified galaxy environment into ``cluster'', ``group'' and ``field'', and examined the environmental dependence of the faint end of the CMR as well. They suggest that the build-up of the CMR depends also on environment in the sense that it is more delayed in lower-density environment. The deficit of the faint end of the CMR is often discussed in the line of a currently favoured observational phenomenon called ``down-sizing''. This trend was first noted for field galaxies by Cowie et al.\ (1996). They showed in their Hawaii Deep Field that most massive galaxies tend to show low star formation rates while less massive galaxies still show on-going star formation activity at $z\lsim1$. Such a trend has been extended in both redshift space and magnitude range. Kauffmann et al.\ (2003) showed in the local SDSS data that massive galaxies are dominated by red old galaxies. By contrast, less massive galaxies show bluer colours due to some on-going star formation, and galaxies below a few times $10^{10}$ \msun {} in stellar mass are predominantly blue. A very similar trend was reported at $z\sim1$ by Kodama et al.\ (2004). They looked in the Subaru/XMM Deep Field and showed the distribution of galaxies at $z\sim1$ on the colour-magnitude diagram. A clear bi-modality on the colour-magnitude diagram was observed again. Since then, a large number of papers have discussed this interesting issue. One of the most convincing cases is based on $\sim$8,000 galaxies with spectroscopic redshifts within $0.7<z<1.4$ in the DEEP2 survey (Bundy et al.\ 2006). They derived stellar mass functions of red and blue galaxies and showed that the mass where the dominant contribution is switched from red to blue galaxies shifts to smaller masses as time goes on. This down-sizing trend is found also in clusters as already mentioned above. Recently, however, Andreon (2006) claimed that the faint end of the CMR is fully in place in the rich cluster MS1054--0321 (hereafter MS1054) at $z=0.83$ and questioned the universality of the deficiency of faint red galaxies at $z \sim 0.8$. A critical problem is that the number of galaxy clusters having deep enough imaging data so that we can discuss the faint end of the CMR is still very limited at high redshifts. In fact, so far only a few clusters at $z \sim 0.8$ (of which some are optically-selected clusters) have been studied in this respect (i.e., MS1054 by \citealt{and06}, RXJ0152 by \citealt{tan05}, and some optically-selected clusters by EDisCS in \citealt{del04}, 2007). Therefore, it is crucial to increase the number of clusters and see if the deficit of faint red galaxies is universally observed or not and see what determines the degree of the deficit. In this paper, we examine another cluster at $z=0.81$, RXJ1716.4+6708 (hereafter RXJ1716), in order to obtain a more general picture of $z\sim0.8$ clusters. We will also discuss a possible origin of the cluster-to-cluster variations. The structure of this paper is the following. In Section 2, we introduce our PISCES project, and also summarize the properties of the RXJ1716 cluster shown by some previous works. We present a deficit of faint red galaxies in Section 3, and we discuss the environmental dependence of the nature of faint galaxies in Section 4. Finally, we summarize our results in Section 5. Throughout this paper we use $\Omega_M =0.3$, $\Omega_{\Lambda} =0.7$, and $H_0 =70$ km s$^{-1}$Mpc$^{-1}$. Magnitudes are all given in the AB system, unless otherwise stated. | \label{sec:summary} Using a deep, multi-colour, panoramic imaging data set of the distant cluster RXJ1716.4+6708 at $z=0.81$, newly taken with the Prime Focus Camera (Suprime-Cam) on the Subaru Telescope, we have carried out an analysis of red-sequence galaxies with a care for incompleteness. We have found that there is a sharp decline in the number of the red galaxies toward the faint end of the CMR below $M^*+2$. We compared our results with those for other clusters at $z \sim 0.8$ taken from the literature, by calculating the luminous-to-faint ratio to quantify the degree of the ``deficit'' and by combining the information on richness of the individual clusters from X-ray properties. We suggest that the deficit of faint red galaxies is dependent on the richness or mass of the clusters in the sense that poorer systems show stronger deficits. This indicates that the evolutionary stage of less massive galaxies depends critically on environment. In order to confirm this interesting trend, we need a much larger sample of galaxy clusters over a wide range in richness, and not only at similar redshifts but also at other redshifts. | 7 | 10 | 0710.0632 |
0710 | 0710.4407_arXiv.txt | Numerical simulations of planetesimal accretion in circumprimary and circumbinary orbits are described. The secular perturbations by the companion star and gas drag are included in our models. We derive limits on the parameters of the binary system for which accretion and then planetary formation are possible. In the circumbinary case we also outline the radial distance from the baricenter of the stars beyond which accumulation always occurs. Hydrodynamical simulations are also presented to validate our N--body approach based on the axisymmetric approximation for the gas of the disk. | The formation of terrestrial planets and cores of giant planets within circumstellar disks involves the accumulation of a large number of planetesimals, solid bodies with initial sizes of roughly several kilometers (Lissauer 1993; Wetherill \& Stewart 1993). The initial growth of the planetesimals can follow different paths depending on the their mutual velocities. If runaway growth occurs, a limited number of large planetary embryos form on a short timescale (about $10^{4} - 10^5$ years) followed by a period of violent mutual collisions until the planets reach their final mass. If the encounter velocities exceeds the planetesimal's escape velocities, the size distribution of the entire population exhibits an orderly growth until larger bodies are formed on a much longer timescale. Most observed extrasolar planets are believed to have formed from planetesimals. The core--accretion model (Pollack et al., 1996) seems to explain a large fraction of the observed physical and dynamical properties of extrasolar gaseous giants in particular after the inclusion of migration by interaction with an evolving disk and gap formation (Alibert et al. (2005)). Neptune--size extrasolar planets possibly formed directly by planetesimal accumulation without reaching the critical mass to accrete a massive gaseous envelope. Around single stars the efficiency of planetesimal accumulation is very high, leading easily to planet formation. The influence of collective perturbations like stirring by mutual gravitational perturbations and damping by collisions and gas drag effects has been studied in detail in order to understand the conditions favoring runaway growth. Radial velocity surveys have shown that exoplanets are found also in binary or higher multiplicity stellar systems (Raghavan et al. 2006, Desidera \& Barbieri 2007). Planetesimal accumulation and then planet formation in binary (or multiple) stellar systems appears to be a more complex process than around single stars. The gravitational secular perturbations by the companion star may overcome the mutual planetesimal interactions and significantly affect the initial stage of accretion by exciting high eccentricities and then affecting significantly the relative velocity distribution. In this paper we explore the velocity evolution of planetesimals in S or C--type orbits under both, the perturbing effects of the companion star and gas drag. We recall that planetesimals revolving just about one star in a binary pair are on so-called "S-type" pr "circumprimary" orbits, whereas those that revolve about both stars have "P-type" or "circumbinary" orbits. | Numerical simulations of planetesimal evolution support the scenario in which planet formation may undergo even in binary star systems. Planetesimals in both S--type and P--type orbits keep their relative velocities low enough to allow accumulation rather than fragmentation for a wide range of binary orbital and physical parameters even if orderly growth or the so--called Type II runaway growth \citep{kort01} are possibly more common than the conventional fast runaway growth presumed to occur around single stars. Both terrestrial planets and giant planets are supposed to form in binary systems unless extreme orbital conditions for the two stars are met like large binary eccentricity or very short separation (in the case of circumprimary disks). The potential lower rate of planet discovery around double stars may be ascribed to these cases rather than to a general effect related to the presence of a companion star. | 7 | 10 | 0710.4407 |
0710 | 0710.4893_arXiv.txt | There is a large body of work that has used the excellent Chandra observations of nearby galaxies with neglible low mass X-ray binary (LMXB) populations. This has culminated in a ``Universal'' X-ray luminosity function (XLF) for high mass X-ray binaries (HMXBs). However, a number of methods have been used to convert from source intensities to luminosities when creating these XLFs. We have taken advantage of the XMM-Newton observations of the nearby starbursting spiral galaxy NGC 253 to test some of these methods. We find the luminosities derived from these various methods to vary by a factor of $\sim$3. We also find the most influential factor in the conversion from intensity to luminosity to be the absorption. We therefore conclude that a more consistent approach is required for determining the true Universal XLF for HMXBs. Ideally, this would involve individual spectral fitting of each X-ray source. Certainly, the line-of-sight absorption should be determined from the observations rather than assuming Galactic absorption. We find the best approach for obtaining an XLF from low-count data to be the splitting of the X-ray sources into two or more intensity intervals, and obtaining a conversion from intensity to flux for each group from spectral modelling of the summed spectrum of that group. | The X-ray populations of external galaxies have been well studied for the last $\sim$20 years. Historically, studies of the individual sources have been severely limited by low count rates and signal to noise, and several methods have been used to derive the X-ray luminosity of a source from its intensity. Grimm et al. (2003, hereafter known as G03) used Chandra and ASCA surveys of nearby starburst galaxies, along with ASCA, MIR-KVANT/TTM and RXTE/ASM observations of HMXBs in our Galaxy and the Magellanic Clouds to obtain a correlation between the X-ray properties of HMXBs and the star formation rate (SFR) of their host galaxies. They chose their sample of galaxies to have sufficiently high SFR to total mass ratios so that their X-ray populations would be dominated by HMXBs, with negligible LMXB contributions. G03 used published Chandra X-ray luminosity functions (XLFs), scaled assuming the Hubble constant to be 70 km s$^{-1}$ Mpc$^{-1}$. They found the XLFs of these galaxies to be strikingly similar, when normalised by the SFR of the galaxy, and proposed a universal HMXB XLF. Additionally, they found that the number of sources with 2--10 keV luminosities $>$2$\times$10$^{38}$ erg s$^{-1}$ to be proportional to SFR$^{1.06\pm0.07}$. Furthermore they discovered a linear relation between the total HMXB X-ray flux of a galaxy and its SFR, for SFRs $\ga$4 M$_{\odot}$ yr$^{-1}$. Several different methods were used to convert from X-ray intensity to luminosity when creating the XLFs for galaxies in the G03 sample. Some XLFs were derived assuming a standard X-ray binary emission model (a power law with spectral index, $\Gamma$, $\sim$1.7 or a 5 keV bremsstrahlung) with Galactic line-of sight absorption (e.g. Zezas et al., 2002; Soria \& Kong, 2002). Others used the X-ray colours to estimate the emission spectrum, using Galactic absorption (e.g. Eracleous et al., 2002), or deriving the absorption from the colours also (e.g. Lira et al., 2002). Also, some XLFs were derived using best fit spectra to individual bright sources (Smith \& Wilson, 2001), or to the stacked X-ray population (Roberts et al., 2002). XMM-Newton (XMM) is the most sensitive X-ray imaging telescope in the 0.3--10 keV band. We can therefore use deep XMM observations of nearby galaxies to glean the accuracy of these methods by comparing their resulting XLFs with the XLF derived from freely fitting each X-ray source. To do this, we chose XMM-Newton observations of NGC 253. NGC 253 is a star-bursting spiral galaxy in the Sculptor Group at a distance of $\sim$4 Mpc; it is almost edge on, with a D$_{25}$ region of $\sim$25$'$$\times$7$'$. We fully describe our analysis of the observations in Barnard et al. (2007, MNRAS submitted, hereafter known as B07), and concentrate here on the 2003, 110 ks XMM observation of NGC253. | Grimm et al. (2003) report a Universal HMXB XLF derived from published XLFs of several nearby galaxies. They also derive relations between the star formation rate and (i) the total luminosity of the point X-ray sources in the galaxies and (ii) the number of X-ray sources in a galaxy with 2--10 keV luminosity $>$2$\times$10$^{38}$ erg s$^{-1}$. However, the published XLFs were produced using a number of methods, in most cases assuming Galactic line-of-sight absorption. We have tested several of these models using a deep XMM-Newton observation of the nearby galaxy NGC253, included in a secondary sample of Grimm et al. (2003). We obtained freely modelled luminosities for the 140 brightest sources in the field and also obtained the conversion factors from intensity to flux for some of these different models. We found them to vary by a factor of $\sim$3. We found the biggest influence on the conversion factor to be the absorption; Model I assumed Galactic line-of-sight absorption, while absorptions 20--50 times higher than this were obtained for the other methods. Since the universal XLF and relations between SFR and X-ray properties were obtained using a mixture of methods, we find them to be inconclusive. We have also found from freely fitting the spectra of 140 bright sources that the corresponence between count-rate and flux is non-linear; this is largely due to the systematic softening of the spectra of more luminous sources. Hence it is unwise to employ a single emission model when describing the X-ray populations of nearby galaxies. Ideally, one would construct XLFs only from luminosities derived from free spectral modelling. To get the most out of the low-photon data, we recommend the stacking method of e.g. Roberts et al. (2002). | 7 | 10 | 0710.4893 |
0710 | 0710.4588_arXiv.txt | {Stability properties of magnetic-field configurations containing the toroidal and axial field are considered. The stability is treated by making use of linear analysis. It is shown that the conditions required for the onset of instability are essentially different from those given by the necessary condition $d (s B_{\varphi})/ds > 0$, where $s$ is the cylindrical radius. The growth rate of instability is calculated for a wide range of the parameters. We argue that the instability can operate in two different regimes depending on the strength of the axial field and the profile of the toroidal field. | Turbulence generated by MHD instabilities can play an important role in enhancing transport processes in various astrophysical bodies, such as accretion and protoplanetary disks, galaxies, stellar radiative zones, etc. The anomalous turbulent transport can be particularly important in magnetized gas where a wide variety of MHD instabilities can occur (see, e.g., Barnes et al. 1999). In this case, the onset of instability can be caused both by hydrodynamic motions (for instance, differential rotation; see, e.g., Velikhov 1959; Chandrasekhar 1960) or unstable magnetic configurations. { Which field strength and topology can sustain a stable magnetic configuration is still rather uncertain despite extensive work (see Borra et al. 1982; Mestel 1999 for review).} Most likely, the best-studied magnetic configuration is one with a purely toroidal field. Ever since the paper by Tayler (1973), it has been known that toroidal fields can be unstable close to the axis of symmetry, if there is a non-zero electric current density on the axis. The growth rate of this instability is expected to be of the order of the time taken for an Alfv\'en wave to travel around the star on a toroidal field line. However, even a purely toroidal field is stable if it decreases rapidly with the cylindrical radius $s$. For instance, Tayler (1973; see also Chanmugam 1979) argued that the toroidal field $B_{\varphi}$ is stable against axisymmetric perturbations if it satisfies the condition $d ( B_{\varphi}/s)/ds < 0$ and to non-axisymmetric perturbations if $d (s B_{\varphi}^2)/ds <0$. Note that a purely toroidal field can also be subject to the magnetic buoyancy instability (Parker 1955; Gilman 1970; Acheson 1978) but the Tayler instability likely appears first as the strength of the toroidal field increases (Spruit 1999). The stability of a purely toroidal field in the radiative zones of stars and accretion disks has been studied by a number of authors. Numerical modeling by Braithwaite (2006) confirms that the toroidal field with $B_{\varphi} \propto s$ or $\propto s^2$ is unstable to the $m=1$ mode ($m$ is the azimuthal wave number) as predicted by Tayler (1973). The linear stability of the toroidal field in rotating stellar interiors has been considered by Kitchatinov \& R\"udiger (2007), who conclude that the magnetic instability is essentially three-dimensional and that the finite thermal conductivity has a strong destabilizing effect. Terquem \& Papaloizou (1996) and Papaloizou \& Terquem (1997) considered the stability of an accretion disk with an embedded toroidal magnetic field. These authors find that the disks containing a purely toroidal field are always unstable and obtained spectra of unstable modes in the local approximation. They argue that one type of modes is driven primarily by buoyancy, while the other is driven by shear independently of the magnetic configuration. { Stability properties of purely poloidal magnetic configurations have also been well-studied. Since the papers by Wright (1973) and Markey \& Tayler (1973, 1974), it is known that a poloidal field is subject to dynamical instabilities in the neighbourhood of neutral points/lines if the field lines are closed inside the star. These authors recognizes that the magnetic field in the neighbourhood of a neutral line resembles that of a toroidal pinched discharge that is known to be unstable. Although instabilities involving significant displacements in the direction of gravity were strongy inhibited, other instabilities were not affected. The instability of poloidal configurations is rather fast: its growth time can reach a few Alfv\'en crossing time (Van Assche et al. 1982; Braithwaite \& Spruit 2006). With numerical simulations, the stability of poloidal magnetic configurations has been studied by Braithwaite \& Spruit (2006), who apply the results to the internal magnetic configuration of neutron stars. Note, however, that a toroidal field might exert a stabilising influence on the instabilities of a poloidal field in the neighbourhood of neutral points (Tayler 1980).} { On the contrary, the addition of even a relatively weak poloidal field alters the stability of the toroidal field substantially. For example, as first shown by Howard \& Gupta (1962; see also Knobloch 1992; Dubrulle \& Knobloch 1993), a necessary (but not sufficient) condition for the instability of a toroidal field in the presence of the axial field reads} \begin{equation}\label{1} \frac{d}{ds} (s B_{\varphi}) > 0. \end{equation} Howard \& Gupta (1962) argue that, for a fixed value of $m$, the growth rate of instability caused by condition (1) must vanish in the limit of a vanishing axial magnetic field, thereby providing a connection with the stability criterion obtained by Tayler (1973). Note that the presence of a radial field is also crucial for stability properties of rotating magnetic configurations (Bonanno \& Urpin 2006). { It turns out that configurations containing both toroidal and poloidal fields are more stable than purely toroidal or purely poloidal ones (Prendergast 1956; Tayler 1980). With numerical simulations, Braithwaite \& Nordlund (2006) studied the stability of a random initial field in the stellar radiative zone. The star was modeled on a Cartesian grid, and the authors found that the stable magnetic configurations generally have the form of tori with comparable poloidal and toroidal field strengths.} In the present paper, we address the stability properties of magnetic configurations by considering the stability of the toroidal magnetic field with respect to axisymmetric perturbations in the presence of axial fields of a various strength. We show that the instability may occur in such magnetic field configurations under the conditions that differ substantially from those imposed by the Tayler criterion or the necessary condition (\ref{1}). We argue that, in some cases, the instability is caused by the new type of MHD waves with the growth rate proportional to $\sqrt{B_z B_{\varphi}}$ where $B_z$ is the axial magnetic field. Depending on the profile $B_{\varphi}(s)$ and the ratio $B_z/B_{\varphi}$, the instability can occur in two regimes that have substantially different growth rates. We also show that the range of unstable wavelength in the $z$-direction can be essentially different, depending on the $B_z/B_{\varphi}$ ratio. | { We have considered the hydromagnetic stability of cylindrical configurations containing the toroidal and axial magnetic fields. Dissipative effects were neglected in our study. We treated a linear stability assuming that the behaviour of small perturbations is governed by equations of incompressible hydrodynamics. This approximation is justified if the magnetic field is subthermal and the Alfv\'en velocity is low compared to the sound speed.} The stability of the magnetic configurations is a key issue for understanding the properties of various astrophysical bodies, such as peculiar A and B stars, magnetic white dwarfs, neutron stars, etc. { Even though various dynamo models predict that the toroidal field should be typically stronger than the poloidal one, the effect of a poloidal field on the stability usually cannot be neglected.} To demonstrate this, we treated the simplest model of a highly conducting fluid between two cylindrical surfaces. We assumed that the toroidal and axial fields depends on the cylindrical radius alone. In a short-wavelength approximation, we derived the growth rate and a sufficient criterion of instability analytically (Eqs.~(30)-(31)). For large-scale perturbations, the condition of instability and its growth rate were calculated numerically. The analytical and numerical results are in good qualitative agreement. The obtained conditions of instability differ substantially from what is predicted by the necessary condition (1). For instance, according to Eq.~(1), if the instability occurs in the magnetic configuration, then the toroidal field profile satisfies the condition $\alpha > -1$. In fact, the instability occurs only if the toroidal field decreases with $s$ much slower (or even increases): the critical value of $\alpha$ is $\approx -0.1$ if $B_z/ B_{\varphi 0} =0.1$ and $\approx 1$ if $B_z/ B_{\varphi 0} =0.5$. { If $B_{z}$ depends on $s$, then the critical values of $\alpha$ should be even higher.} Depending on the profile of the toroidal field and the strength of the axial field, the instability can arise in two essentially different regimes. In the case of a weak axial field, $B_{\varphi 0} \gg B_z$, the value of $\alpha$ that distinguishes between the regimes is $\approx 1$. If $\alpha > 1$, then the instability grows on the Alfv\'en timescale determined by the toroidal field and is rather fast. If $\alpha < 1$, then the instability is much slower and grows on the timescale determined by the axial field. The transition between these two regimes occurs at larger $\alpha$ if the axial field increases. The efficiency of the considered instability turns out to be rather low if $\alpha < 1$ and $B_z$ is weak. It is worth noticing the very particular properties of instability in the case $\alpha \approx 1$. In such a configuration, the particular type of MHD waves is given by the dispersion equation (19). The growth rate of these waves (or the frequency, if the waves are stable) is proportional to the product of $B_z$ and $B_{\varphi}$. These waves cannot exist in purely toroidal or purely poloidal fields. The instability of the configuration with $\alpha \approx 1$ is caused by the generation of this particular type of wave. { These waves can probably determine the instability of magnetic configurations near the axis of symmetry where $B_{\varphi} \rightarrow 0$.} A sufficiently strong axial field always suppresses the instability. For more or less plausible values of $\alpha \leq 1$, the strength of the axial field stabilising the configuration is $\sim 0.1-1 B_{\varphi}$. A much stronger field is required, however, to stabilise the configuration with larger $\alpha$. \vspace{0.5cm} \noindent {\it Acknowledgments.} This research project was supported by a Marie Curie Transfer of Knowledge Fellowship of the European Community's Sixth Framework Programme under contract number MTKD-CT-002995. VU thanks also INAF-Ossevatorio Astrofisico di Catania for hospitality. | 7 | 10 | 0710.4588 |
0710 | 0710.3682_arXiv.txt | We report on the latest (2007 Jan) observations of supernova remnant (SNR) 1987A from the {\it XMM-Newton} mission. Since the 2003 May observations of Haberl et al. (2006), 11 emission lines have experienced increases in flux by factors $\sim 3$ to 10 ($6 \pm 0.6$ on average), with the 775 eV line of O~{\sc viii} showing the greatest increase. Overall, we are able to make Gaussian fits to 17 emission lines in the {\it RGS} spectra and obtain line fluxes; we have observed 6 lines of Fe~{\sc xvii} and Fe~{\sc xviii} previously unreported by {\it XMM-Newton}. A two-shock model representing plasmas in non-equilibrium ionization is fitted to the {\it EPIC-pn} spectra, yielding temperatures of $\sim 0.4$ and $\sim 3$ keV, as well as elemental abundances for N, O, Ne, Mg, Si, S and Fe. We demonstrate that the abundance ratio of N and O can be constrained to $\lesssim 20\%$ accuracy $\left(\mbox{N/O} = 1.17 \pm 0.20 \right)$. Within the same confidence interval, the same analysis suggests that the C+N+O abundance varies from $\sim 1.1$ to $1.4 \times 10^{-4}$, verifying the {\it Chandra} finding by Zhekov et al. (2006) that the C+N+O abundance is lower by a factor $\sim 2$ compared to the value obtained in the optical/ultraviolet study by Lundqvist \& Fransson (1996). Normalizing our obtained abundances by the Large Magellanic Cloud (LMC) values of Hughes, Hayashi \& Koyama (1998), we find that O, Ne, Mg and Fe are under-abundant, while Si and S are over-abundant, consistent with the findings of Aschenbach (2007). Such a result has implications for both the single-star and binary accretion/merger models for the progenitor of SNR 1987A. In the context of the binary merger scenario proposed by Morris \& Podsiadlowski (2006, 2007), material forming the inner, equatorial ring was expelled after the merger, implying that either our derived Fe abundance is inconsistent with typical LMC values or that iron is under-abundant at the site of Sanduleak -69$^\circ$202. | \label{sect:intro} SNR 1987A is the Rosetta Stone (Allen 1960) of Type II supernova remnants, resolved and well-studied in multiple wavebands, including the infrared (Bouchet et al. 2006; Kjaer et al. 2007), optical/ultraviolet (Gr\"{o}ningsson et al. 2007; Heng 2007) and radio (Gaensler et al. 2007). The physical mechanism partially powering optical/ultraviolet emission from the reverse shock is the same as the one at work in Balmer-dominated supernova remnants (Heng \& McCray 2007; Heng et al. 2007). The detection of a neutrino burst confirmed the core collapse nature of the progenitor (Koshiba et al. 1987; Svoboda et al. 1987), though a pulsar has yet to be detected (Manchester 2007). A system of three rings may be the result of a binary merger between two massive stars about 20,000 years prior to the supernova explosion (Morris \& Podsiadlowski 2006, 2007; hereafter MP0607). Reviews of the multi-wavelength studies of SNR 1987A can be found in McCray (1993, 2005, 2007). Mixing of the stellar envelope and core by Rayleigh-Taylor instabilities within the progenitor star, Sanduleak -69$^\circ$202, has been invoked to explain the early emergence of the 847 keV $\gamma$-ray line from SNR 1987A, which was predicted by Shibazaki \& Ebisuzaki (1988) to reach its peak around 1.1 years after the explosion, if one assumes a mixed mass of about 5$M_\sun$ (Ebisuzaki \& Shibazaki 1988). Instead, Matz et al. (1988) observed the 847 keV line $\sim$ 6 to 8 months post-explosion, suggesting even more extensive mixing of $^{56}$Co than assumed. A similar explanation (Ebisuzaki \& Shibazaki 1988) was given for the early emergence of 16 to 28 keV X-rays (Sunyaev et al. 1990; Inoue et al. 1991). The $\gamma$-rays originate from the radioactive decay of $^{56}$Co, while the X-rays are from the Compton degradation of the $\gamma$-rays (McCray, Shull \& Sutherland 1987). In the soft X-ray, SNR 1987A was first observed by Beuermann, Brandt \& Pietsch (1994). Subsequently, Hasinger, Aschenbach \& Tr\"{u}mper (1996) tracked a steady increase of the soft X-ray flux over 4 years with {\it ROSAT}. Extensive work has since been done by the {\it Chandra} (Michael et al. 2002; Park et al. 2002, 2004, 2005 [hereafter P05], 2006, 2007 [hereafter P07]; Zhekov et al. 2005, 2006 [hereafter Z06]) and {\it XMM-Newton} groups (Haberl et al. 2006, hereafter H06; Aschenbach 2007, hereafter A07). The general picture gleaned from these studies is of a bimodal plasma distribution present in the region between the forward and reverse shocks (Fig. \ref{fig:87a}). The soft X-rays ($\sim$ 0.3 to 0.5 keV) are from the decelerated shock front interacting with dense protrusions (``fingers'') on the inner, equatorial ring, while the ``hard'' X-rays ($\sim$ 2 to 3 keV) are from a fast shock propagating into more tenuous material. The soft X-rays appear to be correlated with optical ``hot spots'', believed to be emission from the shocked fingers, appearing around the equatorial ring, while the hard X-ray and radio images exhibit structures that coincide. Between Days 6000 and 6200, the soft X-ray light curve experienced an upturn and departure from an exponentially increasing profile, which P05 interpreted as evidence that the blast wave had reached the main body of the dense circumstellar material of the equatorial ring. Future studies on the nature of the soft X-ray light curve are relevant to the issue of pre-ionization of the supernova ejecta, which can potentially extinguish H$\alpha$ and Ly$\alpha$ emission from the reverse shock (Smith et al. 2005; Heng et al. 2006). In the studies described, relatively little attention has been paid to the subject of inferring elemental abundances from X-ray analyses. Fits to the X-ray spectra yield N, O, Ne, Mg, S, Si and Fe abundances; such results have been tabulated and studied by Z06 and H06, using {\it Chandra} and {\it XMM}, respectively. Here, we examine such an approach using the latest {\it XMM} data set of SNR 1987A, taken in early 2007. In \S\ref{sect:data}, we describe our observations and data reduction techniques. Our results are presented in \S\ref{sect:results}. In \S\ref{sect:discussion}, we perform a detailed error analysis of individual abundances and their ratios. We find that the N/O ratio as well as the individual N and O abundances can be constrained to $\sim 20\%$ accuracy. Our derived elemental abundances for O, Ne, Mg and Fe are under-abundant, while Si and S are over-abundant, relative to typical values for the Large Magellanic Cloud (LMC; Hughes, Hayashi \& Koyama 1998). Such a result has implications for modeling the progenitor of SNR 1987A, which we discuss. | \label{sect:discussion} \subsection{INDIVIDUAL ABUNDANCES \& THEIR RATIOS} \label{subsect:abund} We compare the derived elemental abundances to those of Z06 and H06 and list them {\it relative to hydrogen} in Table \ref{tab:abundances}; uncertainties in the AG89 and W00 abundance tables are not propagated. We emphasize that care must be taken to specify the abundance table used, as this may lead to widely differing values of the derived abundances (relative to hydrogen). In modeling the LMC absorption, Z06 used the elemental abundance table of AG89, while H06 chose the table of W00 because the lower oxygen abundance fitted the K absorption edge in the {\it EPIC} data better. Fransson et al. (1989) found N/O = 1.6 $\pm$ 0.8, about 12 times higher\footnote{The AG89 and W00 values for solar nitrogen-to-oxygen abundance are N/O = 0.132 and 0.110 by number, respectively.} than the AG89 solar value, which they interpret as evidence of substantial CNO processing. LF96 found N/O = 1.1 $\pm$ 0.4, while Sonneborn et al. (1997) found N/O = 1.7 $\pm$ 0.5. All three of these values were derived from optical/ultraviolet data. Next, we turn our attention to the N/O ratio derived from X-ray studies. Linearly propagating the errors listed by Z06, we find that their results yield N/O = $1.10^{+0.47}_{-0.45}$; they remark that their derived C, N and O abundance, C+N+O $\approx 1.98 \times 10^{-4}$, is lower by about a factor of 2 compared to the $3.72 \times 10^{-4}$ value of LF96. Model A ({\tt VNEI+VRAYMOND}) in H06 yields C+N+O $\approx 1.67 \times 10^{-4}$ and a rather wide range in the nitrogen-to-oxygen ratio, N/O = $1.33^{+1.47}_{-0.71}$. Our {\tt VPSHOCK+VPSHOCK} fit to the {\it EPIC-pn} data yields C+N+O $\approx 1.29 \times 10^{-4}$ and N/O = $1.17^{+0.37}_{-0.34}$ (using the W00 table); we call this combination of N and O the ``best fit point''. Note that the C abundance is held fixed at 0.09 relative to solar ($\sim 3 \times 10^{-5}$ relative to hydrogen) for the C+N+O values derived from the X-ray studies. We next perform a more careful analysis of the N/O ratio. We first generate a $\chi^2$ map quantifying the inter-dependence of the fits to the N and O abundance. Contour lines in the $\chi^2$ map form ``error ellipses'', which are shown for different $\Delta \chi^2$ values from the best fit point (Fig. \ref{fig:NO}). At the $\Delta \chi^2 = 2.706$ level, N/O $= 1.17 \pm 0.20$ for the {\it EPIC-pn} data. In linearly propagating the errors in the individual abundances, one is in essence adopting the largest possible range of ratios, which can be visualized as the edges of a rectangle in the contour map. Our error analysis improves the uncertainties because it considers only values of the abundance ratio within the specified contour. Within the same confidence interval considered, the corresponding C+N+O value is from $\sim$ 1.1 to $1.4 \times 10^{-4}$. We see that the errors in the N/O ratio and the individual N and O abundances can be constrained at the $\sim 20\%$ level. We thus confirm the C+N+O under-abundance noted by the {\it Chandra} studies of Z06, who suggest a couple of physical reasons for such a result: the sub-LMC abundance of C+N+O within the progenitor star, Sanduleak -69$^\circ$202, and/or an extra source of possibly non-thermal, X-ray continuum. We perform the same analysis for N/S (Fig. \ref{fig:othercontours}). Again using the W00 table, linear propagation of the errors in the abundances obtained from the {\tt VPSHOCK+VPSHOCK} fit yields N/S = $4.59^{+1.51}_{-1.36}$, while the error ellipse analysis gives N/S = $4.59^{+1.59}_{-1.33}$. The iron, nitrogen and oxygen lines are predominantly from the lower energy part ($\lesssim 1$ keV) of the spectrum, while the sulphur lines are situated between $\sim 2.2$ and 3.1 keV. (The silicon lines are located between $\sim 1.8$ and 2.2 keV.) Abundance ratios based on lines of widely differing energies lead to rounder error ellipses --- ``error circles''. In such cases, a more thorough error analysis will not constrain the abundance ratio better, as in the case of N/S. This is partially an instrumental effect --- when the energy resolution is comparable to the spacing of the lines, the line complexes overlap and are only partially resolved. There is also the issue of choosing a coordinate system in which the fitting parameters are ``orthogonal''. Hydrogen and helium have no X-ray lines --- their abundances are derived from the strength of the X-ray continuum. When the pair of elements considered are situated close to each other in energy, the hydrogen abundance has to first order a linear dependence on the continuum strength. Thus, the ratio of the considered abundances is tightly constrained as the individual abundances track each other closely. By contrast, when the pair of abundances considered are located far apart in energy, this linear dependence of hydrogen abundance on the continuum is broken as the dominant uncertainties are in temperature rather than in flux. Orthogonality is now absent. To attain orthogonality for such pairs of lines, one has to construct models that directly fit to the abundance ratio considered, an approach which is not explored in this paper. The error in the abundance ratios increases as one moves a given $\Delta \chi^2$ away from the minimum point. In Fig. \ref{fig:ratioerror}, we compute the mean error sustained by the various ratios as a function of $\Delta \chi^2$. According to Avni (1976), if three interesting parameters are considered (see \S\ref{subsect:spectral}), the 90\% confidence interval is situated at $\Delta \chi^2 = 6.25$. In this case, the N/O abundance ratio suffers from errors $\sim 25\%$. \subsection{THE PROGENITOR OF SNR 1987A: SINGLE-STAR OR BINARY MODEL?} \label{subsect:metallicity} A more revealing approach to analyze the elemental abundances is to normalize the results listed in Table \ref{tab:abundances} by the ``canonical'' values of the LMC abundances (Hughes, Hayashi \& Koyama 1998). These were derived using a sample of 7 middle-aged SNRs in the LMC (N23, N49, N63A, DEM 71, N132D, 0453-68.5 and N49B). This approach was first explored by A07, who showed that the normalized abundances appear to cluster in two groups: N, O, Ne, Mg and Fe are slightly more than half their LMC values, while Si, S and Ni exceed their LMC values. We generalize the A07 approach by considering both sets of abundances derived from using the AG89 and W00 tables. The normalized abundance, relative to its respective LMC value, is $R_{\rm{87A/LMC}}$; it is plotted as a function of the elemental mass number in Fig. \ref{fig:abundratios}. The error bars for $R_{\rm{87A/LMC}}$ are computed by linearly propagating the errors listed in Table \ref{tab:abundances} and in Hughes, Hayashi \& Koyama (1998). We caution that additional systematic errors may be present that are not taken into account. For example, the derived abundance for Fe is a sensitive function of temperature and may vary substantially when small changes are made to $T_{\rm{low}}$ \footnote{The Fe abundance obtained from the {\tt VPSHOCK+VPSHOCK} fit (W00 table) varies by $\sim 16\%$ when $T_{\rm{low}}$ is changed by $\sim 10\%$.}. We see that the elements O, Ne, Mg and Fe are under-abundant, while Si and S are over-abundant, consistent with the findings of A07. With the exception of Fe, there is a tendency for $R_{\rm{87A/LMC}}$ to increase with larger elemental mass number, a trend that is independent of the AG89 or W00 tables, though we note that it is more pronounced with the latter. The Fe abundance derived is essentially independent of the AG89 or W00 tables, and is under-abundant by about 70\% relative to the LMC \footnote{The O abundance relative to H for the AG89 and W00 tables are the same.}. The under-abundance of Fe and O was previously noted by Hasinger et al. (2006), who argued for the existence of iron-oxygen ``rust grains''. The reduced abundance of Fe alone suggests that the iron is locked up in dust grains. However, Dwek \& Arendt (2007) showed from an analysis of the infrared-to-X-ray flux ratio --- ${\cal R}_{\rm{IRX}} < 1$ versus the theoretically expected value of $\sim 10^2$ to $10^3$ --- that the dust in SNR 1987A is severely depleted compared to standard dust-to-gas mass ratios in the LMC, suggesting low dust condensation efficiency or dust destruction in the hot X-ray gas. In fact, ${\cal R}_{\rm{IRX}}$ was shown to {\it decrease} with time, which is direct evidence for dust destruction. Our derived plasma temperatures are consistent with this scenario --- even if dust could form, it would be destroyed at these temperatures. In light of Fig. \ref{fig:abundratios}, the central question to ask is whether the progenitor of SNR 1987A arose from a single star or a binary system? Sanduleak -69$^\circ$202 was known to be a blue supergiant (BSG) at the time of the supernova explosion, contrary to the expectation that massive stars end their lives as red supergiants (RSGs). Observations of low-velocity, nitrogen-rich circumstellar material are interpreted as the progenitor star being a RSG until about $\sim 20,000$ years before its death (Fransson et al. 1989). Such a time scale has in turn been interpreted as the Kelvin-Helmholtz time of the helium core (Woosley et al. 1997). This BSG-RSG-BSG evolution remains one of the greatest challenges for the single-star model (Woosley et al. 1997; Woosley, Heger \& Weaver 2002), as is the observed system of three rings produced $\sim 20,000$ years before the explosion. The favored single-star models require a combination of reduced metallicity and ``restricted semi-convection'' (Woosley 1988), the former of which is supported by our derived abundance ratios. An additional challenge for the single-star model is to reproduce the over-abundance of Si and S (Fig. \ref{fig:abundratios}), which may require the invoking of some non-standard mixing process (H.-T. Janka 2007, private communication). Binary solutions to Sanduleak -69$^\circ$202 are sub-divided into accretion and merger models (see Woosley, Heger \& Weaver [2002] and references therein). The binary accretion models allow helium- and nitrogen-rich material to be added to the progenitor star and require the disappearance of the mass donor in an earlier supernova event. In the binary merger scenario proposed by MP0607, two stars with masses $\sim 5 M_\sun$ and $\sim 15 M_\sun$ are initially orbiting each other with a period $\sim 10$ years. The more massive companion transfers mass to the less massive star only after the former has completed helium burning in the core. A common envelope (Paczy\'{n}ski 1976) is formed, during which core material from the primary star is dredged up to the surface. The merger process takes a few hundred years, culminating in an initially over-sized RSG, which loses its excess thermal energy over a few thousand years to become a BSG. The spun-up, rapidly rotating BSG produces a fast stellar wind, sweeping up ejecta associated with the merger, producing the triple ring nebula we now see in projection. The nearly axi-symmetric but highly non-spherical nature of the rings suggests that rotation played a role in their formation and is consistent with the proposed scenario. The beauty of the model lies in the fact that it requires no physically ad hoc assumptions --- apart from a small kick of $\sim 2$ km s$^{-1}$ given to the ejecta to displace the center of the outer rings from their symmetry axis --- and makes a number of predictions. In their favored model, MP0607 assert that the outer rings are ejected before the stellar core material is dredged up, while the inner, equatorial ring --- the site of the observed X-ray emission --- is ejected {\it afterwards} \footnote{As noted by MP0607, this hypothesis is verifiable/refutable, as the inner ring should exhibit helium enhancement and more CNO processing relative to the outer rings.}. A relevant consequence of this model is that the dredged-up heavy elements will manifest themselves in the form of X-ray emission lines. This may explain the trend we see in Fig. \ref{fig:abundratios} --- the challenge for binary merger models is to reproduce the derived $R_{\rm{87A/LMC}}$ values. Stellar nucleosynthesis can add and {\it not} subtract iron --- in the context of the MP0607 binary merger model, we expect $R_{\rm{87A/LMC}}$(Fe) $\ge 1$. We are again led to the question: where is the iron? If the Fe abundance derived is an upper limit on the iron used to form Sanduleak -69$^\circ$202, then it is clearly inconsistent with ``standard'' Fe abundances in the LMC. An alternative interpretation is that there is a strong spatial variation in the Fe abundance throughout the LMC, such that the iron is sub-LMC at the site of SN 1987A and equal to its LMC abundance elsewhere. The C+N+O under-abundance suggested in \S\ref{subsect:abund} supports such a conclusion. An improvement over using the Hughes, Hayashi \& Koyama (1998) {\it ASCA} abundance values is to re-analyze and expand upon their SNR sample using {\it Chandra} and {\it XMM}. Existing studies (e.g., Hughes et al. 2006) tend to pick out regions of interest that may include ejecta enrichment; instead, X-ray emission should be extracted from the {\it entire} blast wave, from which average abundances can be inferred. Future studies will be invaluable towards resolving these issues. \scriptsize K.H. acknowledges the kind hospitality and financial support of: the Max Planck Institutes for Astrophysics (MPA) and Extraterrestrial Physics (MPE) during June to October 2007, where he was a visiting postdoctoral scientist; and the Lorentz Center (Leiden) during the August 2007 workshop, ``From Massive Stars to Supernova Remnants''. He thanks Dick McCray, Sangwook Park, Svet Zhekov, Jack Hughes, John Raymond, Philipp Podsiadlowski, Rashid Sunyaev, Peter Lundqvist, Claes Fransson, Dmitrijs Docenko, Thomas Janka, Carlos Badenes, Roger Chevalier, Nathan Smith and Jacco Vink for engaging and helpful discussions. Special mentions go out to: Dick and John, who provided a crash course on non-equilibrium ionization plasmas during a sunny bicycle ride in Leiden; Svet, who pointed out relevant material on X-ray physics, as well as for critical comments following his meticulous scrutiny of the manuscript. The authors are collectively grateful to Kazik Borkowski for providing the updated {\tt VPSHOCK} model files for use in {\tt XSPEC}. The {\it XMM-Newton} project is supported by the {\it Bundesministerium f\"{u}r Wirtschaft und Technologie/Deutsches Zentrum f\"{u}r Luft- und Raumfahrt} (BMWI/DLR, FKZ 50 OX 001) and the Max Planck Society. \normalsize | 7 | 10 | 0710.3682 |
0710 | 0710.4294_arXiv.txt | Infrared surveys indicate that the dust content in debris disks gradually declines with stellar age. We simulated the long-term collisional depletion of debris disks around solar-type (G2~V) stars with our collisional code. The numerical results were supplemented by, and interpreted through, a new analytic model. General scaling rules for the disk evolution are suggested. The timescale of the collisional evolution is inversely proportional to the initial disk mass and scales with radial distance as $r^{4.3}$ and with eccentricities of planetesimals as $e^{-2.3}$. Further, we show that at actual ages of debris disks between 10~Myr and 10~Gyr, the decay laws of the dust mass and the total disk mass are different. The reason is that the collisional lifetime of planetesimals is size-dependent. At any moment, there exists a transitional size, which separates larger objects that still retain the ``primordial'' size distribution set in the growth phase from smaller objects whose size distribution is already set by disruptive collisions. The dust mass and its decay rate evolve as that transition affects objects of ever-larger sizes. Under standard assumptions, the dust mass, fractional luminosity, and thermal fluxes all decrease as $t^\xi$ with $\xi = -0.3$...$-0.4$. Specific decay laws of the total disk mass and the dust mass, including the value of $\xi$, largely depend on a few model parameters, such as the critical fragmentation energy as a function of size, the primordial size distribution of largest planetesimals, as well as the characteristic eccentricity and inclination of their orbits. With standard material prescriptions and a distribution of disk masses and extents, a synthetic population of disks generated with our analytic model agrees quite well with the observed Spitzer/MIPS statistics of 24 and 70 \micron\ fluxes and colors versus age. | Since the IRAS discovery of the excess infrared emission around Vega by \citet{aumann-et-al-1984}, subsequent infrared surveys with ISO, Spitzer and other instruments have shown the Vega phenomenon to be common for main-sequence stars. The observed excess is attributed to second-generation circumstellar dust, produced in a collisional cascade from planetesimals and comets down to smallest grains that are blown away by the stellar radiation. While the bulk of such a debris disk's mass is hidden in invisible parent bodies, the observed luminosity is dominated by small particles at dust sizes. Hence the studies of dust emission offer a natural tool to gain insight into the properties of planetesimal populations as well as planets that may shape them and, ultimately, into the evolutionary history of circumstellar planetary systems. In recent years, various photometric surveys of hundreds of nearby stars have been conducted with the Spitzer Space Telescope. These are the GTO survey of FGK stars \citep{beichman-et-al-2005,bryden-et-al-2006,beichman-et-al-2006b}, the FEPS Legacy project \citep{meyer-et-al-2004,kim-et-al-2005}, the A star GTO programs \citep{rieke-et-al-2005,su-et-al-2006}, the young cluster programs \citep{gorlova-et-al-2006}, and others. These observations were done mostly at 24 and 70~\textmu{m} with the MIPS photometer, but also between 5 and 40~\textmu{m} with the IRS spectrometer \citep{jura-et-al-2004,chen-et-al-2006}. Based on these studies, about 15\% of mature solar-type (F0--K0) stars have been found to harbor cold debris disks at 70~\textmu{m}. For cooler stars, the fraction drops to 0\%--4\% \citep{beichman-et-al-2006b}. For earlier spectral types, the proportion increases to about 33\% \citep{su-et-al-2006}. At 24~\textmu{m}, the fraction of systems with detected excess stays similar for A~stars, but appreciably decreases for FGK ones. Similar results in the sub-millimeter range are expected to become available soon from a survey with SCUBA and SCUBA2 on JCMT \citep{matthews-et-al-2007}. Preliminary SCUBA results for M dwarfs suggest, in particular, that the proportion of debris disks might actually be higher than suggested by Spitzer \citep{lestrade-et-al-2006}. All authors point out a decay of the observed infrared excesses with systems' age. However, the values reported for the slope of the decay, assuming a power-law dependence $t^{-\alpha}$, span a wide range. \citet{greaves-wyatt-2003} suggest $\alpha \la 0.5$, \citet{liu-et-al-2004} give $0.5 < \alpha < 1.0$, \citet{spangler-et-al-2001} report $\alpha \approx 1.8$, and \citet{greaves-2005} and \citet{moor-et-al-2006} derive $\alpha \approx 1.0$. Fits of the upper envelope of the distribution of luminosities over the age yield $\alpha \approx 1.0$ as well \citep{rieke-et-al-2005}. Besides, the dust fractional luminosity exhibits a large dispersion at any given age. In an attempt to gain theoretical understanding of the observed evolution, \citet{dominik-decin-2003} assumed that equally-sized ``comets'' produce dust through a cascade of subsequent collisions among ever-smaller objects. If this dust is removed by the same mechanism, the steady-state amount of dust in such a system is proportional to the number of comets. This results in an $M/M_0 \approx \tau/t$ dependence for the amount of dust and for the number of comets or the total mass of the disk. Under the assumption of a steady state, this result is valid even for more complex systems with continuous size distributions from planetesimals to dust. Tenuous disks, where the lifetime of dust grains is not limited by collisions but by transport processes like the Poynting-Robertson drag \citep{artymowicz-1997,krivov-et-al-2000,wyatt-2005}, follow $M \propto t^{-2}$ rather than $M \propto t^{-1}$. More recently, \citet{wyatt-et-al-2007a} lifted the most severe simplifying assumption of the Dominik-Decin model, that of equal-sized parent bodies, and included them into the collisional cascade. A debris disk they consider is no longer a two-component system ``comets + dust''. Instead, it is a population of solids with a continuous size distribution, from planetesimals down to dust. A key parameter of the description by \citet{dominik-decin-2003} is the collisional lifetime of comets, $\tau$. \citet{wyatt-et-al-2007a} replaced it with the lifetime of the largest planetesimals and worked out the dependencies on this parameter in great detail. Since the collisional timescale is inversely proportional to the amount of material, $\tau \propto 1/M_0$, the asymptotic disk mass becomes independent of its initial mass. Only dynamical quantities, i.e. the disk's radial position and extent, the orbiting objects' eccentricities and inclinations, and material properties, i.e. the critical specific energy and the disruption threshold, as well as the type of the central star determine the very-long-term evolution. Still, there are two important simplifications made in the model by \citet{wyatt-et-al-2007a}: (i) the disk is assumed to be in collisional equilibrium at all sizes, from dust up to the largest planetesimals and (ii) the minimum specific energy needed to disrupt colliding objects is independent of their size. As a consequence of (i) and (ii), the size distribution of solids is a single power-law. To check how reasonable these assumptions are, realistic simulations of the disks with collisional codes are necessary \citep[e.g.,][]{thebault-et-al-2003,krivov-et-al-2005,krivov-et-al-2006,thebault-augereau-2007}. The aim of this paper is two-fold. First, we follow the evolution of debris disks with our elaborate numerical code \citep{krivov-et-al-2005,krivov-et-al-2006} to check the existing analytic models and the assumptions (i) and (ii) they are based upon. Second, in order to make these numerical results easier to use, we develop a new analytic model for the evolution of disk mass and dust mass that relaxes both assumptions (i) and (ii) above. Section~\ref{secNumerics} summarizes the basic ideas and assumptions and describes our numerical model and the runs of the collisional code. In Section~\ref{secScalings} the numerical results are presented and dependences of the collisional timescale on the disk mass, distance to the star, and mean eccentricity of parent bodies are derived. In section~\ref{secAnalytics}, the analytic model for the evolution of disk mass and dust mass is developed. Section~\ref{secEvolLuminosity} analyzes the evolution of dust luminosities. In Section~\ref{secObservations}, we use the analytic model to synthesize representative populations of debris disks and compare them with statistics of debris disks derived from the Spitzer surveys. A summary is given and conclusions are drawn in Section~\ref{secConclusions}. \pagebreak | \begin{enumerate} \item The timescale of the collisional evolution is inversely proportional to the initial disk mass. For example, halving the total mass doubles all collisional timescales. This rule is valid for systems where collisions are the only loss mechanism of particles and only as long as $\beta$-meteoroids are unimportant for the collisional budget. \item Numerics and analytics consistently yield a $\tau\propto r^{4.3}$ dependence of the timescale of the collisional evolution on the radial distance. \item Numerical simulations show that the collisional timescale varies with the average eccentricity of dust parent bodies as $\tau \propto e^{-2.3}$. The analytic approach suggests a somewhat weaker dependence, $\tau\propto e^{-5/3}$. \item An evolving three-slope size distribution is proposed to approximate the numerical results. The biggest objects are still distributed primordially, with a slope $q\sbs{p}$. The objects below a certain transitional size are already reprocessed by collisions and thus have a quasi-steady-state size distribution, determined by their self-gravity (for intermediate-sized objects, slope $q\sbs{g}$) or by material strength (for smallest objects, slope $q\sbs{s}$). That transitional size corresponds to the largest objects for which the collisional lifetime is still shorter than the age of the system. The transitional size increases with time, meaning that ever-larger planetesimals get involved into the collisional cascade. \item At actual ages of debris disks, $\sim$10~Myr to $\sim$10~Gyr, the decay of the dust mass and the total disk mass follow {\em different} laws. The reason is that, in all conceivable debris disks, the largest planetesimals have longer collisional lifetimes than the system's age, and therefore did not have enough time to reach collisional equilibrium. If the system were let to evolve for sufficiently long time, both dust mass and disk mass would start to follow $t^{-1}$. However, this requires time spans of much longer than 10~Gyr. \item The loss rate of the dust mass, and the decay rate of fractional luminosity, primarily depend on the difference between the slope $q\sbs{p}$ of the ``primordial'' size distribution of largest planetesimals and the slope $q\sbs{g}$ of the size distribution of somewhat smaller, yet gravity-dominated, planetesimals that already underwent sufficient collisional evolution. With ``standard'' values of $q\sbs{p}$ and $q\sbs{g}$, the dust mass and the thermal fluxes follow approximately $t^\xi$ with $\xi = -0.3\ldots -0.4$. \item Specific decay laws of the total disk mass and the dust mass largely depend on a few model parameters. Most important are: the critical fragmentation energy $Q\sbs{D}^*$ as a function of size, the slope of the ``primordial'' size distribution of planetesimals $q\sbs{p}$ and their maximum size $s\sbs{max}$, and the characteristic eccentricity $e$ and inclination $I$ of planetesimals. \item The property that the maximum possible dust luminosity for a given age does not depend on the initial disk mass, established by \citet{wyatt-et-al-2007a}, is only valid in cases of very rapid collisional evolution, i.e. in closer-in or dynamically very hot disks. For most of the systems at ages $<10$~Gyr, an increase of the initial disk mass leads to an increase of the dust luminosity, unless that initial mass is assigned extreme values, incompatible with our understanding of planetesimal disks. \item Assuming standard material prescriptions and disk masses and extents, a synthetic population of disks generated with our analytic model generally agrees with the observed statistics of 24 and 70~\textmu{m} fluxes versus age. Similarly, the synthetic [24]-[70] colors are consistent with the observed disk colors. \end{enumerate} As every model, our numerical model makes a number of general simplifying assumptions; the analytic one imposes further simplifications: \begin{itemize} \item The collisional evolution is assumed to be smooth and unperturbed. Singular episodes like the aftermath of giant break-ups or special periods of the dynamical evolution such as the late heavy bombardment are not included. \item Effects of possible perturbing planets are taken into account only indirectly: through the eccentricities of planetesimals (dynamical excitation) and confinement of planetesimal belts (truncation of disks). Further effects such as resonant trapping or ejection of material by planets are neglected. \item We only consider disruptive collisions. This is a reasonable approximation for disks that are sufficiently ``hot'' dynamically. However, cratering collisions become important when the relative velocities are insufficient for disruption to occur. \item Neither dilute disks under the regime of Poynting-Robertson drag nor very dense disks with collisional timescales shorter than orbital timescales and with avalanches \citep{grigorieva-et-al-2007} are covered by the present work. \item Explaining the initial conditions or deriving them from the dynamical history of the systems at early stages of planetesimal and planetary accretion was out of the scope of this paper. Correlations between disk masses, disk radii, and the presence of planets, for example, were not considered, although they might alter the scalings we found here. \end{itemize} Despite these limitations, our models reproduce, in essential part, the observed evolution of dust in debris disks. We hope that they may serve as a starting point for in-depth studies that will certainly be undertaken in the future, motivated by questions that remain unanswered, as well as by new data expected from ongoing and planned observational programs. | 7 | 10 | 0710.4294 |
0710 | 0710.2367_arXiv.txt | We present measurements of the neutron-capture elements Rb and Pb for bright giants in the globular clusters M4 and M5. The clusters are of similar metallicity ([Fe/H] $\simeq -1.2)$ but M4 is decidedly $s$-process enriched relative to M5: [Ba/Fe] = +0.6 for M4 but 0.0 for M5. The Rb and Pb abundances were derived by comparing synthetic spectra with high-resolution, high signal-to-noise ratio spectra obtained with MIKE on the Magellan telescope. Abundances of Y, Zr, La, and Eu were also obtained. In M4, the mean abundances from 12 giants are [Rb/Fe] = 0.39 $\pm$ 0.02 ($\sigma$ = 0.07), [Rb/Zr] = 0.17 $\pm$ 0.03 ($\sigma$ = 0.08), and [Pb/Fe] = 0.30 $\pm$ 0.02 ($\sigma$ = 0.07). In M5, the mean abundances from two giants are [Rb/Fe] = 0.00 $\pm$ 0.05 ($\sigma$ = 0.06), [Rb/Zr] = 0.08 $\pm$ 0.08 ($\sigma$ = 0.11), and [Pb/Fe] = $-$0.35 $\pm$ 0.02 ($\sigma$ = 0.04). Within the measurement uncertainties, the abundance ratios [Rb/Fe], [Pb/Fe] and [Rb/X] for X = Y, Zr, La are constant from star-to-star in each cluster and none of these ratios are correlated with O or Na abundances. While M4 has a higher Rb abundance than M5, the ratios [Rb/X] are similar in both clusters indicating that the nature of the $s$-products are very similar for each cluster but the gas from which M4's stars formed had a higher concentration of these products. | \label{sec:intro} Globular clusters continue to provide a source of fascination and frustration to both theorists and observers. Two notable accomplishments include the use of globular clusters to (a) check the age of the Universe (e.g., \citealt{gratton03c}) and to (b) test and refine our understanding of stellar evolution (e.g., \citealt{renzini88}). Despite these successes, globular clusters continue to present bewildering puzzles. The most persistent puzzle relates to chemical composition. For many years, globular clusters have been known to exhibit star-to-star abundance variations for the light elements C, N, O, Na, Mg, and Al (e.g., see reviews by \citealt{smith87}, \citealt{kraft94}, and \citealt{gratton04}). While the amplitude of the star-to-star abundance dispersion can vary from cluster to cluster, the now familiar anticorrelations between C and N, O and Na, and Mg and Al reveal that the abundance variations are likely produced during hydrogen burning at high temperatures via the CNO, Ne-Na, and Mg-Al cycles. (The O-Na and Mg-Al anticorrelations are not seen in field stars.) However, the stars responsible for the nucleosynthesis and the nature of the pollution mechanism(s) remain poorly understood (see \citealt{lattanzio06} for a recent summary). One possible explanation for the observed abundance anomalies is internal mixing and nucleosynthesis (e.g., \citealt{sm79,charbonnel95}) within the present cluster members, the so-called evolutionary scenario. The systematic variation of the C and N \citep{ss91} and Li \citep{grundahl02} abundances with luminosity along the red giant branch demand an evolutionary component to the star-to-star abundance variations. Dredge-up of CN-cycled material accounts for the C and N variations. Development of a giant's convective envelope leading to mixing with highly Li-depleted gas accounts for the decline of the Li abundance with increasing luminosity. The proton-capture reactions causing the O, Na, Mg, and Al variations demand much higher temperatures and much deeper mixing than those required for CN-cycling. Such mixing is not predicted by standard theoretical models of red giants and the discovery of the O, Na, Mg, and Al anomalies in main sequence stars (e.g., \citealt{briley96}, \citealt{gratton01}) eliminates deep mixing as a viable explanation for the O-Al variations. The interiors of main sequence stars are too cool to process Ne to Na or Mg to Al. Therefore, the cluster gas must have been inhomogeneous when the present stars were formed. This alternative explanation for the abundance anomalies is the so-called primordial scenario. In the primordial scenario, intermediate-mass asymptotic giant branch stars (IM-AGBs) from the generation to which the observed stars belong have long been considered candidates for synthesizing the abundance variations \citep{cottrell81}. In IM-AGBs, the convective envelope can reach the top of the hydrogen-burning shell, a process called hot-bottom burning. For sufficiently massive and metal-poor AGBs, the temperatures at the base of the convective envelope can exceed 100 million degrees thereby allowing the efficient operation of the CNO, Ne-Na, and Mg-Al cycles (e.g., \citealt{karakas03}). That IM-AGBs do not alter the abundances of the alpha or iron-peak elements (as required by observations) adds to their qualitative appeal. However, quantitative tests reveal problems with the IM-AGB primordial scenario. Theoretical yields from IM-AGBs combined with a chemical evolution model \citep{fenner04} suggest that O is not sufficiently depleted, Na is overproduced, Mg is produced rather than destroyed, the isotope ratios of Mg do not match the observations, and the sum of C+N+O increases substantially in contrast to the observations. \citet{ventura05a,ventura05b,ventura05c} find that many of the flaws noted above can be alleviated when IM-AGB yields are calculated using a revised treatment for convection and mass-loss. However, \citeauthor{ventura05a} note that problems persist, namely with the Mg isotope ratios, and warn that the predictive power of the current AGB models is limited. Recently, \citet{prantzos06}, \citet{smith06}, and \citet{decressin06} suggest that the winds from massive stars may be more promising candidates than IM-AGBs. There is no satisfactory explanation for the complex patterns for the light element abundances exhibited by every well studied Galactic globular cluster. Therefore, our present understanding of globular cluster chemical evolution and/or stellar nucleosynthesis is incomplete. Determinations of the stellar abundances of the trans-iron or heavy elements offer clues to the history behind the chemical evolution of globular clusters. Here, we provide novel information -- the Rb and Pb abundances -- for giants in M4 and M5, a pair of clusters of similar metallicity but with distinctly different levels of $s$-process products. The quintessential $r$-process element Eu has similar abundances in the two clusters and, indeed, across the collection of Galactic globular clusters. In sharp contrast, the $s$-process products are more evident in M4 than in M5 and other clusters of similar metallicity: [Ba/Fe] is about +0.6 in M4 but 0.0 in M5. The questions - Are there differences in the Rb and Pb abundances between this pair of clusters? and Are the star-to-star variations in the abundances of light elements (O, Na, Mg, and Al) reflected in variations among the abundances of Rb and Pb? -- seem likely to probe the origins of the $s$- and $r$-process products for globular clusters. Due to a critical branching point in the $s$-process path at $^{85}$Kr, the abundance of Rb relative to Sr, Y, or Zr can differ by a factor of 10 depending upon the neutron density at the $s$-process site. In the case of AGB stars, the neutron density in the He-shell is dependent on the stellar mass (e.g., see \citealt{tomkin83}, \citealt{lambert95}, \citealt{busso99}, and \citealt{abia01} for further details). Since the isotopes of Pb and Bi are the last stable nuclei on the $s$-process path, the $s$-process terminates at these elements and overabundances of Pb and Bi will arise if seed nuclei are shuffled by neutron captures down the entire $s$-process path. In particular, metal-poor AGB stars may produce large overabundances of Pb and Bi if the neutron supply per seed exceeds a critical value (e.g., see \citealt{goriely01}, \citealt{travaglio01}, and \citealt{busso01} for further details). The suspicion is that the star-to-star abundance variations for light elements are due to contamination by IM-AGBs. Some contend that IM-AGBs also synthesize $s$-process nuclides and then one might expect to see star-to-star variations in the Rb and Pb abundances as well as correlations with light element abundances. To further examine the possible role of IM-AGBs in the chemical evolution of globular clusters, \citet{rbpbsubaru} measured Rb and Pb in NGC 6752 and M13, the two clusters that exhibit the largest amplitude for Al variations. It was found that the abundance ratios [Rb/Zr] and [Pb/Fe] were constant from star-to-star within the measurement uncertainties. If IM-AGBs do synthesize Rb and Pb, then they may not be responsible for the abundance variations. On the other hand, if IM-AGBs are responsible for the abundance variations, they cannot synthesize Rb or Pb. In this paper, we extend the measurements of Rb and Pb to the globular clusters M4 and M5. While these clusters are more metal-rich than NGC 6752 or M13, both M4 and M5 are known to exhibit large dispersions and correlations for the light element abundances [see pioneering studies on CN bimodality by \citet{norris81a} and \citet{smith83} as well as recent high-resolution spectroscopic studies by \citet{M4,M5}, \citet{ramirez03}, and references therein]. In particular, as noted above, M4 is remarkably, perhaps uniquely among globular clusters, enriched in $s$-process products. | \label{sec:summary} In this paper we present measurements of the neutron-capture elements Rb and Pb in the globular clusters M4 and M5. While both clusters exhibit star-to-star abundance variations for the light elements, we find that the abundances of Rb and Pb are constant. None of the abundance ratios [Rb/Fe], [Rb/Zr], and [Pb/Fe] are correlated with O or Na abundances. In the primordial scenario, the abundance variations for the light elements are attributed to different levels of accretion of ejecta from IM-AGBs or massive stars. The fact that the heavy elements including Rb and Pb do not show abundance variations implies that the accreted material has the same composition as the ambient material for the heavy elements (i.e., the accreted material cannot be highly underabundant or overabundant in these elements). That the ratios [Rb/X] for X = Y, Zr, La are similar for M4 and M5 suggests that the source of the $s$-process elements are similar and that M4 had a greater concentration of these products. There remains a need to pursue additional observational tests of the primordial scenario. In particular, present data on the Rb and Pb abundances in field and cluster stars are sparse. The indication that the Mg isotopic ratios of unpolluted or normal cluster stars differ from those of field stars of the same metallicity deserves closer scrutiny by, in particular, extending the measurement of these isotopic ratios to additional clusters. | 7 | 10 | 0710.2367 |
0710 | 0710.5637_arXiv.txt | { A new method for the determination of open cluster membership based on a cumulative effect is proposed. In the field of a plate the relative $x$ and $y$ coordinate positions of each star with respect to all the other stars are added. The procedure is carried out for two epochs $t_1$ and $t_2$ separately, then one sum is subtracted from another. For a field star the differences in its relative coordinate positions of two epochs will be accumulated. For a cluster star, on the contrary, the changes in relative positions of cluster members at $t_1$ and $t_2$ will be very small. On the histogram of sums the cluster stars will gather to the left of the diagram, while the field stars will form a tail to the right. The procedure allows us to efficiently discriminate one group from another. The greater the distance between $t_1$ and $t_2$ and the more cluster stars present, the greater is the effect. The accumulation method does not require reference stars, determination of centroids and modelling the distribution of field stars, necessary in traditional methods. By the proposed method 240 open clusters have been processed, including stars up to $m<13$. The membership probabilities have been calculated and compared to those obtained by the most commonly used Vasilevskis-Sanders method. The similarity of the results acquired the two different approaches is satisfactory for the majority of clusters. | Various methods based on the analysis of positions, proper motions, radial velocities, magnitudes and their combinations have been proposed to determine the members of open clusters. The first mathematically rigorous procedure for determination of open cluster membership was developed by Sanders (\cite{Sanders}) with a statistical analysis of proper motions. It is also the most widely used method. Sanders's approach is based on the model of overlapping distributions of field and cluster stars in the neighborhood and within the region of visible grouping of stars, introduced by Vasilevskis et al. (\cite{Vasilevskis}). Vasilevskis's model implies that proper motion dispersion of cluster members is caused by observational and measurement errors assumed to be normally distributed. Thus the distribution of cluster stars is represented by a bivariate normal frequency function. The dispersion of field stars is due to not only the errors referred to above, but also to peculiar motion and differential galactic rotation. Therefore the field star distribution is not expected to be random, but rather to have a preferential direction and not normal distribution. However, in a first approximation a bivariate normal ellipsoidal distribution function was assumed for field stars, the major axis of the ellipse being parallel to the galactic plane. Thus, Sanders's equations contain 8 unknown parameters to be determined: the number of cluster members, x and y components of two centroids, one circular and two elliptical dispersions. This system is solved by a maximum likelihood method (Sanders \cite{Sanders}). The probability of stars being cluster members is calculated by frequency functions with determined parameters. Slovak (\cite{Slovak}) tested the Vasilevskis-Sanders method by modelling the proper motion distribution in the surroundings of a cluster. He proved the uniqueness and convergence of solutions of the system, provided that errors are represented by a Gaussian distribution and open clusters have no significant internal motion. But if there is noticeable motion within a cluster or it rotates then the above method fails. Cabrera-Ca\~{n}o and Alfaro (\cite{Cabrera-Cano Alfaro1}) improved the numerical techniques for obtaining the above parameters. McNamara and Schneeberger (\cite{McNamara Schneeberger}) showed that the final probabilities could be influenced by various weight groups. Zhao and He (\cite{Zhao He}) provided a method for treating data with different accuracies. However all these improvements did not treat the problem of star distribution model on which the membership probabilities are based. Even if the hypothesis of a normal distribution of field stars is realistic for some clusters, the centroids of field and cluster stars sometimes are too close to be well discriminated. The parametric model also fails when the cluster member-to-field star ratio is small. The Vasilevskis-Sanders method does not work in the case of significant internal motion in a cluster or its rotation. To overcome some of the problems arising from the parametric Vasilevskis-Sanders method, especially the star distribution modelling, Cabrera-Ca\~{n}o and Alfaro (\cite{Cabrera-Cano Alfaro2}) developed a more general, non-parametric method of membership determination. Here no assumptions were made about the nature of cluster and field star distributions and allowed for the use of photometric data. Each cluster needed careful individual study. So, the actual distribution of cluster and field stars, especially the latter, may not be fitted by a Vaselevskis-Sanders model; the centroids for the two groups may be too close to be distinguished; the reliability of the method depends on the cluster-to-field star ratio; in the case of significant internal motions or rotation of a cluster the traditional method fails. In the next section we introduce a new method for the discrimination of cluster stars from surrounding field stars based on a cumulative effect using positions and proper motions. We do not try to overcome all the difficulties of traditional methods, though we offer another approach that enlarges the statistical distance between the two populations - cluster members and field stars - by revealing the group of stars with the least relative velocities. The advantage of this method is that no assumption is made about the distribution of field stars and determination of centroids is avoided. However, in order to determine probabilities we still have to assume a normal bivariate distribution for clusters. Another advantage of the method is that reference stars are not necessary: the discrimination of cluster members is most effective by rectangular coordinates. These features of the method allow us to increase the statistical distance between the two populations. The most noticeable advantage of the accumulation method is its ability to reveal dynamic structures within the clusters if there are any. | In the presented method of accumulation we assume that the cluster members moving in the Galaxy as a whole have similar velocities, while field stars show a wide range of velocities. The procedure - adding close velocities while compensating for disperced ones - should enhance the assembling of cluster stars, while distributing more sparsely the field ones, due to the cumulative effect. Thus this method effectively enlarges the statistical distance between physical members and field stars, so that a membership as well as a non-membership is more pronounced. This effect is better for larger cluster-to-field star ratios. For field stars no particular distribution is assumed, thus the centroids are not determined. However, when calculating probabilities, for physical members a normal bivariate distribution function is assumed. For rectangular coordinates no reference stars are needed in the procedure. The most interesting feature of the accumulation mathod is its ability to reveal more then one group of velocities, which has been shown with the example of Pleades. The modified accumulation method was applied to 240 clusters from Dias's list. The probabilities calculated by the accumulation method showed satisfactory agreement with those obtained by the Vasilevskis-Sanders method for the majority of clusters. The poor agreements or disagreements can be ascribed to low cluster-to-field star ratios, or multiple dynamic structures. The accumulation method can be further expanded using other variables like radial velocities and photometric quantities. | 7 | 10 | 0710.5637 |
0710 | 0710.0224_arXiv.txt | { We discuss the large scale properties of standard cold dark matter cosmological models characterizing the main features of the power-spectrum, of the two-point correlation function and of the mass variance. Both the real-space statistics have a very well defined behavior on large enough scales, where their amplitudes become smaller than unity. The correlation function, in the range $0<\xi(r)<1$, is characterized by a typical length-scale $r_c$, at which $\xi(r_c)=0$, which is fixed by the physics of the early universe: beyond this scale it becomes negative, going to zero with a tail proportional to $-(r^{-4})$. These anti-correlations represent thus an important observational challenge to verify models in real space. The same length scale $r_c$ characterizes the behavior of the mass variance which decays, for $r>r_c$, as $r^{-4}$, the fastest decay for any mass distribution. The length-scale $r_c$ defines the maximum extension of (positively correlated) structures in these models. These are the features expected for the dark matter field: galaxies, which represent a biased field, however may have differences with respect to these behaviors, which we analyze. We then discuss the detectability of these real space features by considering several estimators of the two-point correlation function. By making tests on numerical simulations we emphasize the important role of finite size effects which should always be controlled for careful measurements. | In contemporary cosmological models the structures observed today at large scales in the distribution of galaxies in the universe are explained by the dynamical evolution of purely self-gravitating matter (dark matter) from an initial state with low amplitude density fluctuations, the latter strongly constrained by satellite observations of the fluctuations in the temperature of the cosmic microwave background radiation. The other main observational elements for the understanding of the large scale structure of the universe is represented by the studies of galaxy correlations. Any theoretical model aiming to explain the formation of structures must be tested against the data provided by galaxy surveys which give the important bridge between the regimes characterized by large and small fluctuations. Models of the early universe (see e.g. Padmanabhan, 1993 and references therein) predict certain primordial fluctuations in the matter density field, defining the correlations of the initial conditions, i.e. at the time of decoupling between matter and radiation. In the regime where density fluctuations are small enough, the correlation function of the present matter density field is simply related to one describing the initial conditions. In fact, according to the growth of gravitational instabilities in an expanding universe in the linear regime perturbations are simply amplified (see e.g., Peebles, 1980 and references therein). Thus today at some large scales where the correlation function is still positive but with $\xi(r)<1$ the imprint of primordial fluctuations should be preserved. In the region of strong non-linear fluctuations an analytical treatment to predict the behavior of the two-point correlation function has not been developed yet and, in general, one makes use of numerical simulations which provide a rich, but phenomenological, description of structure in the non-linear regime. It is in this regime, at small enough scales, where most observations have been performed until now. We focus here on the type of correlations predicted in the linear regime by models of the early universe. While the characterization of correlations is usually done in terms of the power-spectrum of the density fluctuations a real space analysis turns out to be useful to point out some relevant features from an observational point of view (see, e.g., the discussion in Gabrielli et al., 2004). Theoretical models of primordial matter density fields in the expanding universe are characterized by a single well-defined length scale, which is an imprint of the physics of the early universe at the time of the decoupling between matter and radiation (see e.g. Bond and Efstathiou 1984, and Padmanabhan 1993 for a general introduction to the problem). The redshift characterizing the decoupling is directly related to the scale at which the change of slope of the power-spectrum of matter density fluctuations $P(k)$ occurs, i.e. it defines the wavenumber $k_c$ at which there is the turnover of the power-spectrum between a regime, at large enough $k$, where it behaves as a negative power-law of the wave number $P(k) \sim k^{m}$ with $-1<m\le-3$, and a regime at small $k$ where $P(k)\sim k$ as predicted by inflationary theories. Given the generality of this prediction, it is clearly extremely important to look for this scale in the data. The exact location of this scale is related to several parameters, including the cosmological ones which describe the geometry of the universe at large scales (see e.g. Padmanabhan 1993, Tegmark et al. 2004 and Spergel et al. 2007 for a recent determination). We discuss in what follows that the scale $r_c$ corresponding to the wave-number $k_c$, in a particular variant of Cold Dark Matter (CDM) models --- the so-called $\Lambda$CDM vanilla model --- is predicted to be $r_c \approx 124$ Mpc/h\footnote{For seek of clarity we have chosen the scale of distances normalized to the adimensional Hubble parameter $h$, which is defined from the Hubble's constant $H_0 = 100 h$ km/sec/Mpc.}. At this scale the real space correlation function crosses zero, becoming negative at larger scales. In particular the correlation function presents a positive power-law behavior at scales $r \ll r_c$ and a negative power-law behavior at scales $r \gg r_c$. Positive and negative correlations are exactly balanced in way such that the integral over the whole space of the correlation function is equal to zero. This is a global condition on the system fluctuations which corresponds to the fact that the distribution is super-homogeneous (or hyper-uniform), i.e. characterized by a sort of stochastic order and by fluctuations which are depressed with respect to, for example, a purely uncorrelated distribution of matter (Gabrielli, Joyce and Sylos Labini, 2002 --- see discussion below). Note that the scale $r_c$ marks the maximum extension of positively correlated structures: beyond $r_c$ the distribution must be anti-correlated since the beginning, as the evolution time was not sufficient for the positive correlations to be developed. Thus this scale can be regarded as an upper limit to the maximum size of structures (with large of weak correlations) in the present universe. The possible discoveries of structures of larger size is still a challenging task for observational cosmology. A relevant problem for the measurements of small amplitude values of the correlation function, i.e. when $\xi(r) <1$, is represented by the characterization and the understanding of both the systematic biases which may affect the estimators of $\xi(r)$ and the stochastic noise which perturbs any real determination. A study of this problems can be found, for example, in Kerscher (1999) and Kerscher et al. (2000) where it is shown that in general the biases in several estimators of the two-point correlation function are not negligible. In particular when there are structures of large spatial extension inside a given sample there can be non negligible biases affecting the determination of two-point properties. We focus here on the systematic bias related to the effect of the so-called integral constraint, which distorts any estimator of the correlation function at large scales in any given sample. The integral-constraint represents an overall condition on any estimator of the correlation function which is due to the fact that the average density, estimated in any given sample, is in general different from its ensemble average value. Here we treat explicitly the case for the simplest estimator of the two-point correlation function, {the so-called full-shell or minus estimator and and we illustrate the situation for the other estimatorsby studying artificial distributions. In particular we devote most attention to the estimator introduced by Davis and Peebles (1983), which is still very used in the literature, and to the estimator introduced by Landy and Szalay (1993), which is the most popular one. Kerscher et al. (2000) considered also other estimators, like the Hewett estimator (Hewett, 1982) and the Hamilton estimator (Hamilton, 1993) and have shown that the results obtained with he Landy and Szalay estimator are almost indistinguishable from the Hamilton estimator. In this way we will be able to identify the problems related to the identification of correlations above the mentioned scale $r_c$: we will then propose several tests to be applied to the galaxy data, in order to define the strategy to study the correlation function at small amplitudes and larger distances in order to eventually detect the length scale $r_c$. Up to now studies of the correlation function $\xi(r)$ in galaxy samples have been limited to small scales, i.e. $0.1 < r \ltapprox 30$ Mpc/h (i.e. Totsuji \& Kihara, 1969, Davis and Peebles, 1983, Davis et al., 1988, Benoist et al., 1996 Park et al., 1994, Scranton et al., 2002}, Zehavi et al., 2002, Zehavi et al., 2004, Ross et al., 2007) and only recently the volume covered by galaxy redshift samples is approaching a size which is large enough to make a robust estimation of the correlation function at scales of order 100 Mpc/h. When the Sloan Digital Sky Survey (SDSS) (York et al., 2000) will be completed by filling up the gap between the two main angular regions of observations, which are nowadays disjointed, the volume of the survey and the statistics of the number of objects in the samples would be large enough to test space correlations on scales of order $r_c$ or more. An exception to this situation is represented by the paper by Eisenstein, et al., (2005), who, by studying a sample of Luminous Red Galaxies (LRG) of the SDSS, have estimated the correlation function on scales of order 100 Mpc/h. These authors have however focused their attention to another real space feature of theoretical models: the so-called ``bump'' of the correlation function which corresponds in real space to the so-called Doppler peaks in the matter power-spectrum generated by the baryonic acoustic oscillations in the early universe. As we discuss below this bump, corresponding to a singular point of the correlation function (Gabrielli et al., 2004), is localized at scales of order of 100 Mpc/h and characterized by a small amplitude. This is a second important real-space scale of the theoretical correlation function which is localized at a scale slightly smaller than $r_c$. The detection of the baryonic bump is thus related to the detection of the scale $r_c$ as any finite-size effect perturbing the determination of the scale $r_c$ will, inevitably, also affect the determination of the baryonic bump. In fact the baryonic bump can be seen as a small modification to the overall shape of the correlation function at scales of order $r_c$, to which we focus our attention here. Note that, because of the very large scales, the acoustic signature and the zero point scale remain in the linear regime even today and they are weakly affected by non-linear effects (see Eisenstein et al., 2006). Thus real space and redshift space properties, at such large scales, should not differ substantially. In Section 2 we introduce the basic definitions of the statistical quantities usually employed to characterize two-point properties in real and Fourier space. In Section 3 we discuss a simple functional behavior of the power-spectrum of matter density fluctuations which captures the main elements of a more realistic CDM power-spectrum. We discuss the real-space properties as represented by the two-point correlation function and we consider the problem of selection or biasing in the simplest theoretical scheme of biasing a correlated Gaussian field. In Section 4 we treat explicitly the case of a $\Lambda$CDM matter density field characterizing in detail real space properties. The main estimators of the two-point correlation function are discussed in Section 5 and in Section 6 we test these estimators in artificial distributions. Finally in Section 7 we draw our main conclusions discussing the problems related to the estimations of two-point correlations in real galaxy samples. | We have considered the real-space properties of CDM density fields, focusing in detail in a particular variant known as $\Lambda$CDM (vanilla) model. It is well known that the power-spectrum has typically a behavior $P(k) \sim k^{m}$ with $-1<m\le-3$ for large wavelengths $k > k_c$, and $P(k)\sim k$ at smaller wavelengths $k <k_c$. We discussed that, correspondingly, the two-point correlation function shows approximatively a positive power-law behavior $\xi(r) \sim r^{-2}$ at small scales $r < r_c \approx k_c^{-1}$ and a negative power-law behavior $\xi(r) \propto - r^{-4}$ at large scales $r>r_c$, where the zero-crossing occurs at about $r_c \approx 124$ Mpc/h in the model considered. We discussed the fact that, globally, a system with this type of correlations belong to the category of super-homogeneous distributions, which are configurations of points more ordered than a purely uncorrelated (Poisson) distribution. Correspondingly fluctuations are depressed with respect to the Poisson case, and the normalized mass variance, for instance, decay faster ($\sigma^2(r) \sim r^{-4}$) than for the Poisson case ($\sigma^2(r) \sim r^{-3}$). The condition of super-homogeneity is expressed by the condition that $P(k) \rightarrow 0$ for $k \rightarrow 0$, or alternatively that \[ \int_0^{\infty} \xi(r) r^2 dr = 0 \;. \] Following the work of Durrer et al. (2003) we have pointed out that the above condition is broken when one samples the distribution, as for example when the simplest biasing scheme of correlated Gaussian fields (introduced by Kaiser, 1984) is applied. This is particularly important for the behavior of the power-spectrum for $k<k_c$, which, under biasing, remains constant instead of going as $P(k) \sim k$. The correlation function at large scales $r>r_c$ is instead expected to be linearly amplified with respect to the original one of the whole matter field. Thus the large scale negative tail $\xi(r) \sim -r ^{-4}$ is the main feature which one would like to detect in order to test theoretical models. Given the fact that when $\xi(r)$ becomes negative, it is characterized by a very small amplitude, the determination of the negative power-law tail represents a very challenging problem. We have discussed the fact that, at first approximation in a real measurement, one may treat the system as having positive correlations at small scales with an exponential cut-off at the scale $r_c$ and then it becomes uncorrelated (a situation which can be regarded as upper limit to the presence of anti-correlations). This implies that for $r_c > 124$ Mpc/h galaxy distribution should not present any positive correlation. Whether this behavior is compatible with the existences of structures of order 200 Mpc/h or more is an open problem which has to be addressed in the studies of forthcoming galaxy catalogs. More in detail, one of the most basic results (see e.g., Peebles 1980) about self-gravitating systems, treated using perturbative approaches to the problem (i.e. the fluid limit), is that the amplitude of small fluctuations grows monotonically in time, in a way which is independent of the scale. This linearized treatment breaks down at any given scale when the relative fluctuation at the same scale becomes of order unity, signaling the onset of the ``non-linear'' phase of gravitational collapse of the mass in regions of the corresponding size. If the initial velocity dispersion of particles is small, non-linear structures start to develop at small scales first and then the evolution becomes ``hierarchical'', i.e., structures build up at successively larger scales. Given the finite time from the initial conditions to the present day, the development of non-linear structures is limited in space, i.e., they can not be more extended than the scale at which the linear approach predicts that the density contrast becomes of order unity at the present time. This scale is fixed by the initial amplitude of fluctuations, constrained by the cosmic microwave background anisotropies (Spergel et al., 2007), by the hypothesized nature of the dominating dark matter component and its correlation properties. According to current models of CDM-type the scales at which non-linear clustering occurs at the present time (of order 10 Mpc) are much smaller than the scale $r_c \approx 124$ Mpc/h (see e.g. Springel et al., 2005). Thus the region where the super-homogeneous features should still be in the linear regime, allowing a direct test of the initial conditions predicted by early universe models. The scale $r_c$ marks the maximum extension of positively correlated structures: beyond $r_c$ the distribution must be anti-correlated since the beginning, as there was no time to develop other correlations. The possible presence of structures, which mark long-range correlations, whether or not of large amplitude, reported both by observations of galaxy distributions (like the Sloan Great Wall --- see Gott et al., 2005), by the detection of dark matter distributions (see e.g. Massey et al., 2007) and by the large void of radius $\sim 140$ Mpc identified by Rudnick et al. (2007), is maybe indicating that positive correlations extend well beyond $r_c$. We have discussed that an important finite size effect must be considered when estimating the correlation function, and which may mimic a break of the power-law behavior similar to the ones of CDM models at a scale of order $r_c$. This is related to the effect of the integral constraint in the estimators, namely the fact that the sample average, estimated in a finite sample, differs from the ensemble average, and can be finite-size dependent. This situation occurs when correlations (weak or strong) extend to scales larger than the sample size. For these reasons, in order to study the two-point correlation function in real galaxy samples when its amplitude becomes smaller than unity, it is crucial to check whether the break of the power-law behavior has a finite size dependence or not, by choosing samples with different depth. In this perspective the assessment of the reality of the break of the two-point correlation function is the main observational point to be considered. Once this will be clarified other features should be considered, as for the example the so-called baryonic bump, which is a very small perturbation to the overall shape of the correlation function at scales of order of the zero-point $r_c$. We will present a detailed analysis of the correlation properties of galaxy distribution in the SDSS catalog, considering specific tests for finite-size effects in the determination of the correlation function, in a forthcoming paper. | 7 | 10 | 0710.0224 |
0710 | 0710.2017_arXiv.txt | { H.E.S.S. observations of the old-age ($>$10$^4$~yr; $\sim 0.5^\circ$ diameter) composite supernova remnant (SNR) W~28 reveal very high energy (VHE) $\gamma$-ray emission situated at its northeastern and southern boundaries. The northeastern VHE source (HESS~J1801$-$233) is in an area where W~28 is interacting with a dense molecular cloud, containing OH masers, local radio and X-ray peaks. The southern VHE sources (HESS~J1800$-$240 with components labelled A, B and C) are found in a region occupied by several HII regions, including the ultracompact HII region W~28A2. Our analysis of NANTEN CO data reveals a dense molecular cloud enveloping this southern region, and our reanalysis of EGRET data reveals MeV/GeV emission centred on HESS~J1801$-$233 and the northeastern interaction region. } \email{[email protected]} \begin{document} | The study of shell-type SNRs at $\gamma$-ray energies is motivated by the idea that they are the dominant sites of hadronic Galactic cosmic-ray (CR) acceleration to energies approaching the \emph{knee} ($\sim 10^{15}$~eV) and beyond, e.g. \cite{Ginzburg:1}. CRs are then accelerated via the diffusive shock acceleration (DSA) process (eg. \cite{Bell:1,Blandford:2}). Gamma-ray production from the interaction of these CRs with ambient matter and/or electromagnetic fields is a tracer of such particle acceleration, and establishing the hadronic or electronic nature of the parent CRs in any $\gamma$-ray source is a key issue. Already, two shell-type SNRs, RX~J1713.7$-$3946 and RX~J0852.0$-$4622, exhibit shell-like morphology in VHE $\gamma$-rays \cite{HESS_RXJ1713_II,HESS_VelaJnr_II,HESS_RXJ1713_III} to 20~TeV and above. Although a hadronic origin of the VHE $\gamma$-ray emission is highly likely in the above cases, an electronic origin is not ruled out. W~28 (G6.4$-$0.1) is a composite or mixed-morphology SNR, with dimensions 50$^\prime$x45$^\prime$ and an estimated distance between 1.8 and 3.3~kpc (eg. \cite{Goudis:1,Lozinskaya:1}). It is an old-age SNR (age 3.5$\times 10^4$ to 15$\times 10^4$~yr \cite{Kaspi:1}), thought to have entered its radiative phase of evolution \cite{Lozinskaya:1}. The shell-like radio emission \cite{Long:1,Dubner:1} peaks at the northern and northeastern boundaries where interaction with a molecular cloud \cite{Wootten:1} is established \cite{Reach:1,Arikawa:1}. The X-ray emission, which overall is well-explained by a thermal model, peaks in the SNR centre but has local enhancements in the northeastern SNR/molecular cloud interaction region \cite{Rho:2}. Additional SNRs in the vicinity of W~28 have also been identified: G6.67$-$0.42 and G7.06$-$0.12 \cite{Yusef:1}. The pulsar PSR~J1801$-$23 with spin-down luminosity $\dot{E} \sim 6.2\times 10^{34}$ erg~s$^{-1}$ and distance $d\geq9.4$~kpc \cite{Claussen:3}, is at the northern radio edge. Given its interaction with a molecular cloud, W~28 is an ideal target for VHE observations. This interaction is traced by the high concentration of 1720~MHz OH masers \cite{Frail:2,Claussen:1,Claussen:2}, and also the location of very high-density ($n>10^3$~cm$^{-3}$) shocked gas \cite{Arikawa:1,Reach:1}. Previous observations of the W~28 region at VHE energies by the CANGAROO-I telescope revealed no evidence for such emission \cite{Rowell:1} from this and nearby regions. The High Energy Stereoscopic System (H.E.S.S.: see \cite{Hinton:1} for details and performance) has observed the W~28 region over the 2004, 2005 and 2006 seasons. After quality selection, a total of $\sim$42~hr observations were available for analysis. Data were analysed using the moment-based Hillas analysis procedure employing {\em hard cuts} (image size $>$200~p.e.), the same used in the analysis of the inner Galactic Plane Scan datasets \cite{HESS_GalScan,HESS_GalScan_II}. An energy threshold of $\sim 320$~GeV results from this analysis. The VHE $\gamma$-ray image in Fig.~\ref{fig:tevskymap} shows that two source of VHE $\gamma$-ray emission are located at the northeastern and southern boundaries of W~28. The VHE sources are labelled HESS~J1801$-$233 and HESS~J1801$-$240 where the latter can be further subdivided into three components A, B, and C. The excess significances of both sources exceed $\sim$8$\sigma$ after integrating events within their fitted, arcminute-scale sizes. Similar results were also obtained using an alternative analysis \cite{Mathieu:1}. \begin{figure*}[th] \centering \hbox{ \begin{minipage}{0.55\textwidth} \includegraphics[width=\textwidth]{icrc0129_fig1.eps} \end{minipage} \begin{minipage}{0.45\textwidth} \caption{H.E.S.S. VHE $\gamma$-ray excess counts, corrected for exposure and Gaussian smoothed (with 4.2$^\prime$ std. dev.). Solid green contours represent excess significance levels of 4, 5, and 6$\sigma$, for an integrating radius $\theta$=0.1$^\circ$. The VHE sources HESS~J1801$-$233 and a complex of sources HESS~J1800-240 (A, B \& C) are indicated. The thin-dashed circle depicts the approximate radio boundary of the SNR W~28 \cite{Dubner:1,Brogan:1}. Additional objects indicated are: HII regions (black stars); W~28A2, {G6.1$-$0.6} % {6.225$-$0.569}; % The 68\% and 95\% location contours (thick-dashed yellow lines) of the $E>100$~MeV EGRET source {GRO~J1801$-$2320}; the pulsar {PSR~J1801$-$23} (white triangle). The inset depicts a pointlike source under identical analysis and smoothing as for the main image.} \label{fig:tevskymap} \end{minipage} } \end{figure*} | H.E.S.S. and NANTEN observations reveal VHE emission in the W~28 region spatially coincident with molecular clouds. The VHE emission and molecular clouds are found at the northeastern boundary, and $\sim 0.5^\circ$ south of W~28 respectively. The SNR W~28 may be a source of power for the VHE sources, although there are additional potential particle accelerators in the region such as other SNR/SNR-candidates, HII regions and open clusters. Further details concerning these results and discussion are presented in \cite{HESS_W28}. | 7 | 10 | 0710.2017 |
0710 | 0710.5727_arXiv.txt | {}{We communicate the detection of soft (20--200\,keV) $\gamma$-rays from the pulsar and pulsar wind nebula of \hbox{PSR~J1846$-$0258} and aim to identify the component of the system which is responsible for the $\gamma$-ray emission.}{We combine spectral information from the \integral $\gamma$-ray mission with archival data from the \chandra X-ray Observatory to pinpoint the source of soft $\gamma$-ray emission.}{Our analysis shows that the soft $\gamma$-rays detected from \hbox{PSR~J1846$-$0258} include emission from both the pulsar and the pulsar wind nebula, but the measured spectral shape is dominated by the pulsar wind nebula. We discuss \hbox{PSR~J1846$-$0258} in the context of rotation powered pulsars with high magnetic field strengths and review the anomalously high fraction of spin-down luminosity converted into X- and $\gamma$-ray emission in light of a possible overestimate of the distance to this pulsar.}{} | The pulsar \hbox{PSR~J1846$-$0258} (also known as \hbox{AX~J1846.4$-$0258}) was discovered by \citet{GotthelfVasishtBoylan-Kolchin2000} in the X-ray band and lies near the centre of the supernova remnant Kes 75 (SNR G29.7-0.3). Recent X-ray imaging \citep{HelfandCollinsGotthelf2003} shows that the pulsar is embedded in a pulsar wind nebula (PWN) which shows distinct physical structure. No radio emission has been observed from \hbox{PSR~J1846$-$0258}, but X-ray timing reveals the pulse period to be $P=324$\,ms and the characteristic age is $P/(2\dot{P})=728$\,years -- the smallest characteristic age of any rotation-powered pulsar. The inferred surface dipole magnetic field strength is $4.8\times10^{13}$\,G -- almost an order of magnitude greater than the magnetic fields for typical pulsars, placing it closer to the field strengths of magnetars. The distance to the system is $\sim19$\,kpc as estimated from the neutral hydrogen density along the line of sight \citep{BeckerHelfand1984}, which implies that the size of the supernova remnant (SNR) is extremely large for such a young pulsar. It also implies that the efficiency of the conversion of the pulsar spin-down luminosity ($8.4\times10^{36}$\,erg\,s$^{-1}$, using $P$ and $\dot P$ from \citealt{GotthelfVasishtBoylan-Kolchin2000}) to combined pulsar and PWN X-ray luminosity in the 0.5--10\,keV energy range, is 20\% (using the flux from this work) -- the largest of any rotation-powered pulsar. This source is one of a growing class of rotation-powered pulsars with B-field strengths approaching those of magnetars. In this paper we present the \integral observations and results for \hbox{PSR~J1846$-$0258}, link these results to previously published \chandra results and follow up with a discussion of the properties of this source in comparison to other members of this class and examine the possibility of an overestimate in the distance to this pulsar. \begin{figure*} \sidecaption \includegraphics[width=12cm]{8432fig1.eps} \caption{IBIS/ISGRI significance mosaic of the 20--100\,keV energy band, showing \hbox{PSR~J1846$-$0258} in the centre of the field. False colour representation of the significance is displayed on a logarithmic scale. The top right inset is a $50\arcsec\times50\arcsec$ image from our reduction of \chandra data in the 0.3--10\,keV energy band. One can clearly distinguish the bright pulsar and the surrounding synchrotron nebulosity. The white ellipses on the \chandra image indicate the extraction regions for the PWN and pulsar spectra as explained in Sect.~\ref{chanresults}.} \label{FigIntimg} \end{figure*} | \label{discuss} PSR~J1846$-$0258 is one of a recently emerged and growing class of rotation-powered pulsars with inferred surface dipole magnetic fields approaching those of the magnetars. A number of these pulsars have been discovered as a result of the Parkes multi-beam pulsar survey of the Galactic plane \citep{ManchesterLyneCamilo2001}, for example \hbox{PSR~J1119$-$6127} and \hbox{PSR~J1814$-$1744} \citep{CamiloKaspiLyne2000}, which have inferred magnetic fields of $4.1\times10^{13}$\,G and $5.5\times10^{13}$\,G respectively. These are all young pulsars with characteristic ages of order a few thousand years and approximately half of them show X-ray emission (\hbox{PSR~J1119$-$6127}, \citealt{GonzalezSafi-Harb2003}; \hbox{PSR~J1718$-$3718}, \citealt{KaspiMcLaughlin2005}) with spectra harder than those of magnetars. In fact, \hbox{PSR~J1119$-$6127} shows a remarkable resemblance to \hbox{PSR~J1846$-$0258} as far as spin characteristics and the properties inferred from them are concerned. \hbox{PSR~J1119$-$6127} has $P=408$\,ms and $\dot{P}=4.1\times10^{-12}$\,s\,s$^{-1}$, giving a spin-down luminosity, characteristic age and inferred surface dipole magnetic field very close to those of \hbox{PSR~J1846$-$0258} \citep{CamiloKaspiLyne2000}. This, however, is where the similarity ends. Across the electromagnetic spectrum, these pulsars have very different attributes. In the soft $\gamma$-ray band covered by IBIS/ISGRI, \hbox{PSR~J1846$-$0258} is clearly detected in the 20--100\,keV band (see Table~\ref{TabSpecPar}), while with similar exposure we can place a $2\sigma$ upper limit of $3.3\times10^{33}$\,erg\,s$^{-1}$ on any emission from the location of \hbox{PSR~J1119$-$6127} (using a distance of 8.4\,kpc, \citealt{CaswellMcClure-GriffithsCheung2004}) in the same energy band. Also in the X-ray band (0.5--10\,keV), these two pulsars show contrasting spectral behaviour. While \hbox{PSR~J1846$-$0258} shows absorbed power law spectra from both the pulsar and PWN \citep{HelfandCollinsGotthelf2003}, \hbox{PSR~J1119$-$6127} shows a strong thermal spectrum superimposed on a hard power law \citep{GonzalezKaspiCamilo2005}. Although \citet{GonzalezKaspiCamilo2005} attribute the hard power law component to the PWN, the thermal X-rays from the pulsar are difficult to interpret through conventional emission models, e.g. models for initial cooling of the entire neutron star result in a radius smaller than allowed by neutron star equations of state, while hot spot models result in unusually high temperature estimates. It may be the case that, while these two pulsars show similar current spin characteristics, their evolutionary paths and physical properties may be very different, as reflected in the significantly different emission properties. The most unusual feature of \hbox{PSR~J1846$-$0258} is its efficiency in converting spin-down power, $\dot E$, into X- and $\gamma$-ray luminosity $L_{\rm X}$ and $L_{\rm \gamma}$. Using data from 41 pulsars, \citet{PossentiCeruttiColpi2002} determined a general relation between the 2--10\,keV X-ray luminosity of the combined pulsar and PWN, and the spin-down power. According to this relation, and the value of $\dot E$ determined from the spin characteristics, $L_{\rm 2-10\,keV}/\dot{E}$ should be $\sim 0.2$\% for \hbox{PSR~J1846$-$0258}. The measured efficiency, at $L_{\rm 2-10\,keV}/\dot{E}\sim 12$\%, when combined with the 20--100\,keV measured in this work indicates $L_{\rm 2-10,20-100\,keV}/\dot{E}\sim27$\%. It is expected that the total efficiency in converting spin-down power into luminosity is actually much larger than this, as the luminosity contribution between 10--20\,keV is not included and the soft $\gamma$-ray flux extends beyond 100\,keV (see Fig.~\ref{FigSpec}). The observation that the 2--10\,keV spin-down conversion efficiency is so much larger than average, and the fact that this efficiency becomes even larger upon inclusion of the 20--100\,keV luminosity, may be a reason to call the distance estimate into question. To this effect, we can compare the efficiency of \hbox{PSR~J1846$-$0258} with two similar young pulsars in the Large Magellanic Cloud, for which the distance is well known at $\sim50$\,kpc. Including luminosity contributions from both the pulsar and PWN \hbox{PSR~J0537$-$6910} has $L_{\rm 0.5-10\,keV}/\dot{E}=0.9$\% \citep{ChenWangGotthelf2006} and \hbox{PSR~B0540$-$69} has $L_{\rm 0.5-10\,keV}/\dot{E}=9.1$\% \citep{KaaretMarshallAldcroft2001}. Even though the estimate for \hbox{PSR~B0540$-$69} is much larger than the X-ray efficiency of most pulsars, it is still only around half that of \hbox{PSR~J1846$-$0258} ($L_{\rm 0.5-10\,keV}/\dot{E}=20$\%). \citet{BeckerHelfand1984} determined a distance of 19\,kpc through neutral hydrogen measurements. This distance, which agrees with the estimate of \citet{Milne1979} based on the surface brightness -- diameter relation for supernovae and the neutral hydrogen measurements of \citet{CaswellMurrayRoger1975}, does imply a very high mean expansion velocity for the supernova remnant \citep{HelfandCollinsGotthelf2003}. On the other hand, if the distance were overestimated, this would lead to inconsistency between the neutral hydrogen column as measured by \citet{BeckerHelfand1984} and that deduced from fits to X-ray data of the PWN \citep{HelfandCollinsGotthelf2003} -- such inconsistency is not observed. \citet{MortonSlaneBorkowski2007} also discuss the possibility of an incorrect distance estimate, but dismiss this option as it would double the already high inferred density of postshock gas in the SNR. We can estimate a lower limit on the distance of the system by employing the half width at half maximum of the Si line as measured by \citet{HelfandCollinsGotthelf2003}. Assuming that the SNR has been expanding at this measured rate (1850\,km\,s$^{-1}$) for its entire lifetime (an upper limit of 884 years, \citealt{LivingstonKaspiGotthelf2006}), we can estimate the distance from the angular size of the SNR on the sky to be $\sim3$\,kpc. The real distance is certainly larger than this lower limit due to the fact that the SNR expansion velocity at earlier times was most likely larger than our current estimate. Using this lower limit on the distance would provide $L_{\rm 0.5-10\,keV}/\dot{E}\sim 0.5$\% -- more in line with the efficiencies of other rotation-powered pulsar systems. This makes the prospect of a distance smaller than 19\,kpc to \hbox{PSR~J1846$-$0258} appealing, but difficult to substantiate in terms of the existing estimates \citep{CaswellMurrayRoger1975,Milne1979,BeckerHelfand1984}. Of the rotation-powered pulsars that exhibit X-ray emission, only very few have been detected above 10\,keV. Although a consistent treatment exploring the general properties of PWN in the 20--100\,keV range will be presented as a future publication, we can see from the existing literature that the \integral spectrum of \hbox{PSR~J1617$-$5055}, a young rotation powered pulsar, shows a power law slope of $\sim 2$ \citep{LandiDeRosaDean2007}, while a combined IBIS/ISGRI and BeppoSAX spectrum of the PWN in \hbox{PSR~B1509$-$58} \citep{ForotHermsenRenaud2006} can be fit with a power law of photon index $\sim2.1$ -- both close to the spectral shape of \hbox{PSR~J1846$-$0258}. In addition, the EGRET instrument aboard the Compton Gamma-Ray Observatory detected emission from $\sim7$ rotation-powered pulsars and/or their PWN in the 30\,MeV--20\,GeV energy range \citep{Roberts2005}. These pulsars have photon indices harder than 2.2, also consistent with the soft \gm ray spectral slope of \hbox{PSR~J1846$-$0258} in this work and with those \integral observations quoted above. However, smooth spectral coverage above 100\,keV and into the GeV energy range is required to ascertain the presence or absence of spectral turnovers predicted by models of PWN emission mechanisms \citep{ZhangHarding2000,ChengHoRuderman1986a}. | 7 | 10 | 0710.5727 |
0710 | 0710.5511_arXiv.txt | % In hierarchical galaxy formation the stellar halos of galaxies are formed by the accretion of minor satellites and therefore contain valuable information about the (early) assembly process of galaxies. Our GHOSTS survey measures the stellar envelope properties of 14 nearby disk galaxies by imaging their resolved stellar populations with HST/ACS\&WFPC2. Most of the massive galaxies in the sample ($V_{\rm rot}$$>$200 km/s) have very extended stellar envelopes with $\mu(r)$$\sim$$r^{-2.5}$ power law profiles in the outer regions. For these massive galaxies there is some evidence that the stellar surface density of the profiles correlates with Hubble type and bulge-to-disk ratio, begging the question whether these envelopes are more related to bulges than to a Milky Way-type stellar halo. Smaller galaxies ($V_{\rm rot}$$\sim$100 km/s) have much smaller stellar envelopes, but depending on geometry, they could still be more luminous than expected from satellite remnants in hierarchical galaxy formation models. Alternatively, they could be created by disk heating through the bombardment of small dark matter sub-halos. We find that galaxies show varying amounts of halo substructure. | We select RGB stars from our CMDs and use those to trace the stellar surface density along the minor axis. RGB stars are ideal as they are abundant in our CMDs, are indicative of old stellar populations (as expected to be found in the outskirts of galaxies), and are representative of the underlying stellar mass. To map the surface brightness profiles in the central regions of the galaxies we use the integrated light from Spitzer/IRAC 4.5 micron observations. The Spitzer images provide near unobscured light profiles, even for our edge-on galaxies. We scale the RGB surface density star counts such that they match the IR luminosity profiles in the overlapping region. In this way we derive equivalent surface brightness profiles directly from the RGB star counts. \begin{figure} \mbox{ \epsfclipon \epsfxsize=0.49\textwidth \epsfbox[83 207 500 550]{prof_min_linr00.eps} \epsfclipon \epsfxsize=0.49\textwidth \epsfbox[83 207 500 550]{prof_min_logr01.eps} } \vspace*{-5mm} \caption{ Minor axis surface density profiles of GHOSTS galaxies. The thin solid lines indicate the profiles derived from Spitzer/IRAC 4.5 micron images calibrated to Vega magnitudes (add about 3.5 mag to convert to Vega $V$-mag). The symbols connected with dotted lines represent RGB star count profiles, scaled to match the Spitzer data. To reduce confusion at small radii we only plot star counts beyond 12 kpc for the non-edge-on galaxies NGC\,3031/M81 and NGC\,5236/M83. On a linear radial scale (left diagram) exponential disks appear as straight lines, as indicated for IC5052 by the dot-dashed thick line. In the log-log plot on the right, where we have removed low mass galaxies for clarity, a straight line indicates a power law profile (e.g. thin dashed line = r$^{-2.5}$). Also show are an exponential disk (for NGC\,0891, dot-dashed line) and a S\'ersic profile with the typical parameters for a flattened stellar halo as modeled by \citet{AbaNav06} (thick dashed line). \vspace{-1mm} \label{profs} } \end{figure} In Fig.\,\ref{profs} we show profiles of the edge-on galaxies analyzed so far, along with outer profiles for the more face-on galaxies NGC\,3031/M81 and NGC\,5236/M83% . The exponential thin disks only dominate the inner $\sim$2--3 kpc (5-10 scale heights), while the extended components are evident at larger radii. We find that eight of the nine galaxies analyzed thus far show components that are more extended than the exponential disks detected at small radii. Several of the most massive galaxies have very extended envelopes with stellar densities at 30 kpc that are 10--100 times higher than the contamination background, equivalent to $\sim$29 $V$-mag arcsec$^{-2}$. NGC\,5236/M83 is the only galaxy with a pure exponential disk to the last measured point at 20 kpc (more than 10 disk scale lengths). \subsubsection{The bulge-halo connection} In this section we explore the connection between bulges and the extended components. We model the minor axis profiles by combining a S\'ersic profile and an exponential disk. Merger models show that the hot components resulting after a violent relaxation generally exhibit a S\'ersic profile \citep[e.g.,][and reference therein]{BarHer92,AbaNav06}. If bulges and stellar envelopes are created by a collisionless merger processes, we thus expect their light to follow a S\'ersic profile. For a number of massive galaxies ($V_{\rm rot}$$>$200 km\,s$^{-1}$) we can fit the entire minor axis profile over a factor of 1000 in size ($\sim$10$^{4.5}$ in surface density) by an exponential disk and a single S\'ersic profile, representing both the inner bulge-like region and the outer envelope. This is, for instance, the case for the bulge dominated NGC\,7814 or a galaxy like NGC\,3031/M81, which has a power law outer envelope of rather steep slope. Other galaxies, like NGC\,4565% , have too shallow an outer slope compared to their concentrated bulge region to be fitted by a S\'ersic profile. NGC\,891 can be fitted by an exponential disk and a S\'ersic profile from central bulge to outer envelope if we ignore our outermost field and presume that the higher star density at 30 kpc is due to substructure. Finally NGC\,5236/M83, which has only a small bulge, shows no sign of an outer envelope out to ten disk scale lengths. The smallest galaxies in the sample ($V_{\rm rot}$$\simeq$100--120 km\,s$^{-1}$) have small extended components, barely discernible above the background contamination (see Fig.\,\ref{profs}). The shape of the extended component is thus poorly constrained due to both the uncertainty in the background and low number statistics. The star counts can be fitted equally well by exponential, power law, and S\'ersic law profiles. This feature could be the thick disk, as observations of the NGC\,4244 major axis suggest the component is very flattened. However, the RGB main disk scale height is already twice that of the main sequence population and has been argued already represent the thick disk \citep{Seth05II} with the feature observed here being an additional component. These additional components are most likely (depending on exact shape) more luminous than predicted in the hierarchical models of \citet{PurBul07}, but could have been created by the bombardment of small dark matter sub-halos \citep{KazBul07}. Therefore, a number of the observed extended envelopes seem structurally directly related to the central bulge regions, like in NGC\,7814. In other galaxies, where the bulge is too concentrated to be simply related to the outer envelope, we can suspect that secular evolution (e.g., bar driven central enhancement and thickening) can account for the extra (pseudo-)bulge light. NGC\,4565 with its boxy bulge could nicely fall in this category. In small galaxies the extended envelope is unrelated to the central region, as these small galaxies have no bulge. \subsubsection{Envelope properties and halo models} Comparing the envelopes of the different galaxies we find that the two small galaxies have much smaller extended components than the larger galaxies, with surface densities that are lower relative to their disks. The more massive galaxies in our sample are very similar in terms of mass, luminosity, and scale size. Still, there is significant variation in outer envelope properties. At 20 kpc NGC\,891, NGC\,4565, and NGC\,7814 have power law profiles with a slope of about -2.5. NGC\,3031/M81 has a steeper profile, while at 20 kpc NGC\,5236 is still dominated by the (face-on) disk. At first sight, the envelope luminosity at 20 kpc seems correlated with Hubble type and bulge-to-disk ratio, with the bulge dominated NGC\,7814 being the brightest and the late-type spiral NGC\,5236 showing no sign of an envelope at all. Although M31 also fits this trend, the Sab galaxy NGC\,3031/M81 does not, as a steeper and fainter profile is evident at 20 kpc. In Fig.\,\ref{profs} we also show a typical profile from the \citet{AbaNav06} model of accreted stars. Our profiles are a bit shallower and mostly fainter than these models between 10 and 30 kpc. While the surface density normalization may be somewhat uncertain in the models, the shape is quite well constrained. It could be that the true halos only dominate at even larger radii and the slope becomes even shallower at larger radii. However, in hierarchical galaxy formation the halo and ``classical'' bulge are formed by the same merging process, so there is no reason to suspect a large structural change between bulge and halo. The S\'ersic radii derived for our combined envelope and bulge fits are typically eight times smaller than those of \citet{AbaNav06}. However, a number of simulation parameters affect the concentration of the accreted halos. Increasing star formation suppression in small sub-halos, such that only the most massive dark matter sub-halos contain stars, yields steeper and fainter envelopes \citep{BekChi05}. Alternatively, the stars in the accreted satellites could sit deeper in the potential wells of their dark matter sub-halos than simulated in these models, thereby only being tidally stripped closer to the main galaxy, also resulting in more concentrated halos \citep{BulJoh05}. | 7 | 10 | 0710.5511 |
|
0710 | 0710.0272_arXiv.txt | {ANTARES is a large volume neutrino telescope currently under construction off La Seyne-sur-mer, France, at 2475m depth. Neutrino telescopes aim at detecting neutrinos as a new probe for a sky study at energies greater than 1 TeV. The detection principle relies on the observation, using photomultipliers, of the Cherenkov light emitted by charged leptons induced by neutrino interactions in the surrounding detector medium. Since late January 2007, the ANTARES detector consists of 5 lines, comprising 75 optical detectors each, connected to the shore via a 40 km long undersea cable. The data from these lines not only allow an extensive study of the detector properties but also the reconstruction of downward going cosmic ray muons and the search for the first upward going neutrino induced muons.The operation of these lines follows on from that of the ANTARES instrumentation line, which has provided data for more than a year on the detector stability and the environmental conditions. The full 12 line detector is planned to be fully operational early 2008.} \begin{document} | Great achievements have been made by the Antares collaboration in the last year. The detector is working in nominal mode with 5 lines and should be complete early 2008. Upward neutrino candidates have been found that validate the conceptual method and the chosen techniques. Very exciting times have started with a detector looking for neutrinos in a region of the celestial sky which has never been studied with such a level of sensitivity. | 7 | 10 | 0710.0272 |
|
0710 | 0710.0758_arXiv.txt | The star HDE 226868 known as an optical counterpart of the black hole candidate Cyg X-1 has been observed in H$_\alpha$ region using spectrograph at Ond\v{r}ejov 2-m telescope. The orbital parameters are determined from He\,I-line by means of the author's method of Fourier disentangling. Preliminary results are also presented of disentangling the H$_\alpha$-line into a P-Cyg profile of the (optical) primary and an emission profile of the circumstellar matter (and a telluric component). | The bright X-ray source Cyg X-1 has been identified with the star denoted as HDE 226868, V1357 Cyg or BD+34$^{\circ}$3815 etc. An improvement of instrumentation of the Ond\v{r}ejov 2-m telescope enabled to start with systematic observations of this target of magnitude V$\simeq$8.9, B$\simeq$9.6. With coordinates $\alpha_{2000}=19^{h}58^{m}21.7^{s}$, $\delta_{2000}=+35^{\circ}12'6''$ it is well observable from Ond\v{r}ejov mainly at summer. It is known to be an interacting binary with period $P\simeq 5.6 d$. The primary component is a supergiant of spectral type classified as B0 (or O9.7) Iab with temperature $T_{\rm eff}=30400\pm500$ K and log $g=3.31\pm0.07$. This primary, which nearly fills its Roche lobe, shows signs of variable strong stellar wind and an overabundace of He and heavier elements (cf. e.g. Karitskaya et al. 2007). The secondary component invisible in optical radiation is a compact object, most probably a black hole. This companion, or its neighborhood emits a variable X-radiation, which is supposed to originate from an accretion disk fed by the stellar wind from the primary. The X-radiation switches chaotically between two states. In the low/hard state the total X-ray flux is low and the spectrum is flat, so that the hard tail of X-radiation prevails. In the high/soft state the soft radiation is enhanced more, and consequently the spectrum has a steeper decrease toward the higher energies and hence the radiation is softer in the mean. Some intermediate states may also appear temporarily. The X-ray flux is anticorrelated with the strength of emission in the H$_{\alpha}$-line of the primary: in the X-low/hard state the H$_{\alpha}$ emission is strong, while in the X-high/soft state the H$_{\alpha}$ emission is weak. The aim of the observational campaign at Ond\v{r}ejov observatory was to improve orbital parameters of the system, to check a possible spectroscopic features connected with the circumstellar matter (either accretion disk around the black hole, gaseous streams or stellar wind) or with a possible third body, and to get line-profiles enabling a quantitative comparison with a model of the atmosphere and stellar wind of the primary. The first part of obtained spectra was provided for a study on Cyg X-1 organized in a wide international collaboration, the results of which should appear in Gies et al. (2007). In the present contribution, results obtained using the author's method of spectra disentangling from the same set of Ond\v{r}ejov spectra are given. A more detailed study taking into account also recently obtained spectra is in progress. | 7 | 10 | 0710.0758 |
|
0710 | 0710.0875_arXiv.txt | Theoretical arguments and indirect observational evidence suggest that the stellar initial mass function (IMF) may evolve with time, such that it is more weighted toward high mass stars at higher redshift. Here we test this idea by comparing the rate of luminosity evolution of massive early-type galaxies in clusters at $0.02\leq z\leq 0.83$ to the rate of their color evolution. A combined fit to the rest-frame $U-V$ color evolution and the previously measured evolution of the $M/L_B$ ratio gives $x = -0.3^{+0.4}_{-0.7}$ for the logarithmic slope of the IMF in the region around 1\,\msun, significantly flatter than the present-day value in the Milky Way disk of $x= 1.3\pm 0.3$. The best-fitting luminosity-weighted formation redshift of the stars in massive cluster galaxies is $3.7^{+2.3}_{-0.8}$, and a possible interpretation is that the characteristic mass $m_c$ had a value of $\sim 2$\,\msun\ at $z\sim 4$ (compared to $m_c \sim 0.1$\,\msun\ today), in qualitative agreement with models in which the characteristic mass is a function of the Jeans mass in molecular clouds. Such a ``bottom-light'' IMF for massive cluster galaxies has significant implications for the interpretation of measurements of galaxy formation and evolution. Applying a simple form of IMF evolution to literature data, we find that the volume-averaged star formation rate at high redshift may have been overestimated (by a factor of $3-4$ at $z> 4$), and the cosmic star formation history may have a fairly well-defined peak at $z\sim 1.5$. The $M/L_V$ ratios of galaxies are less affected than their star formation rates, and future data on the stellar mass density at $z>3$ will provide further constraints on IMF evolution. The formal errors likely underestimate the uncertainties, and confirmation of these results requires a larger sample of clusters and the inclusion of redder rest-frame colors in the analysis. | The form of the stellar initial mass function (IMF) is of fundamental importance for many areas of astrophysics and a topic of considerable debate (see, e.g., {Schmidt} 1959; {Miller} \& {Scalo} 1979; {Scalo} 1986; {Larson} 1998, 2003; {Kroupa} 2002; {Chabrier} 2003, for reviews). Measurements of the IMF are difficult and somewhat model-dependent as they require the conversion of the observed present-day luminosity function of a stellar population to its mass function at birth. Best estimates for the Galactic disk suggest that the IMF has a powerlaw slope at $m\gtrsim 1$\,\msun, and turns over at lower masses ({Kroupa} 2001; {Chabrier} 2003). This turnover can be modeled by a broken powerlaw ({Kroupa} 2001) or by a log-normal distribution with a characteristic mass $m_c$ ({Chabrier} 2003). The value of $m_c$ is $\sim 0.1$\,\msun\ in the disk of the Milky Way, with considerable uncertainty. The powerlaw slope at high masses is probably close to the {Salpeter} (1955) value of $x=1.35$, with an uncertainty of $\sim 0.3$ ({Scalo} 1986; {Chabrier} 2003). Although there is no direct evidence for dramatic variations of the IMF within the present-day Milky Way disk (e.g., {Kroupa} 2001; {Chabrier} 2003), this does not preclude variations with time, metallicity, and/or environment. In particular, {Larson} (1998, 2005) has argued that the characteristic turnover mass may be largely determined by the thermal Jeans mass, which strongly depends on temperature ($\propto T^{3/2}$ at fixed density). In the context of this model one might expect that heating by ambient far-infrared radiation would disfavor the formation of low mass stars in extreme environments, such as in super star clusters and in the center of the Milky Way. Other models emphasize the role of turbulence as opposed to temperature in determining the distribution of protostellar clumps (e.g., {Padoan} \& {Nordlund} 2002), and in these models the role of the environment may be less direct (see {McKee} \& {Ostriker} [2007] for a recent review of various models to explain the characteristics of the IMF). Observations may support the notion of a top-heavy (or ``bottom-light'') IMF in extreme environments. Some young super star clusters in M82 appear to have a top-heavy mass function (e.g., {Rieke} {et~al.} 1993; {McCrady}, {Gilbert}, \& {Graham} 2003), as do clusters in the Galactic center region (e.g., {Figer} {et~al.} 1999; {Stolte} {et~al.} 2005; {Maness} {et~al.} 2007). The interpretation of observed mass functions is complicated by dynamical effects, which tend to make the mass function more top-heavy over time, in particular in the central regions of star clusters (see, e.g., {McCrady} {et~al.} 2003; {McCrady}, {Graham}, \& {Vacca} 2005; {Kim} {et~al.} 2006). Recently Harayama, Eisenhauer, \& Martins (2007) studied the IMF of NGC 3603, one of the most massive Galactic star-forming regions, out to large radii and conclude that its IMF is substantially flatter than Salpeter for masses $0.4 -20$\,\msun. The IMF may also depend on redshift. At earlier times star formation presumably occurred more often in a burst mode than in a relatively gradual ``disk'' mode (e.g., {Steidel} {et~al.} 1996; {Blain} {et~al.} 1999b; {Lacey} {et~al.} 2007), which means that the IMF could generally be more skewed toward high mass stars at redshifts 1--3 and beyond. Furthermore, the average metallicity in star forming clouds was lower at higher redshift, which may have led to an extremely top-heavy IMF for the first generation of stars (e.g., {Abel}, {Bryan}, \& {Norman} 2002; {Bromm}, {Coppi}, \& {Larson} 2002). Finally, the cosmic microwave background (CMB) radiation sets a floor to the ambient temperature, and hence the Jeans mass, which scales with $(1+z)$. Beyond $z\sim 2$ the CMB temperature exceeds the typical temperatures of dense prestellar cores in Galactic molecular clouds (e.g. {Evans} {et~al.} 2001; {Tafalla} {et~al.} 2004). Therefore, at sufficiently high redshift the characteristic mass may be expected to evolve roughly as $m_c \propto (1+z)^{3/2}$, leading to IMFs which have a reduced fraction of low mass stars ({Larson} 1998). The effects of the CMB are even more pronounced when its influence on the pressure in star-forming clouds is taken into account, and {Larson} (2005) suggests that at $z=5$ the characteristic mass may be higher than today's value by as much as an order of magnitude. Such rapid evolution of the IMF would have important consequences for determinations of masses and star formation rates of distant galaxies, and for measurements of evolution in these properties. It is very difficult to constrain the IMF at early times directly, as the light of high redshift galaxies is completely dominated by massive stars. The extremely blue rest-frame UV colors of galaxies at $z\sim 6$ may imply a top-heavy IMF ({Stanway}, {McMahon}, \& {Bunker} 2005), although this is just one of several possible explanations. From observations of a lensed Lyman break galaxy there is some evidence that the {\em slope} of the IMF at $z\sim 3$ is similar to the Salpeter value at the high mass end ({Pettini} {et~al.} 2000), but there is essentially no information on stars with masses near or below 1\,\msun. Fortunately, the form of the high redshift IMF has implications for the properties of galaxies at much lower redshift, as all stars with masses $\lesssim 0.8$\,\msun\ that formed in the history of the Universe are still with us today. Tumlinson (2007) finds that the properties of carbon-enhanced metal-poor stars in our Galaxy are best explained with a relatively high number of stars in the mass range 1--8\,\msun\ at high redshift. Various other constraints obtained from galaxies at low redshift (including our own) are reviewed in {Chabrier} (2003). Of particular interest are the stellar populations of massive early-type galaxies, as they are very homogeneous and should reflect conditions in star forming regions at $z> 2$. Recently, {Cappellari} {et~al.} (2006) used the kinematics of elliptical galaxies to constrain the IMF, as the dynamical $M/L$ ratio provides an upper limit to the amount of mass that can be locked up in low mass stars. Current data appear to rule out a {Salpeter} (1955) (or steeper) IMF, but are consistent with {Kroupa} (2001) and {Chabrier} (2003) IMFs ({Cappellari} {et~al.} 2006). In this paper, we provide new constraints on the IMF at high redshift by comparing the {\em evolution} of the $M/L$ ratios of early-type galaxies to their color evolution. This method was first suggested by {Tinsley} (1980), but data of sufficient accuracy are only now becoming available. The method is sensitive to the IMF in the important mass range around 1\,\msun, where the effects of an evolving characteristic mass might be expected to manifest themselves. A plan of the paper follows. In \S\,2 a relation between color evolution, luminosity evolution, and the logarithmic slope of the IMF $x$ is derived using stellar population synthesis models. In \S\,3 published data and archival HST images of galaxy clusters are used to construct the redshift evolution of the $U-V$ color-mass relation. In \S\,4 the color evolution from \S\,3 is combined with the previously measured evolution of the $M/L_B$ ratio. The relations from \S\,2 are then used to derive constraints on the IMF slope $x$ from the combined color and luminosity evolution. Section 5 is devoted to the (many) systematic uncertainties in the methodology and in the data, and \S\,6 asks whether our results are consistent with other constraints on the stellar populations of massive early-type galaxies. Although the constraints we derive in this paper are subject to many uncertainties, it is interesting to explore their consequences. In \S\,7 the fitting results of \S\,4 are interpreted in the context of an evolving characteristic mass $m_c$. The data on cluster galaxies are combined with previous constraints on $m_c$ for globular clusters and submm-galaxies, and a simple form of IMF evolution is proposed. This evolution is then applied to literature data on the evolution of the volume-averaged star formation rate and stellar mass density. The key results are summarized in \S\,8. We assume $\Omega_m=0.3$, $\Omega_{\Lambda}=0.7$, and $H_0=71$\,\kms\,Mpc$^{-1}$ where needed. | \label{conclusion.sec} This paper compares the color evolution of massive cluster galaxies to their luminosity evolution, with the aim of constraining the form of the IMF at the time when the stars in these galaxies were formed. It is found that the evolution of the rest-frame $U-V$ color is not consistent with the previously determined evolution of the rest-frame $M/L_B$ ratio, unless the IMF slope is significantly flatter than the Salpeter value around 1\,\msun. For standard IMFs with a slope of 1.3 at $m\geq 1$\,\msun\ the luminosity evolution is too fast for the measured color evolution, and the implied stellar ages derived from $M/L$ evolution and color evolution are not consistent with each other. The only models that are able to fit the color evolution and the luminosity evolution simultaneously have IMF slopes of $\sim 0$ around 1\,\msun\ and mean luminosity-weighted stellar formation redshifts of $\sim 4$ (for Solar metallicity). This result is somewhat uncertain, as the currently available sample of cluster galaxies with accurate rest-frame $U-V$ colors and dynamical masses is somewhat limited and there are many systematic effects which may play a role. In particular, it is an open question whether stellar population synthesis models are able to predict color evolution with the required accuracy. The commonly used {Bruzual} \& {Charlot} (2003) and {Maraston} (2005) models give broadly similar answers, but that may be because they share many of the same assumptions. As discussed in \S\,\ref{comp.sec} the higher stellar ages implied by a flat IMF are consistent with many other studies, which lends some credibility to the results presented here. Of particular importance is the agreement with the data on Balmer line strengths of {Kelson} {et~al.} (2001), as they do not suffer from the same systematic uncertainties as the color data. Formation redshifts substantially larger than two also fit more comfortably with the direct detection of old galaxies at high redshifts. A firm independent measurement of the star formation epoch of massive cluster galaxies, combined with a better understanding of selection effects at high redshift, would leave the IMF as the only free parameter and greatly simplify the problem. The implications discussed in \S\,\ref{imply.sec} are obviously somewhat speculative. Although the interpretation in terms of an evolving characteristic mass is physically plausible according to some models (e.g., {Larson} 2005), many other forms of the IMF are consistent with the data. The observations described in this paper are only sensitive to a narrow mass range near 1\,\msun, and the IMF proposed in Eq.\ \ref{mod.eq} represents a very substantial extrapolation. This is illustrated in Fig.\ \ref{summary.plot}: both the top-heavy IMF (red line) and the ``bottom-light'' IMF (green line) are consistent with the data presented in this paper. The main reason for preferring the bottom-light IMF over a top-heavy form is that the absolute $M/L$ ratios of galaxies are within a reasonable range. As shown in Fig.\ \ref{mlmc.plot} the $M/L_V$ ratios are similar to those implied by a standard {Chabrier} (2003) IMF, which means that they are consistent with dynamical measurements at $z=0$ ({Cappellari} {et~al.} 2006). \vbox{ \begin{center} \leavevmode \hbox{% \epsfxsize=8.5cm \epsffile{f15.eps}} \figcaption{\small Illustration of the key results of this paper. In \S\,4 the slope of the IMF of massive cluster galaxies $x$ is estimated to be approximately $-0.3$ in a narrow region around 1\,\msun\ (thick black line). The red dashed line and the green solid line show two possible interpretations: a global change of the slope of the IMF (with respect to a standard Chabrier or Salpeter IMF) at all masses (red dashed line), or a change in the characteristic mass (solid green line). The ``top-heavy'' interpretation is inconsistent with the dynamical $M/L$ ratios of nearby elliptical galaxies (see \S\,\ref{absml.sec}), whereas the ``bottom-light'' interpretation is consistent with all data that we are aware of. The blue and yellow areas illustrate that stars with masses $\gtrsim 10$\,\msun\ drive star formation measurements, whereas stars with masses $1-5$\,\msun\ drive $M/L$ measurements (see \S\,7.3-7.5). \label{summary.plot}} \end{center}} An ``unintended'' effect of an evolving IMF of the form proposed here is that it reduces the discrepancy between the observed stellar mass density and the density implied by the cosmic star formation history. This result is in excellent (albeit qualitative) agreement with several other recent studies (e.g., Hopkins \& Beacom 2006, Fardal et al.\ 2007, P\'erez-Gonz\'alez et al.\ 2007; Wilkins, Trentham, \& Hopkins 2007; Dav\'e 2007; see also, e.g., Fields 1999). The differences between a non-evolving IMF and an evolving IMF are fairly large at $z\sim 4$ (as the effect on the star formation rate is strong and the effect on $M/L$ ratios is weak at that redshift), and it will be interesting to see where future measurements of the mass density at high redshift will fall in Fig.\ \ref{mdens.plot}. We note that the effects on the $M/L_V$ ratios are somewhat uncertain, as they rely on rather rudimentary stellar population synthesis modeling. The effects on star formation rates are more robust, and suggest that the cosmic star formation rate peaked at $z\sim 1.5$. This paper adds to previous theoretically and observationally motivated suggestions that the IMF may evolve with redshift (e.g., Worthey et al.\ 1992; Larson 1998, 2005; Fields 1999; Blain et al.\ 1999a; Baugh et al.\ 2005; Stanway et al.\ 2005; Hopkins \& Beacom 2006; Tumlinson 2007; Lacey et al.\ 2007; Fardal et al.\ 2007; P\'erez-Gonz\'alez et al.\ 2007). Although these studies vary greatly in their parameterization of IMF evolution and the range of stellar masses that are considered, they all suggest that the ratio of high-mass stars to low-mass stars was higher in the past. It should be pointed out that most of these papers invoke a change in the IMF as a ``last resort'' possibility, to explain data that are otherwise difficult to interpret. In the present study a different approach was followed, in that we set out with the specific purpose of constraining the slope of the IMF. An advantage of the applied method is that it is fairly direct, as the rate of luminosity evolution is determined by the number of stars as a function of mass. Disadvantages are that it is only sensitive to a very limited mass range (see Fig.\ \ref{summary.plot}); that it relies on stellar population synthesis models, which are not well calibrated in the relevant parameter range; that the progenitors of early-type galaxies may not be representative for the general population of high redshift galaxies; and that the currently available data are somewhat limited. Accepting the possibility of an evolving characteristic mass, it is interesting to speculate what could be the cause, or causes. The proximate cause may well be a higher temperature in molecular clouds at high redshift, which would raise the Jeans mass and could inhibit the formation of low mass stars ({Larson} 1998, 2005). The ultimate cause could be the higher temperature of the cosmic microwave background, the fact that star formation tends to proceed in more extreme environments at higher redshift, or a combination. Available information on IR-bright galaxies suggests that dust temperatures in star burst galaxies are of order $30-40$\,K ({Dunne} {et~al.} 2000; {Chapman} {et~al.} 2005), and hence exceed the CMB temperature for all relevant redshifts. However, it is as yet unclear what fraction of the total star formation has taken place in these extreme environments (see, e.g., {Reddy} {et~al.} 2007). The analysis presented here can be improved in various ways. The number of clusters with accurate rest-frame $U-V$ is currently smaller than the number of clusters with accurate $M/L_B$ measurements, and this can be remedied by obtaining accurate (space-based) photometry in well-chosen filters of the remaining clusters in the vv07 sample. It is also important to measure the evolution in a redder rest-frame color, such as $V-I$. Redder color suffer less from possible contributions of hot stars, and their evolution is probably somewhat better calibrated in stellar population synthesis models. This requires very accurate photometry in the near-infrared, which should be possible with WFC3 on HST. On the modeling side, it would be helpful to implement more variations of the IMF in stellar population synthesis codes than the standard Salpeter, Kroupa, and Chabrier forms. Ultimately it may be fruitful (or prove necessary) to have the characteristic mass, or some other parameter describing the form of the IMF, as one of the ``standard'' parameters in these models, on a par with the age and metallicity. | 7 | 10 | 0710.0875 |
0710 | 0710.2041_arXiv.txt | \noindent The generalized Darmois--Israel formalism for Einstein--Gauss--Bonnet theory is applied to construct thin-shell Lorentzian wormholes with spherical symmetry. We calculate the energy localized on the shell, and we find that for certain values of the parameters wormholes could be supported by matter not violating the energy conditions. | For spacetime dimension $D\geq 5$ the Einstein--Hilbert action of gravity admits quadratic corrections constructed from coordinate-invariant tensors which scale as fourth derivatives of the metric. In particular, when $D=5$ the most general theory leading to second order equations for the metric is the so-called Einstein--Gauss--Bonnet theory or Lovelock theory up to second order. This class of model for higher dimensional gravity has been widely studied, in particular because it can be obtained in the low energy limit of string theory \cite{1}. For spacetime dimensions $D<5$ the Gauss--Bonnet terms in the action represent a topological invariant. The equations of gravitation admit solutions, known as Lorentzian wormholes, which connect two regions of the same universe (or of two universes) by a throat, which is a minimal area surface \cite{motho,book}. Such kind of geometries would present some features of particular interest, as for example the possibility of time travel (see Refs. \cite{morris-novikov}). But a central objection against the actual existence of wormholes is that in Einstein gravity the flare-out condition \cite{hovis1} to be satisfied at the throat requires the presence of exotic matter, that is, matter violating the energy conditions \cite{book}. In this sense, thin-shell wormholes have the advantage that the exotic matter would be located only at the shell. However, it has recently been shown \cite{gra-wi} that in pure Gauss--Bonnet gravity exotic matter is no needed for wormholes to exist; in fact, they could exist even with no matter (see also Refs. \cite{7,maeda}). In this work we thus study thin-shell wormholes in Einstein--Gauss--Bonnet gravity. We focus in the amount of matter necessary for supporting the wormholes, without analyzing the microphysics explaining this matter. Differing from the approach in the related work Ref. \cite{grg06}, where the Gauss--Bonnet terms were treated as an effective source for the Einstein's field equations, here the Gauss--Bonnet contribution is treated as an essentially geometrical object. This requires a generalization \cite{4} of the Darmois--Israel formalism \cite{3} for thin shells, but provides a better physical understanding. In particular, we show that for certain values of the parameters, thin-shell wormholes could be supported by matter not violating the energy conditions. | Motivated by the results within pure Gauss--Bonnet gravity (i.e. without Einstein term) in Ref. \cite{gra-wi}, here we evaluate the amount of exotic matter and the energy conditions, following the approach presented above in which the Gauss--Bonnet term is treated as a geometrical contribution in the field equations. Coming this contribution from the curvature tensor, this approach is clearly the most suitable to give a precise meaning to the characterization of matter supporting the wormhole. As we shall see, the results will considerably differ from those in Ref. \cite{grg06}, where the Gauss--Bonnet term was treated as an effective source for the Einstein's field equations. The {\it weak energy condition} (WEC) states that for any timelike vector $U^{\mu}$ it must be $T_{\mu\nu}U^{\mu}U^{\nu}\geq0$; the WEC also implies, by continuity, the {\it null energy condition} (NEC), which means that for any null vector $k^{\mu}$ it mus be $T_{\mu\nu}k^{\mu}k^{\nu}\geq0$ \cite{book}. In an orthonormal basis the WEC reads $\rho\geq0$, $\rho+p_{l}\geq0\ \forall\, l$, while the NEC takes the form $\rho+p_{l}\geq0 \ \forall\, l$. In the case of thin-shell wormholes the radial pressure $p_{r}$ is zero, while within Einstein gravity or even with the inclusion of a Gauss--Bonnet term in the way proposed in \cite{grg06}, the surface energy density must fulfill $\sigma<0$, so that both energy conditions would be violated. The sign of $\sigma+p_{t}$ where $p_{t}$ is the transverse pressure is not fixed, but it depends on the values of the parameters of the system. In what follows we restrict to static configurations. The surface energy density $\sigma_{0}$ and the transverse pressure $p_{0}$ for a static configuration ($b=b_0$, $\dot{b}=0$, $\ddot{b}=0$) are given by \begin{equation} \sigma_{0}=-\frac{1}{8\pi}\left[ 6\frac{\sqrt{f(b_{0})}}{b_{0}}-2\alpha\sqrt{f(b_{0})}\left(4\frac{f(b_{0})}{b^{3}_{0}}-\frac{12}{b^{3}_{0}}\right)\right],\label{s0} \end{equation} \begin{equation} p_{0}=\frac{1}{8\pi}\left[ 4\frac{\sqrt{f(b_{0})}}{b_{0}}+\frac{f^{'}(b_{0})}{\sqrt{f(b_{0})}}-2\alpha\left(2f^{'}(b_{0})\frac{\sqrt{f(b_{0})}}{b^{2}_{0}} -2\frac{f^{'}(b_{0})}{b^{2}_{0}\sqrt{f(b_{0})} }\right)\right].\label{p0} \end{equation} Note that the sign of the surface energy density is, in principle, not fixed. The most usual choice for quantifying the amount of exotic matter in a Lorentzian wormhole is the integral \cite{nandi}: \begin{equation} \Omega= \int (\rho + p_{r})\sqrt{|g|}\,d^{4}x. \end{equation} We can introduce a new radial coordinate ${\cal R}=\pm(r-b_{0})$ with $\pm$ corresponding to each side of the shell. Then, because in our construction the energy density is located on the surface, we can write $\rho=\delta({\cal R})\,\sigma_{0}$, and because the shell does not exert radial pressure the amount of exotic matter reads \begin{equation} \Omega=\int^{2\pi}_{0} \int^{\pi}_{0}\int^{\pi}_{0}\int^{+\infty}_{-\infty}\delta({\cal R})\,\sigma_{0} \sqrt{|g|}\, d{\cal R}\,d\theta\,d\chi\ d\varphi =2\pi^{2} b_{0}^3 \sigma_{0}. \end{equation} Replacing the explicit form of $\sigma_{0}$ and $g$, we obtain the exotic matter amount as a function of the parameters that characterize the configurations: \begin{equation} {\Omega}= -\frac{3}{2}\pi b^{2}_{0}\sqrt{f(b_{0})} +2\pi\alpha\sqrt{f(b_{0})}\left[f(b_{0})-3)\right], \end{equation} where $f$ is given by (\ref{f}). For $\Lambda=0$, in the limit $\alpha\rightarrow 0$ and Taylor expanding up to zeroth order we obtain the exotic matter for the Reissner--N\"{o}rdstrom ($Q\neq 0$) and Schwarzschild ($Q= 0$) geometries. For $\alpha\neq 0$ we now find that there exist positive contributions to $\Omega$; these come from the different signs in the expression (\ref{s0}) for the surface energy density, because $\Omega$ is proportional to $\sigma_0$. We stress that this would not be possible if the standard Darmois--Israel formalism was applied, treating the Gauss--Bonnet contribution as an effective energy-momentum tensor, because this leads to $\sigma_0\sim - \sqrt{f(b_0)}/b_0$ \cite{grg06}. Now, once the explicit form of the function $f$ (with $\Lambda=0$) is introduced, the condition $\sigma_0>0$ leads to \begin{equation} -8\alpha-2b_0^2-b_0^2\sqrt{1+\frac{16M\alpha}{\pi b_0^4}-\frac{8Q^2\alpha}{3b_0^6}}>0,\label{s+} \end{equation} which can hold only for $\alpha <0$. The subsequent analysis is simplified by considering the case $Q=Q_c, \,\Lambda=0$; then there would be at most only one horizon in the original manifold, its radius being independent of the charge. But for this charge it can be shown that, for values of $\alpha$ such that the horizon exists, it is not possible to fulfil Eq. (\ref{s+}) for any wormhole radius larger than $r_{hor}$; the reason is that the horizon exists only for $\alpha > -M/(3\pi)$, which is not compatible with condition (\ref{s+}) if $b_0$ is to be larger than the corresponding horizon radius. Instead, a simple numerical analysis shows that for $\alpha$ slightly below $-M/(3\pi)$ both the singularity at $r\neq 0$ and the horizon dissapear in the original manifold, so that the only condition to be fulfilled is that given by Eq. (\ref{s+}). And for $\alpha <0$ it is always possible to choose $b_0$ such that this indeed happens, so that the existence of thin-shell wormholes is compatible with a positive surface energy density \footnote{It is not difficult to see that for $Q$ slightly below $Q_c$ the same happens for larger $|\alpha|$.}. In figures 1 and 2 we show the amount $\Omega$ as a function of the wormhole radius for this relatively large value of $|\alpha|$ (that is, $\pi |\alpha|$ of order $M$); though this would imply microscopic configurations or a scenario different from that suggested by present day observation, the analysis shows that this is the most interesting situation. Besides the fact that $\Omega$ results to be smaller when calculated by treating the Gauss--Bonnet contribution as a geometric object than in the case that it was treated as an effective energy-momentum tensor, this amount is smaller than which would be necessary in the five-dimensional pure Reissner--N\"{o}rdstrom case (see Fig. 1). However, the central, remarkable, result is that we have a region with $\Omega >0$ (see Fig. 2), corresponding to $\sigma_0>0$; and that besides, from Eqs. (\ref{s0}) and (\ref{p0}) we have $\sigma_0+p_0=-(b_0/3)\,d\sigma_0/db_0$, which shows that for wormhole radii such that $\sigma_0 >0$ and ${d\sigma_0}/{db_0}<0$ ($r_{wh}$ within the maximun and the zero of $\sigma_0$ in Fig. 3) both the WEC and the NEC are satisfied. Thus, in the picture providing a clear meaning to matter in the shell, in Einstein--Gauss--Bonnet gravity the violation of the energy conditions could be avoided, and wormholes could be supported by ordinary matter. \begin{figure}[htp] \centering \includegraphics[height=8cm, width=12cm]{compara.eps} \caption{The amount $\Omega$ is shown as a function of $r^2_{wh}/M$, for $Q=Q_c$ and $\alpha=-0.11 M$. The dashed line corresponds to the five-dimensional Reissner--N\"{o}rdstrom case, the dotted line corresponds to considering the Gauss--Bonnet term as a kind of effective source for the field equations, and the solid line shows the result obtained here with the generalized Darmois--Israel formalism for Einstein--Gauss--Bonnet theory.} \end{figure} \begin{figure}[htp] \centering \includegraphics[height=8.cm, width=8.5cm]{omega.eps} \caption{The amount $\Omega$ is shown as a function of $r^2_{wh}/M$, for $Q=Q_c$ and $\alpha=-0.11 M$. The plot shows the result obtained here with the generalized Darmois--Israel formalism for Einstein--Gauss--Bonnet theory.} \end{figure} \begin{figure}[htp] \centering \includegraphics[height=8cm, width=8cm]{wec.eps} \caption{Energy conditions: the dashed line shows $\tilde \sigma_0=\sqrt{M}\sigma_0$, the dashed-dotted line shows $\tilde p_0=\sqrt{M}p_0$ and the solid line shows the sum $\tilde \sigma_0+\tilde p_0$.} \end{figure} | 7 | 10 | 0710.2041 |
0710 | 0710.1873_arXiv.txt | Precision measurement of the scalar perturbation spectral index, $n_s$, from the \emph{Wilkinson Microwave Anisotropy Probe} temperature angular power spectrum requires the subtraction of unresolved point source power. Here we reconsider this issue, attempting to resolve inconsistencies found in the literature. First, we note a peculiarity in the \emph{WMAP} temperature likelihood's response to the source correction: Cosmological parameters do not respond to increased source errors. An alternative and more direct method for treating this error term acts more sensibly, and also shifts $n_s$ by $\sim0.3\sigma$ closer to unity. Second, we re-examine the source fit used to correct the power spectrum. This fit depends strongly on the galactic cut and the weighting of the map, indicating that either the source population or masking procedure is not isotropic. Jackknife tests appear inconsistent, causing us to assign large uncertainties to account for possible systematics. Third, we note that the \emph{WMAP} team's spectrum was computed with two different weighting schemes: uniform weights transition to inverse noise variance weights at $l=500$. The fit depends on such weighting schemes, so different corrections apply to each multipole range. For the Kp2 mask used in cosmological analysis, we prefer source corrections {$A=0.012\pm0.005$ $\mu$K$^2$} for uniform weighting and {$A=0.015\pm0.005$ $\mu$K$^2$} for $N_{\rm obs}$ weighting. Correcting \emph{WMAP}'s spectrum correspondingly, we compute cosmological parameters with our alternative likelihood, finding $n_s=0.970\pm0.017$ and $\sigma_8=0.778\pm0.045$. This $n_s$ is only $1.8\sigma$ from unity, compared to the $\sim 2.6\sigma$ \emph{WMAP} 3-year result. Finally, an anomalous feature in the source spectrum at $l<200$ remains, most strongly associated with W-band. | Measuring $n_s$, the spectral index of initial scalar fluctuations, which is scale invariant ($n_s = 1$) in the Harrison-Zeldovich model and slightly shallower in inflation models, is difficult, primarily because experimental systematics require control over a broad range of spatial scales. In inflation, the deviation from unity closely relates to the inflationary potential and the number of $e$-folds of expansion, so a statistically robust measurement of $n_s \neq 1$ places compelling constraints on the physics of the inflationary epoch. Because all-sky measurements of the cosmic microwave background (CMB) access the largest observable scales in the universe, the angular power spectrum of the CMB, with a long lever arm, is crucial to such studies. Indeed, the latest data release from the \textit{Wilkinson Microwave Anisotropy Probe} (\emph{WMAP}) claims $\sim 2.6\sigma$ deviation from the Harrison-Zeldovich spectrum \citep{Spergel2007}. Unfortunately, the CMB is not a totally clean measurement. For example, the well-known degeneracy with the optical depth since recombination ($\tau$) makes precision measurement of $n_s$ impossible using CMB temperature anisotropies alone, and polarization is required to break it. Complicated noise properties and hints of unknown systematics in the \emph{WMAP} measurement of large-scale polarization indicate that the systematic uncertainty in both $\tau$ and $n_s$ should still be considered significant \citep{Eriksen2007b}. Another important, but under-appreciated, complication for the measurement of $n_s$ is additional power in the angular spectrum from unresolved, and unmasked, point sources. At high $l$, this shot noise can significantly bias the power spectrum, and consequently $n_s$. The \emph{WMAP} team devised a sensible prescription for dealing with this contaminant: 1) Use the spectral energy distribution measured from detected sources (and distinct from the CMB) to infer it for undetected ones; 2) measure the contamination using multifrequency data; 3) correct the spectrum; and 4) marginalize over the measurement error when computing the likelihood \citep{Hinshaw2003,Hinshaw2007}. \citet{Huffenberger2004} found a level of source contamination consistent with the level in the first \emph{WMAP} data release \citep{Hinshaw2003}. However, based on the three year temperature data \citep{Hinshaw2007}, \citet{Huffenberger2006} measured a point source spectrum with two irregularities. First, at $l>200$ the spectrum is white, but with an amplitude below the value in the original preprint of \citet{Hinshaw2007}. In the present work, we discovered a small error in the power spectra used for the \citet{Huffenberger2006} estimate, which should have reported $A = 0.013 \pm 0.001$ $\mu$K$^2$ instead of $A = 0.011 \pm 0.001$ $\mu$K$^2$, still below the original WMAP value of $A = 0.017 \pm 0.002$ $\mu$K$^2$. Prompted by our result, \citet{Hinshaw2007} re-examined the issue, revising their value down somewhat and increasing the error bars, to $A = 0.014 \pm 0.003$ $\mu$K$^2$. The \citet{Spergel2007} bispectrum analysis indicates a non-Gaussianity consistent with these values, but lacks the statistical power of the multifrequency power spectrum comparison. The second peculiarity is that the power at $100<l<200$ in \citet{Huffenberger2006} was inconsistent, at strong statistical significance, with the rest of the white spectrum. This paper again considers the power spectrum source correction procedure in detail. We begin in Section \ref{sec:ns_impact} with a study of the impact of the source correction on the scalar spectral index through the likelihood. Following this, we probe the source amplitude in Section \ref{sec:source_spectra}, examining the dependence of the fit on the sky weighting, mask, year of observation, and frequency dependence, and present our best estimates of the cosmological parameters. These same tests probe the robustness of the $l<200$ feature. Finally, we conclude in Section \ref{sec:conclusions}. | \label{sec:conclusions} We have reanalyzed the source correction procedure for the three-year data release of \emph{WMAP}. First, we considered the impact of this procedure in the \emph{WMAP} likelihood code. Surprisingly, we found that the \emph{WMAP} likelihood does not react to changes in the point source correction error. We have devised a modified likelihood which does respond as expected, although we note that more work is needed to completely validate this approach, as it couples to the important problem of how to approximate a non-Gaussian likelihood with fitting formulas over a wide multipole range. To conclude $n_s < 1$, a precision measurement of the source contamination is required. We note that the modes not contaminated favor $n_s$ consistent with unity. Second, we found several indications that the unmasked source population in \emph{WMAP} data is anisotropic. This implies that the combined spectrum should be corrected differently in two multipole regions, based on the weighting of the map. Anisotropy in the unmasked sources is unexpected, but can be turned to an advantage: By very carefully masking near the ecliptic poles and galaxy, or employing a wide galactic cut, the point source contamination can be cut substantially. This gain must be weighed against the reduction of the sky area. We note irregularities in jackknife tests of the source fit, grouped by time of observation or frequency band. This prompted us to adopt large errors on the source fit, to account for systematics beyond the estimate of statistical error which accompanies our measurement. This step is necessary to treat the source correction conservatively. With the modified likelihood and reduced source amplitude from ignoring the excess at $l<200$, these enlarged error bars are responsible for raising our values of $n_s$ and $\sigma_8$. Finally, the previously noted anomalous $l < 200$ feature is still present, shows signs of being spatially associated with the galaxy, and is most strongly associated with the W band. On the other hand, it does not appear to be associated with calibration or beam errors. It may represent an over-subtraction of the foreground template in W, although further investigation is warranted. However, the immediate conclusion is that this part of the spectrum should not be used to infer the point source amplitude at higher $l$'s. | 7 | 10 | 0710.1873 |
0710 | 0710.0514_arXiv.txt | We study the phase-space behaviour of nearby trajectories in integrable potentials. We show that the separation of nearby orbits initially diverges very fast, mimicking a nearly exponential behaviour, while at late times it grows linearly. This initial exponential phase, known as Miller's instability, is commonly found in N-body simulations, and has been attributed to short-term (microscopic) N-body chaos. However we show here analytically that the initial divergence is simply due to the shape of an orbit in phase-space. This result confirms previous suspicions that this transient phenomenon is not related to an instability in the sense of non-integrable behaviour in the dynamics of N-body systems. | The problem of how exactly galaxies reach their final equilibrium configuration is still unsolved. It is clear that, unlike for gases, two-body collisions between stars in galaxies are not the driving mechanism to reach a relaxed state, since the associated timescales are exceedingly large \citep{bt}. In an attempt to explain the road to equilibrium from a statistical mechanics point of view, Lynden-Bell (1967) introduced the concept of ``violent relaxation''. In this context, the relaxation is reached through the effects of a ``violently changing'' gravitational field. However, the detailed physics of this process also remain to be understood \citep{arad,valluri}. Besides the statistical mechanics approach, it is also possible to study the problem of ``relaxation'' at the level of orbits. In this case, it is useful to introduce the concept of mixing, by which we mean how quickly nearby particle trajectories diverge in (phase-)space as a function of time. In the case of time-independent gravitational potentials it is customary to classify mixing into two types. If the particles move in an integrable potential, nearby orbits will diverge as a power-law in time, e.g. \citet{hw}. This process is known as phase-mixing \citep{bt}. However, when the potential admits a certain amount of chaos, there exist regions of phase-space where nearby orbits diverge exponentially, evidencing an extreme sensitivity to small changes in the initial conditions \citep{Arnold}. This process is known as chaotic-mixing \citep{Kandrup-Mahon,Kandrup}. These mixing processes can also take place in a time-dependent gravitational potential, in which case the energies of the particles will not be constant. The degree of ``stickiness'', quantified by the time-evolution of the divergence of nearby orbits would then measure the degree of ergodicity of the mixing process. In the case of chaotic mixing, this could lead to a system that does not have much memory of its evolutionary history. The timescales for evolution could be relatively short, and in principle, this process could be important in the path towards equilibrium for galaxies in the Universe \citep{merritt,vm}. Since the 1970s N-body simulations have become the standard tool for studies of the formation and dynamics of structures in the Universe. The question of whether they are a faithful representation of the Universe has always attracted significant attention. This is especially true in recent years \citep{diemand,binney}, particularly with the finding that dark-matter halos have universal density profiles \citep{nfw,moore,weinberg-a,weinberg-b}. One of the first works to focus on how N-body systems evolve was \citet{miller}. Using what must have been the very first computers in the world, he simulated a self-consistent system in virial equilibrium of 8 upto 32 particles distributed randomly in a cubic volume. Miller found that the trajectories of neighbouring particles initially diverged exponentially. This initial transient has been confirmed using numerical experiments with a significantly larger number of particles \citep{lecar,Kandrup-Smith, vm, hm}, as well as with various degrees of numerical softening \citep{ks01}. This implies that the initial exponential divergence cannot be purely attributed to the very grainy nature of the gravitational potential in Miller's experiments. Furthermore, even in high-resolution N-body realizations of well-behaved integrable systems such as the Plummer sphere, nearby orbits experiment a phase of exponential separation at very early times \citep{kandrup-sideris}. This initial exponential divergence present in N-body simulations is now known as ``Miller's instability''. Understanding this puzzle is the focus of this paper. That N-body systems would show a certain degree of chaoticity is not necessarily unexpected. However, it seems natural to expect that the larger the number of particles used to represent an otherwise integrable smooth gravitational potential, the more faithful the representation, and hence the lesser the degree of ``numerical'' chaos \citep{quinlan-tremaine}. There is now significant evidence that when such a system is represented by a sufficiently large number of particles, it does tend to the behaviour expected from the collisionless Boltzmann equation \citep{goodman,elzant,kandrup-sideris,sideris}. Nevertheless, even in these high-resolution experiments the initial exponential growth phase is present \citep{Kandrup-Smith,vm,ks01}. Furthermore, there is evidence \citep{goodman,hm} that the rate of divergence associated to this phase increases in proportion to the number of particles used. Because Miller's instability only lasts for a very short timescale this does not imply that the system is (macroscopically) chaotic \citep{vm}. As stated by \citet{elzant} it is likely that the ``mechanism leading to the short e-folding time in point particle systems is physically unimportant''. So, while the existence of a continuum limit in N-body systems appears to be more or less established for long timescales, on short timescales Miller's instability remains a puzzle. The physical mechanism responsible for this was hitherto unknown. It seems quite unlikely that collisions between particles could be important on timescales as short as one-tenth of the crossing time of the system, as measured for example by \citet{hm}. Microscopic chaos arising from ``white-noise'' or poor orbit integrations are also unlikely to be important on those timescales, particularly in integrable (well-behaved) potentials. In this paper, we tackle this paradox by studying the {\it initial} behaviour of nearby characteristics in an integrable smooth (and analytic) potential. Our aim is to understand how these nearby characteristics diverge on short timescales, and if they do so at nearly exponential rates. As we shall demonstrate below, this is indeed the case. The initial behaviour mimics an exponential divergence, but since the system is fully integrable this is not related to the presence of chaos. This near-exponential behaviour merely reflects the time evolution of an orbit in phase-space. This result shows that there is no need to introduce the concept of microscopic chaos, and confirms previous suspicions that this transient phenomenon is not related to an instability in the sense of non-integrable behaviour in the dynamics of N-body systems. In this paper we describe the evolution of nearby orbits in phase-space, and in particular in configuration space, expanding upon a model developed by \citet{hw} (hereafter HW). The details of this formalism are given in Sec.~\ref{sec:aa}. In this Section we focus in detail on the behaviour of nearby orbits in a Plummer potential. In Sec.~\ref{sec:comp} we summarize our results. | \label{sec:comp} Our analysis shows that the initial nearly-exponential divergence of nearby orbits in N-body systems is not due to chaos. It is present also in integrable smooth potentials, and it reflects a power-law divergence modulated by the shape of an orbit in phase-space. It is interesting to note that the rates of divergence that we measure using our formalism are in very good agreement with those obtained by \citet{hm} for N-body realizations of the same Plummer sphere. These authors find a characteristic e-folding time of $t_{cr}/20$ for systems with $N \sim 10^5$ particles, which is very comparable to the values obtained in Section \ref{sec:lyap}. They also find a weak dependence on $N$, which may also be readily understood within our framework. Such a dependence is induced by the very rapid decrease in the spatial density of the system. If a ``relatively'' small number of particles is used in a N-body simulation, then the density cannot be mapped properly. For example, to measure a decline in the density of $10^{-5}$ on a timescale of $\sim 3 t_{cr}$ as observed in Fig.~\ref{fig:ex1}, N-body realizations with at least $10^{5}$ nearby particles are needed. Previous works, including \citet{miller} and \citet{kandrup-sideris} have also noted an oscillatory behaviour in the divergence of nearby orbits. Our analysis, as well as Figure \ref{fig:ex1} show that this is due to the modulation produced by the periodicity of a regular orbit in phase-space. It is not, as suggested by \citet{miller}, due to the formation of tight binaries in an N-body system. The fact that such behaviour was visible in the various N-body studies presented in the literature, in fact demonstrates that such N-body systems were faithful representations of the true (integrable) system, at least on short timescales. As stated by \citet{Kandrup} and \citet{vm}, the Lyapunov exponents need to be measured in the limit of infinite time intervals; short-time exponential-like divergences do not imply chaotic behaviour. | 7 | 10 | 0710.0514 |
0710 | 0710.3384_arXiv.txt | We investigate the stellar populations of a sample of 162 Ly$\alpha$ emitting galaxies (LAEs) at $z = 3.1$ in the Extended Chandra Deep Field South, using deep Spitzer IRAC data available from the GOODS and SIMPLE surveys to derive reliable stellar population estimates. We divide the LAEs according to their rest-frame near-IR luminosities into IRAC-detected and IRAC-undetected samples. About 70\% of the LAEs are undetected in 3.6 \micron\ down to $m_{3.6} = 25.2$ AB. Stacking analysis reveals that the average stellar population of the IRAC-undetected sample has an age of $\sim 200$ Myr and a mass of $\sim \pow{3}{8}$ \Msun, consistent with the expectation that LAEs are mostly young and low-mass galaxies. On the other hand, the IRAC-detected LAEs are on average significantly older and more massive, with an average age $\ga 1$ Gyr and mass $\sim 10^{10}$ \Msun. Comparing the IRAC colors and magnitudes of the LAEs to $z \sim 3$ Lyman break galaxies (LBGs) shows that the IRAC-detected LAEs lie at the faint blue end of the LBG color-magnitude distribution, suggesting that IRAC-detected LAEs may be the low mass extension of the LBG population. We also present tentative evidence for a small fraction ($\sim 5\%$) of obscured AGN within the LAE sample. Our results suggest that LAEs posses a wide range of ages and masses. Additionally, the presence of evolved stellar populations inside LAEs suggests that the Ly$\alpha$ luminous phase of galaxies may either be a long-lasting or recurring phenomenon. | \label{Intro} Lyman-alpha emitting galaxies (LAEs) are important tracers of galaxy formation. The Ly$\alpha$ emission is produced by on-going star formation in the galaxies, and the line emission enables discovery of objects that may be too faint to be seen in the continuum. LAEs therefore offer an opportunity to probe the faint end of galaxy formation at high redshift, and may serve as building blocks of larger galaxies in a hierarchical universe. In recent years, there have been great advancements in narrow-band surveys of LAEs. Large samples of candidate LAEs now exist at redshifts $z = 3 - 7$ \citep[e.g.][]{hu04, malho04, tanig05, shima06, dawso07, muray07, ouchi07}. One outstanding question regarding the LAEs is their stellar population, which is important for understanding the physical nature of the LAEs and their connection with other high redshift galaxy populations such as Lyman break galaxies \citep[LBGs;][]{shapl03, ando06, pente07, stanw07}. A difficulty often encountered when studying the stellar population of high-redshift galaxies is the necessity to have observations at or redward of rest-frame optical in order to constrain the total stellar mass. At $z \ga 3$, this means having either deep near-IR or Spitzer IRAC observations. Much progress has been made recently, however, and several studies have suggested that LAEs are small galaxies with masses $\la 10^9 - 10^{10}$ \Msun\ \citep{gawis06b, lai07, finke07, pirzk07, nilss07}. In this study, we focus on a sample of 162 LAEs at $z = 3.1$ discovered in a narrow-band survey of the Extended Chandra Deep Field South (ECDF-S) by \citet{gronw07}. The narrow-band 4990 \AA\ imaging covers an area of 992 arcmin$^2$ and reaches a completeness limit of $\sim \pow{1.5}{-17}$ erg cm$^{-2}$ s$^{-1}$ (equivalent to a Ly$\alpha$ luminosity $L_{{\rm Ly}\alpha} = \pow{1.3}{42}$ erg s$^{-1}$). Spectroscopic results confirm the robustness of the sample (details of the spectroscopy will be presented in \citealt{gawis07} and Lira et al., in preparation). Of the 52 candidates with sufficient S/N in the spectra to yield redshifts, 51 were confirmed to be $z=3.1$ LAEs (the remaining object is a $z=1.60$ AGN with C~III] 1909 \AA\ emission). The most likely contaminants in our sample are $z=0.34$ [O~II] emitters. However, there should be a negligible number of these objects in our sample because by requiring the observer's frame equivalent width to be $> 80$ \AA, we eliminate all but the strongest (and rarest) [O~II] emitters \citep{gronw07}. The narrow-band imaging in ECDF-S is supplemented by optical and near-IR observations from MUSYC \citep{gawis06a}, GOODS \citep{dicki03}, and GEMS \citep{rix04} surveys, and Spitzer observations from GOODS and SIMPLE\footnotemark\ (Damen et al., in preparation). Using the multi-wavelength data available in ECDF-S, we study the rest-frame UV to near-IR properties of this large sample of LAEs. We place special emphasis on the new Spitzer IRAC observations, which sample the rest-frame $0.9 - 2$ \micron\ emission from the $z = 3.1$ LAEs and provide valuable constraints on their stellar population and star formation history. \footnotetext{http://www.astro.yale.edu/dokkum/SIMPLE/} We assume a ${\rm \Lambda}$CDM cosmology with $\Omega_{\rm M} = 0.3$, $\Omega_{\rm \Lambda} = 0.7$, and $h = 0.7$. All magnitudes are in the AB system. | \label{Disc} There are several previous studies on the stellar population of LAEs at $z \sim 3$ and beyond \citep{gawis06b, finke07, pirzk07, nilss07}. These previous studies found that LAEs have masses ranging from $10^6 - 10^9$ \Msun, and ages ranging from $0.005 - 1$ Gyr. Our best-fit average age of 160 Myr and mass of \pow{3}{8} \Msun\ for the IRAC-undetected sample are broadly consistent with these previous studies. In a further analysis of the IRAC-undetected sample presented in this paper, \citet[submitted]{gawis07} fit two-component models and find that the IRAC-undetected LAEs have a total mass of \pow{1}{9} \Msun, with a young stellar component (age $\sim 20$ Myr) accounting for $\sim 20\%$ of the total mass. These results are consistent with ours which are derived using single-component models and should be interpreted as the average properties of the entire stellar population within the LAEs. \begin{figure} \includegraphics[width=\columnwidth]{f4.eps} \caption{ A comparison of the $R-m_{3.6}$ colors and 3.6 \micron\ magnitudes between 8 \micron\ detected LBGs (green square), regular LBGs (black dots), and the IRAC-detected LAEs (red stars). The solid line shows the $R<25.5$ selection limit for LBGs. The data for the regular LBGs come from IRAC observations of the \citet{steid03} LBG sample (Magdis et al., in preparation), and the 8 \micron\ LBG data come from combining the samples of Magdis et al.\ and \citet{rigop06}. The mean colors and magnitudes of each sample are plotted as large filled symbols, with error bars showing the dispersion within the samples. We additionally plot the stacked photometry of the IRAC-undetected sample, shown as the red filled red star with the faintest 3.6 \micron\ magnitude. \label{compLBG} } \end{figure} The IRAC-detected LAEs are more luminous than the IRAC-undetected LAEs in both rest-frame UV and near-IR, and represent the massive end of the LAE mass spectrum. Potentially, the IRAC-detected LAEs may provide a link between the LAEs and other galaxy populations. For instance, they may be the $z \sim 3$ analogues to the similarly massive IRAC-detected LAEs found at $z \sim 5.7$ \citep{lai07}. In \Fig{compLBG}, we compare the $R-m_{3.6}$ colors and 3.6 \micron\ magnitudes of the LAEs to 8 \micron\ detected LBGs and regular LBGs at $z \sim 3$. The $R-m_{3.6}$ color and 3.6 \micron\ magnitude correlate with stellar mass \citep{rigop06}, with the 8 \micron\ LBGs being the most massive ($\ga 10^{11}$ \Msun), followed by LBGs ($\sim 10^{10} - 10^{11}$ \Msun), IRAC-detected LAEs ($\sim 10^{10}$ \Msun), and finally IRAC-undetected LAEs ($\sim 10^8$ \Msun). The IRAC-detected LAEs occupy the faint and blue end of the LBG color-magnitude distribution, suggesting that they may be the lower mass extension of the LBG population. This interpretation is supported by the fact that the inferred mass of the IRAC-detected LAEs ($\sim 10^{10}$ \Msun) is at the low end of the mass range found for LBGs ($\sim 10^{10} - 10^{11}$ \Msun; \citealt{papov01, shapl01}). Furthermore, most LAEs in the present sample have (observer frame) optical colors that would allow them to be selected as LBGs, although the LAEs tend to be fainter in the rest-frame UV \citep{gronw07, gawis06b}. It has also been observed that $\sim 20\% - 25\%$ of LBGs at $z \sim 3$ exhibit Ly$\alpha$ emission strong enough to be narrow band excess objects \citep{steid00, shapl03}. There is thus accumulating evidence suggesting that the IRAC-detected LAEs may be the bridge connecting the LAE and LBG populations. Using the stellar mass estimates derived in the previous section and assuming a survey volume of \pow{1.1}{5} Mpc$^3$ \citep{gronw07}, we find stellar mass densities of $0.3 \pm 0.3 \times 10^6$ and $3 \pm 1 \times 10^6$ \Msun\ Mpc$^{-3}$ for the IRAC-undetected and IRAC-detected populations, respectively. The errors in the stellar mass densities include uncertainties in the mass from the fits and uncertainties in the number density coming from Poisson fluctuations and a $\sim 20\%$ cosmic variance given the observed LAE bias of $\sim 1.7$ \citep[submitted]{gawis07}. We should stress that the above stellar mass densities only account for LAEs with Ly$\alpha$ luminosities above our survey completeness limit of \pow{1.3}{42} erg s$^{-1}$. The IRAC-undetected LAEs account for about $9 \pm 10\%$ of the total stellar mass in LAEs, even though they make up 2/3 of the total by number. Compared to LBGs and DRGs at $z \sim 3$, which have stellar mass densities of $\sim \pow{1}{7}$ and \pow{7}{6} \Msun\ Mpc$^{-3}$ respectively \citep{grazi07}, the stellar mass contained in LAEs is smaller, but not insignificant. However, it is important to keep in mind that there may be substantial overlap between the LAEs and LBGs. A better understanding of the overlap between these two populations is necessary before a direct comparison of the stellar mass densities can be made. The present results show that the LAEs posses a wide range of masses and ages, from the massive and evolved IRAC-detected LAEs to the young and small IRAC-undetected LAEs. The range of photometric properties shown in the LAE sample suggests that LAEs exhibit a continuum of properties between these two extremes. The presence of both young and evolved stellar populations within the overall LAE population implies that the Ly$\alpha$ luminous phase of galaxies may last $\ga 1$ Gyr, or that the Ly$\alpha$ luminous phase is recurring. Interestingly, \citet{shapl01} found that LBGs with best-fit stellar population ages $\ga 1$ Gyr also show strong Ly$\alpha$ emission in their spectra. The authors suggest that vigorous past star formation has destroyed and/or expelled the dust inside the galaxies, allowing the Ly$\alpha$ photons to escape. One characteristic of these evolved Ly$\alpha$ emitting LBGs is that they tend to have more quiescent SFR than the rest of the population. The IRAC-detected LAEs, with their evolved stellar population and low specific SFR, are consistent with this scenario. If the LAE phase is recurring, then the young age of the IRAC-undetected LAEs implies that their number and stellar mass densities can be as much as a factor $\sim 10$ higher (very roughly the ratio of the age of the universe to the stellar age). Similarly young stellar populations ($\la 100$ Myr) have also been found at higher redshifts \citep{finke07, pirzk07, verma07}. The stochastic nature of galaxies with short lifetimes suggests that there may be a related population of undetected pre or post-starburst galaxies that may contribute significantly to the stellar mass and star formation rate densities at $z \ga 3$. | 7 | 10 | 0710.3384 |
0710 | 0710.1730_arXiv.txt | The evolution of stellar collision products in cluster simulations has usually been modelled using simplified prescriptions. Such prescriptions either replace the collision product with an (evolved) main sequence star, or assume that the collision product was completely mixed during the collision. It is known from hydrodynamical simulations of stellar collisions that collision products are not completely mixed, however. We have calculated the evolution of stellar collision products and find that they are brighter than normal main sequence stars of the same mass, but not as blue as models that assume that the collision product was fully mixed during the collision. | The aim of the MODEST collaboration \citep{article:modest1} is to model and understand dense stellar systems, which requires a good understanding of what happens when two single stars or binary systems undergo a close encounter. A possible outcome of such an encounter is a collision followed by the merging of two or more stars. This is a possible formation channel for blue straggler stars (\emph{e.g.} \citet{article:sills_on_axis}). Understanding the formation and evolution of blue stragglers is important for understanding the Hertzsprung-Russell diagram of clusters. | \begin{figure} \ifpdf \includegraphics[width=\textwidth]{glebbeek_fig_hrd} \else \includegraphics[angle=270,width=\textwidth]{glebbeek_fig_hrd} \fi \caption{Colour-magnitude diagram of the open cluster M67 ($\blacklozenge$). Overplotted are the locations of our models at $4 \mathrm{Gyr}$, the age of M67. The black ({\Large$\bullet$, $\blacktriangle$}) symbols are collisions from the M67 simulation. Two of these are double collisions, which are indicated by {\Large$\blacktriangle$}. The grey ({\Large$\bullet$}) symbols are from our larger grid.} \label{fig:hrd_m67} \end{figure} Compared to normal stars, collision products are helium enhanced. Most of the helium enhancement is in the interior and does not affect the opacity of the envelope. The helium enhancement does increase the mean molecular weight and therefore the luminosity of the star. This decreases the remaining lifetime of collision products compared to normal stars of the same mass and can be important for the predicted number of blue stragglers from cluster simulations. The increased luminosity changes the distribution of blue stragglers in the colour-magnitude diagram, moving it above the extension of the main sequence. The evolution track of a fully mixed model can be significantly bluer than a self-consistently calculated evolution track of a merger remnant. Fully mixed models are closer to the zero age main sequence. Our grid of models covers most of the observed blue straggler region of M67 (Figure \ref{fig:hrd_m67}). A better coverage of the blue part of the region can be achieved by increasing the upper mass limit in the grid. The brightest observed blue straggler falls outside the region of our grid because it requires at least a double collision to explain. | 7 | 10 | 0710.1730 |
0710 | 0710.1506_arXiv.txt | We have modeled the emission of \hdo\ rotational lines from the extreme C-rich star IRC+10216. Our treatment of the excitation of \hdo\ emissions takes into account the excitation of \hdo\ both through collisions, and through the pumping of the $\nu_2$ and $\nu_3$ vibrational states by dust emission and subsequent decay to the ground state. Regardless of the spatial distribution of the water molecules, the \hdo\ $1_{10}-1_{01}$ line at 557 GHz observed by the {\em Submillimeter Wave Astronomy Satellite} (SWAS) is found to be pumped primarily through the absorption of dust-emitted photons at 6 $\mu$m in the $\nu_2$ band. As noted by previous authors, the inclusion of radiative pumping lowers the ortho-\hdo\ abundance required to account for the 557 GHz emission, which is found to be $(0.5-1)\times10^{-7}$ if the presence of \hdo\ is a consequence of vaporization of orbiting comets or Fischer-Tropsch catalysis. Predictions for other submillimeter \hdo\ lines that can be observed by the {\em Herschel Space Observatory} (HSO) are reported. Multitransition HSO observations promise to reveal the spatial distribution of the circumstellar water vapor, discriminating among the several hypotheses that have been proposed for the origin of the \hdo\ vapor in the envelope of IRC+10216. We also show that, for observations with HSO, the \hdo\ $1_{10}-1_{01}$ 557 GHz line affords the greatest sensitivity in searching for \hdo\ in other C-rich AGB stars. | \label{sec:intro} The discovery of water vapor emission in the extreme C-rich AGB star IRC+10216 with SWAS \citep{mel01}, its confirmation by ODIN \citep{has06}, and the subsequent detection of other O-bearing molecules like OH \citep{for03}, H$_2$CO \citep{for04}, and C$_3$O \citep{ten06}, have challenged our current understanding of the chemistry in envelopes around C-rich AGB stars. According to standard models, essentially all oxygen nuclei are predicted to be locked into CO or SiO with no reservoir to form other O-bearing molecules (except for low abundances of species such as HCO$^+$ in the outer envelope where photochemistry is important). Therefore, the unexpectedly high abundances found for H$_2$O, OH, and H$_2$CO, indicate that several processes not included in standard models for C-rich environments are driving the oxygen chemistry, but the dominant water production mechanism is still a source of debate. Four distinct mechanisms have been considered as possible sources of the observed water vapor in IRC+10216, each one making a specific prediction for the \hdo\ spatial distribution in the envelope: $(i)$ chemistry in the inner envelope, which would imply the presence of \hdo\ in the warmest and densest regions; $(ii)$ vaporization of icy orbiting bodies that have survived from the main sequence into the C-rich AGB phase \citep{mel01,for01}, predicting the release of \hdo\ at intermediate radii of $R_{int}=(1-5)\times10^{15}$ cm; $(iii)$ grain surface reactions, such as the Fischer-Tropsch catalysis on the surfaces of small metallic grains \citep{wil04}, which predicts \hdo\ to attain a nearly uniform abundance at radii larger than $R_{int}=(1.5-2)\times10^{15}$ cm; $(iv)$ chemistry involving photodissociation products in the outermost regions of the envelope: a specific mechanism relying on the radiative association O+H$_2$$\rightarrow$\hdo+$\gamma$ has been proposed by Ag\'undez \& Cernicharo (2006, hereafter AC06), and predicts \hdo\ to be present at radii higher than $R_{int}\approx4\times10^{16}$ cm. In all cases, \hdo\ is expected to have a uniform abundance from $R_{int}$ up to the external region where it is photodissociated producing OH. The only water line detected so far in IRC+10216, the ortho-\hdo\ \t110101\ transition at 557 GHz, is the ground-state transition, with the upper level at only 27 K above the ground rotational level, and cannot discriminate -by itself- between the processes listed above. Nevertheless, the launch of the Herschel Space Observatory (HSO) will allow us to observe other \hdo\ lines in the submillimeter and far-infrared domains, permitting us to infer the region where the \hdo\ emission arises, and will potentially favor one of the proposed formation mechanisms. The Heterodyne Instrument for the Far Infrared (HIFI) onboard HSO will provide very high spectral resolution observations (0.14-1.0 MHz), thus permitting the lines to be velocity resolved. In this paper, we model the \hdo\ emission from IRC+10216 to show how the \hdo\ spatial distribution can be inferred from HSO multi-transition observations. Also, we explore which \hdo\ transition provides the most sensitive means of searching for water vapor around AGB stars other than IRC+10216; such a search would determine whether the occurence of \hdo\ in C-rich environments is widespread. | \label{sec:discussion} \subsection{Model uncertainties} Our models for IRC+10216 predict that, whatever the region where \hdo\ is formed or released to the outflow, \hdo\ is radiatively excited, which implies that model results are independent of the gas temperature and density profiles. This result relies on the extrapolation to higher $T_k$ of the collisional rates given by \cite{phi96}, which will require further confirmation from new estimates of \hdo-H$_2$ collisional rates at higher temperatures. Nevertheless, we find that collisional rates would have to be about one order of magnitude higher than estimated to compete efficiently with the radiative rates in the innermost regions of the envelope, which seems somewhat implausible. A relatively uncertain parameter in our models is the assumed distance to IRC+10216, $D=170$ pc. It has been proposed that the distance may be substantially smaller, $D=100-150$ pc \citep*{zuc86,gro92}. At 170 pc, the inferred stellar luminosity is $L_*\approx2\times10^4$ \Lsun, which results in $L_*\approx10^4$ \Lsun\ at $D=120$ pc, a value more similar to the inferred luminosities of other C-rich AGB stars with similar \Mdot\ \citep*{sch06}. If $D=120$ pc, both \Mdot\ and the radiation density at 6 $\mu$m are a factor of 2 lower than assumed, so that the radiative-to-collisional pumping rate ratio still remains unchanged. Since the closer proximity and weaker 6 $\mu$m radiation density have opposite effects on the \hdo\ outflow rate required to match the SWAS 557 GHz flux, the latter will decrease by less than a factor 2, and thus the expected \hdo\ abundance will be a factor of $<2$ higher than in Table~\ref{tab:h2omodels}. Therefore, we conservatively estimate an o-\hdo\ abundance in the range $(0.5-1)\times10^{-7}$ for $R_{int}=2\times10^{15}$ cm. Concerning the line ratios to be observed by HSO, we expect values similar to those obtained at $D=170$ pc if the line emission remains unresolved, i.e. for low values of $R_{int}$. For high values of $R_{int}$, beam effects will be more important at $D=120$ pc, diminishing the low-excitation line fluxes relative to the 557 GHz line. A potentially more important source of uncertainty is the assumption of spherical symmetry in our models. Both the molecular and dust emission from IRC+10216 show evidence for departures from a smooth distribution in a spherically symmetric outflow; incomplete, discrete shells or arcs and clumpy structures are instead observed on a wide range of distances to the star \citep*[e.g.][]{luc99,mau99,fon03,lea06}. Even more important, the OH line shapes observed in IRC+10216 strongly suggest an asymmetric distribution of OH in the outer regions of the envelope \citep{for03}, thus also suggesting an asymmetric distribution of the parent \hdo\ molecule. These asymmetries may alter to some extent the \hdo\ line flux ratios calculated with the use of our spherically symmetric approach. While in spherical symmetry the emission from any line is isotropic, in filamentary structures or slabs of velocity-coherent gas the optically thick lines radiate preferentially in the direction perpendicular to the two faces of the sheet \citep*{eli89}, whereas the emission from thinner lines will approach a more isotropic behavior. Future HSO observations will show the importance of the observed morphological complexity on the line fluxes by showing whether the different \hdo\ line flux ratios are consistent with a single value of $R_{int}$, or indicate a range of values. Finally, the models assume a water shell with uniform \hdo\ abundance and sharp inner and outer radii; however, a finite abundance gradient obviously takes place at both the \hdo\ formation and dissociation regions, and variations of the \hdo\ abundance across the shell are also possible. These effects may also alter to some extent the expected line flux ratios. Uncertainties in the outer radius of the \hdo\ shell may also affect the expected fluxes from stars with low mass loss rates. \subsection{Water formation at inner or intermediate radii} Both the cometary and Fischer-Tropsch catalysis hypothesis predict \hdo\ to be released or formed at intermediate radii of {\it a few} $\times10^{15}$ cm, and although no specific mechanism has been proposed for \hdo\ formation at the innermost regions ($R_{int}<10^{15}$ cm), this possibility cannot be rejected. The models shown in section~\ref{sec:predictions} will permit us to discriminate, from HSO observations of IRC+10216, if there are significant amounts of \hdo\ in the innermost regions through the observation of mid-excitation \hdo\ transitions. More difficult will be a priori to discriminate between the cometary and Fischer-Tropsch catalysis propositions. Both predict similar values for $R_{int}$; nevertheless, if $R_{int}$ were found to be significantly higher than $2\times10^{15}$ cm, the release of \hdo\ from comets could be favoured, unless some additional mechanism were found to shift $R_{int}$, within the Fischer-Tropsch catalysis framework \citep{wil04}, outwards in the envelope. The search for water emission at 557 GHz in C-rich AGB stars other than IRC+10216 will also favour one of the two hypotheses: since IRC+10216 is at the high end of mass loss rates from C-rich stars, and the efficiency of Fischer-Tropsch catalysis to form \hdo\ molecules is expected to decrease with diminishing \Mdot, one would expect in such a case a rate of detection significantly lower than that quantified in section~\ref{sec:crich} for the cometary hypothesis, and one would expect a pronounced decline of the \hdo\ abundance with diminishing \Mdot. The ISO/LWS spectrum of IRC+10216 \citep{cer96} shows an emission feature at 179.5 $\mu$m, coincident with the wavelength of the \t212101\ o-\hdo\ transition, with a flux of $\approx8\times10^{-20}$ W cm$^{-2}$. This flux is very similar to that computed for the quoted line in model $B$ ($R_{int}=2.1\times10^{15}$ cm). However, the far-infrared spectrum of IRC+10216 shows vibrationally excited rotational emission of HCN, and the combined emission of the $\nu_1=1$ $J=19\rightarrow18$ and $\nu_3=1$ $J=19\rightarrow18$ HCN lines, both emitting at 179.5 $\mu$m, is expected to be also comparable to the measured line flux at 179.5 $\mu$m \citep{cer96}. Given the uncertainties inherent to the HCN model in \cite{cer96}, where the excited vibrational states are assumed to be thermalized, and given the high density of spectral lines in the IRC+10216 far-infrared spectrum, which makes it difficult to establish the contribution from the $\nu_1=1$ and $\nu_3=1$ rotational lines to the spectrum, the relative contribution from HCN and \hdo\ to the observed spectral feature is quite uncertain. On the other hand, the expected flux of the \t303212\ line at 174.6 $\mu$m in model $B$ is $\approx4\times10^{-20}$ W cm$^{-2}$, below the 3-$\sigma$ upper limit derived for that line from the ISO/LWS spectrum. These considerations suggest weakly that \hdo\ in IRC+10216 is formed or released at radial distances $R_{int}\gtrsim2\times10^{15}$ cm. Our models show that the required \hdo\ abundance for $R_{int}$ in the range $(2-5)\times10^{15}$ cm is $(0.5-1)\times10^{-7}$ relative to H$_2$, which makes the requirements previously demanded for the cometary and Fischer-Tropsch catalysis hypothesis to work in IRC+10216 less restrictive \citep{for01,wil04}. The water outflow rate is $(0.5-1)\times10^{-5}$ M$_{\oplus}$ yr$^{-1}$, which yields, within the framework of vaporization of icy bodies, a required total initial ice mass of $(0.5-10)$ M$_{\oplus}$ for \Mdot(\hdo)/$M_0({\rm ice})$ in the range $10^{-5}-10^{-6}$ yr$^{-1}$ \citep{for01}. On the other hand, the Fischer-Tropsch catalysis on metallic grains will require a density of iron grains relative to total H nuclei of $n_g({\rm Fe})/n_{{\rm H}}=(0.5-1)\times10^{-13}$ to explain the observed \hdo\ 557 GHz emission \citep{wil04}. \subsection{Water formation in the outermost layers} \label{sec:radassoc} Multitransition HSO observations of \hdo\ in IRC+10216 will easily establish whether or not water is formed in the external layers of the envelope; the predicted fluxes in Fig.~\ref{fig:flujosrint2} and the HSO-HIFI sensitivities in Table~\ref{tab:h2otrans} indicate that only the \t110101\ o-\hdo\ line, and possibly the \t111000\ and \t202111\ p-\hdo\ lines, are detectable in one hour of observing time for $R_{int}\gtrsim4\times10^{16}$ cm. For high $R_{int}$, the expected flux in the 557 GHz line can be obtained analytically. Since essentially all o-\hdo\ molecules are in the ground $1_{01}$ level (Fig.~\ref{fig:pump557ghz}a), the o-\hdo\ \t110101\ line flux is derived from the radiative pumping rate given in eq.~(\ref{gammar}) after multiplying $\Gamma_r$ by 1.35 to account for the two radiative pumping routes that are relevant at high distances from the star (Fig.~\ref{fig:pump557ghz}b): \begin{eqnarray} F(1_{10}\rightarrow1_{01}) = 1.4\times10^{-21} \times \left( \frac{4\times10^{17} \, {\rm cm}}{R_{out}} \right) \nonumber \\ \times \left( \frac{R_{out}}{R_{int}} -1 \right) \times \left( \frac{X({\rm o-H_2O})}{5\times10^{-7}} \right) \,\, {\rm \frac{W}{cm^{2}}}, \label{fluxanal} \end{eqnarray} where $X$ is the abundance relative to H$_2$. Equation~(\ref{fluxanal}) is independent of the assumed distance $D$ to the star because, in order to match the observed mid-IR continuum, $\Gamma_r\propto D^{2}$ (eq.~\ref{eqj}); it assumes a water shell with sharp edges at $R_{int}$ and $R_{out}$ and constant \hdo\ abundance; it ignores slight opacity effects in the ro-vibrational lines (section~\ref{sec:110101}) as well as beam effects (both slightly raise the required abundance), and overestimates by less than 20\% the 557 GHz line flux obtained in model $C$. In the limit $R_{out}\rightarrow\infty$, and using $F(1_{10}\rightarrow1_{01})=10^{-20}$ W cm$^{-2}$ \citep{mel01}, eq.~(\ref{fluxanal}) gives \begin{equation} \chi({\rm H_2O}) \gtrapprox 2.4\times10^{-7} \times \left( \frac{R_{int}}{4\times10^{16} \, {\rm cm}} \right), \label{fluxanallim} \end{equation} where $\chi({\rm H_2O})$ is here the \hdo\ (ortho+para) abundance relative to H nuclei. Equation~\ref{fluxanallim} gives the sharp-inner edge, lower limit for the \hdo\ abundance required to account for the observed 557 GHz line flux, as a function of $R_{int}$. AC06 have reported an abundance of $\chi({\rm H_2O})\sim10^{-7}$ to account for the observed 557 GHz line flux in IRC+10216. However, the quoted abundance would imply an inner radius of $R_{int}\lessapprox2\times10^{16}$ cm, but the \hdo\ abundance shown by AC06 (their Fig. 7) decreases sharply at $r<4\times10^{16}$ cm. Based on detailed modelling, we indicate that the $\chi({\rm H_2O})$ profile given by AC06 has to be shifted up by a factor of $\approx2.5$ to account more accurately for the 557 GHz line flux measured by SWAS. In the model proposed by AC06, atomic oxygen is produced in a shell by the photodissociation of CO (and particularly the $^{13}$CO isotopologue, which shields itself far less effectively than $^{12}$CO.) Since the temperature is low ($\sim 10$~K) within the shell where the atomic oxygen abundance is significant, the neutral-neutral reaction sequence $$\rm O + H_2 \rightarrow OH + H$$ $$\rm OH + H_2 \rightarrow H_2O + H$$ is very slow --and therefore is negligible as a source of H$_2$O-- the first reaction being endothermic and the second --although exothermic-- possessing a substantial activation energy barrier. AC06 therefore proposed the radiative association reaction $$\rm O + H_2 \rightarrow H_2O + \gamma$$ as an alternative source of H$_2$O. To match the SWAS- and ODIN-observed water line fluxes, AC06 had to posit that this reaction is relatively rapid at low temperature, with a rate coefficient $k_{ra} \rm \sim 10^{-15}\,cm^3\,s^{-1}$. According to our estimation above, $k_{ra}$ has to be a factor $\approx2.5$ higher than this value to account for the measured \t110101\ \hdo\ flux. Nevertheless, and even with the use of the $k_{ra}$ estimation given by AC06, we find that this large reaction rate coefficient for the radiative association of O and H$_2$ is inconsistent with observations of H$_2$O and OH in at least one translucent molecular cloud, for which sufficient data exist to disipate any significant ambiguity: observations of the cloud along the sight-line to HD 154368 --carried out by \cite{spa98} with the use of the Goddard High Resolution Spectrograph (GHRS) on the {\it Hubble Space Telescope} (HST) -- yield a 3~$\sigma$ upper limit on the water abundance that lies almost two orders of magnitude below the value that would obtain were $k_{ra}$ as large as $\rm 10^{-15}\,cm^3\,s^{\rm -1}$. The factor of discrepancy would rise above $150$ for $k_{ra} \rm \approx 2.5 \times 10^{-15}\,cm^3\,s^{-1}$. The best-fit model for the HD 154368 cloud obtained by \cite{spa98} posits a plane parallel cloud of total visual extinction $A_V = 2.65$~mag in which the density of H nuclei is $n_{\rm H} = 325\, \rm cm^{-3}$ \citep[this is probably a lower limit at the cloud center, as the density inferred from the CO $J=1\rightarrow0/3\rightarrow2$ ratio is $\sim10^3$ cm$^{-3}$; see][]{van91} and the external ultraviolet radiation field is 3 times mean interstellar value given by \cite{dra78}: $I_{UV} = 3$. Under these conditions, the destruction of H$_2$O is dominated by photodissociation at a rate $\zeta = 5.9 \times 10^{-10}{\rm s}^{-1} \times I_{UV} \times (\exp\{-1.7 A_{V1}\} + \exp\{-1.7 A_{V2}\})/2$ \citep*{let00}\footnote{We note that the factor of 1.7 in the exponential, widely used in the literature, yields \hdo\ photodissociation rates at the midplane of the plane-parallel cloud that are higher than the values reported by \cite{rob91} by factors of 2.2 and 10 for $A_V^{tot}=1$ mag and $A_V^{tot}=10$ mag, respectively; therefore, our estimation for the $N({\rm H_2O})/N({\rm O})$ ratio predicted by the radiative association of O and H$_2$ is probably a conservative lower limit.}, where $A_{V1}$ is the visual extinction to one cloud surface and $A_{V2} = 2.65 - A_{V1}$ is the extinction to the other. At the cloud center, the water photodissociation rate is $1.9 \times 10^{-10}\,\,{\rm s}^{-1}$. For a radiative association rate of $k_{ra}$, the ratio of water vapor to atomic oxygen is therefore given by the expression $$n({\rm H_2O})/n({\rm O}) = k_{ra} n_{\rm H} f_{\rm H_2}/ \zeta,$$ where $f_{\rm H_2} \equiv n({\rm H_2}) / n_{\rm H}$. At the cloud center, the cloud is almost fully molecular, with $f_{\rm H_2} \sim 0.5$, and the above equation yields $n({\rm H_2O})/n({\rm O}) = 8.7 \times 10^{-4} \,(k_{ra}/\rm 10^{-15}\,cm^3\,s^{\rm -1})$. In Fig.~\ref{fig:nh2ono}, we show the predicted $n({\rm H_2}) / n_{\rm H}$ ratio for the best-fit \cite{spa98} model, together with the $n({\rm H_2O})/n({\rm O})$ ratio that would result if $k_{ra}$ were $\rm 10^{-15}\,cm^3\,s^{-1}$. The model predicts hydrogen to be predominantly in molecular form, in agreement with results by \cite{sno96}. Averaging the $n({\rm H_2O})/n({\rm O})$ ratio over the entire sight-line, we obtain a column density ratio $N({\rm H_2O})/N({\rm O}) = 5.3 \times 10^{-4} \,(k_{ra}/\rm 10^{-15}\,cm^3\,s^{\rm -1})$. The atomic oxygen column density along the HD 154368 sight-line is $N({\rm O}) = 1.2 \times 10^{18}\,\,{\rm cm^{-2}}$ (Snow et al. 1996, from absorption line observations of the 1355 \AA\ OI] line). Based on a search for the $C^1B_1 - X^1A_1$ band of water vapor near 1240 \AA, \cite{spa98} obtained 3~$\sigma$ upper limit on the water column density of $9 \times 10^{12}\,\,{\rm cm^{-2}}$, corresponding to $N({\rm H_2O})/N({\rm O}) = 7.5 \times 10^{-6}$. This 3~$\sigma$ upper limit lies a factor 70 below the value that we would obtain were $k_{ra}$ equal to $10^{-15}\rm \,cm^3\,s^{-1}$ as AC06 suggested, and places a 3~$\sigma$ upper limit of $1.4 \times 10^{-17}\rm \,cm^3\,s^{-1}$ on $k_{ra}$. AC06 suggested that the freeze out of oxygen onto grain mantles could diminish the water vapor abundance in molecular clouds, as proposed by \cite{ber00} to explain the low H$_2$O abundances measured by SWAS in dense clouds. Ice absorption line observations of diffuse/translucent sight-lines, however, indicate that water ice is generally present only in clouds of $A_V \ge 3$ \citep{whi01}. Furthermore, in the specific case of HD 154368 under present consideration, the atomic oxygen is known from direct measurement to be $1.2 \times 10^{18}\,\,{\rm cm^{-2}}$. The column density of H nuclei along this sight-line, $N_{\rm H} = N({\rm H}) + 2 N({\rm H}_2)$, has been measured to be $4.2 \times 10^{21}\,\,\rm cm^{-2}$ \citep{sno96}, so the mean line-of-sight oxygen abundance is $n({\rm O})/n_{{\rm H}} \sim 3 \times 10^{-4}$, a value that is entirely consistent with the abundances measured along diffuse sight-lines \citep{mey98} and inconsistent with a significant depletion of oxygen onto ice mantles. In summary, the upper limit on the water vapor abundance observed towards HD 154368 definitively rules out a rate coefficient for the radiative association reaction $\rm O + H_2 \rightarrow H_2O + \gamma$ that is large enough to explain --in the context of the AC06 model-- the H$_2$O \t110101\ line strength measured by SWAS toward IRC+10216. If HSO observations would indicate that \hdo\ is formed in the external layers of IRC+10216, an explanation other than the radiative association proposed by AC06 would be required to avoid incompatibilities with observations toward HD 154368. | 7 | 10 | 0710.1506 |
0710 | 0710.4786_arXiv.txt | The proper motion measurements for 143 previously known L and T dwarfs are presented. From this sample we identify and discuss 8 high velocity L dwarfs. We also find 4 new wide common proper motion binaries/multiple systems. Using the moving cluster methods we have also identified a number of L dwarfs that may be members of the Ursa Major (age $\approx$ 400 Myr), the Hyades (age $\approx$ 625 Myr) and the Pleiades (age $\approx$ 125 Myr) moving groups. | Brown dwarfs may be thought of as failed stars. These low mass ($\leq$70 M$_{\rm Jup}$ \citet{burrows01}), cool objects are the lowest mass objects that the star formation process can produce. The majority of the brown dwarfs that have been discovered to date are field objects discovered using surveys such as the Two Micron All Sky Survey (2MASS; \citet{skrutskie06}, see \citet{leggett02} for examples), the DEep Near-Infrared Sky survey (DENIS; \citet{denis05}, see \citet{delfosse99} for examples), the Sloan Digital Sky Survey (SDSS;\citet{york00} see \citet{hawley02} for examples) and the UKIRT Deep Infrared Sky Survey (UKIDSS; \citet{lawrence06}, see \citet{kendall07} for examples). However, to study brown dwarfs in depth, a knowledge of their age is essential, which means we must study brown dwarfs in open star clusters or moving groups. Once a brown dwarf has been proved to belong to an open star cluster, or a moving group, then the age of the dwarf is known, allowing meaningful comparisons to evolutionary models to be made. The most recent example of this is the study done by \citet{bannister07} who used existing proper motions and parallax measurements to show that a selection of field dwarfs in fact belong to the Ursa Major and Hyades moving groups. The importance of this study, is that these are the first brown dwarfs to be associated with an older cluster or group. Older clusters such as the Hyades are expected to contain very few or no brown dwarfs or low mass members, due to the dynamical evolution of the cluster over time \citep{adams02}. However, these escaped low mass objects may remain members of the much larger moving group that surrounds the cluster. To continue the study started by \citet{bannister07}, proper motions need to be measured for the majority of the field brown dwarfs currently known. This has been accomplished using the wide field camera (WFCAM, \citet{casali07}) of the United Kingdom Infrared Telescope (UKIRT). Using these WFCAM images and existing data we have measured proper motions for 143 L and T dwarfs listed in the online Dwarf Archive\footnote{See http://spider.ipac.caltech.edu/staff/davy/ARCHIVE/, a webpage dedicated to L and T dwarfs maintained by C.\ Gelino, D.\ Kirkpatrick, and A.\ Burgasser.}. This proper motion data may be put to a number of uses. Taken with measured radial velocities and distances, it can yield all three components of velocity (U,V,W). Using reduced proper motion diagrams it can be used as an approximate measure of distance. However we have no radial velocities for these objects. These proper motion measurements can however also be used to help identify objects as members of a star cluster or members of a moving group via the moving cluster method. Our proper motion data is discussed and listed in section 2 of this paper. From the proper motion measurements, we find 5 new wide common proper motion binaries/multiple systems. We also identify 8 high velocity L dwarfs, which are discussed in section 4. We suggest that these L dwarfs are probably old and belong to the thick disc or halo population of the galaxy. This in turn suggests that they are likely to be very faint stars rather than brown dwarfs. Finally in section 5 we identify a number of L and T dwarfs that may be members of the Hyades, Pleiades and Ursa Major moving groups . | We have measured the proper motions for 143 dwarfs from the Dwarf Archive. From these measurements, we find 4 new common proper motion wide binary or multiple systems. We also identify 8 high velocity dwarfs i.e. dwarfs with tangential velocities $\geq$ 100 kms$^{-1}$. These dwarfs also have bluer than average \textit{J}-\textit{K} colours. We argue that these are probably thick disc objects with an age of order 10 Gyr. We estimate their luminosities which are $\approx$10$^{-4}$L/L$_{\odot}$. This suggests that they are probably very low luminosity stars rather than brown dwarfs. If so, they may be some of the dimmest stars found to date. Finally we have found 15 L dwarfs that are potential members of the Hyades moving group, 5 that are potential members of the Ursa Major moving group and 5 that are potential members of the Pleiades moving group. The next obvious step towards confirming membership of these groups is to measure parallaxes for these dwarfs. Parallaxes will allow accurate distances to be used to compare with the moving group distance. \textit{Spitzer} 3.5, 4.49, 5.73 and 7.87 micron magnitudes will also be valuable for a fuller understanding of the high velocity metal poor dwarfs. | 7 | 10 | 0710.4786 |
0710 | 0710.3359_arXiv.txt | The feasibility of a mean-field dynamo in nonhelical turbulence with superimposed linear shear is studied numerically in elongated shearing boxes. Exponential growth of magnetic field at scales much larger than the outer scale of the turbulence is found. The charateristic scale of the field is $\lB\propto S^{-1/2}$ and growth rate is $\gamma\propto S$, where $S$ is the shearing rate. This newly discovered shear dynamo effect potentially represents a very generic mechanism for generating large-scale magnetic fields in a broad class of astrophysical systems with spatially coherent mean flows. | 7 | 10 | 0710.3359 |
||
0710 | 0710.1440_arXiv.txt | % The UV excess shown by elliptical galaxies in their spectra is believed to be caused by evolved low-mass stars, in particular sdB stars. The stellar system most similar to the ellipticals for age and metallicity, in which it is possible to resolve these stars, is the bulge of our Galaxy. sdB star candidates were observed in the color magnitude diagram of a bulge region by Zoccali et al.\ (2003). The follow-up spectroscopic analysis of these stars confirmed that most of these stars are bulge sdBs, while some candidates turned out to be disk sdBs or cool stars. Both spectroscopic and photometric data and a spectral library are used to construct the integrated spectrum of the observed bulge region from the UV to the optical: the stars in the color magnitude diagram are associated to the library spectra, on the basis of their evolutionary status and temperature. The total integrated spectrum is obtained as the sum of the spectra associated to the color magnitude diagram. The comparison of the obtained integrated spectrum with old single stellar population synthetic spectra calculated by Bruzual \& Charlot~(2003) agrees with age and metallicity of the bulge found by previous work. The bulge integrated spectrum shows only a very weak UV excess, but a too strict selection of the sample of the sdB star candidates in the color magnitude diagram and the exclusion of post-Asymptotic Giant Branch stars could have influenced the result. | \label{sec:intro} The UV excess that elliptical galaxies and bulge of spiral galaxies show in their spectra at $\lambda$ shorter than 2300~\AA~ was one of the most puzzling discoveries in the last 30 years, since these stellar systems are believed to be old and metal rich, without young and massive stars emitting most of their flux at short wavelength. It is now widely accepted that this UV emission is caused by evolved low mass stars, in particular Extreme Horizontal Branch stars (EHB), called also sdB stars from their spectral classification. These stars are faint in the optical wavelength range and with the current instrumentation it is impossible to resolve them in the nearest galaxies. The stellar system most similar to the ellipticals for age and metallicity in which it is possible to resolve sdB stars is the bulge of our Galaxy. A sample of sdBs star candidates was observed in the Galactic bulge by Zoccali et al.~(2003)\nocite{zoccali03}, by means of $V$ and $I$ photometry of the region MW05 from the ESO Imaging Survey (EIS\footnote{\tt http://www.eso.org/science/eis/}, the observations were taken with the Wide Field Imager, [email protected]). These stars could be either highly reddened sdBs or cooler stars affected by lower reddening. A follow-up spectroscopic analysis of these stars has been necessary and observations at the Very Large Telescope (VLT) telescope were obtained. The data reduction and the comparison of the obtained spectra with models of hot evolved stars confirmed indeed that most of these stars are bulge sdBs, while some candidates turned out to be disk sdBs or cool stars (for more details, see Busso et al.~2005\nocite{busso05}). To be sure that the observed bulge region was not peculiar, other bulge fields were searched for sdB candidates: EIS photometric data of the bulge fields MW07 and MW08 were reduced and analyzed, finding that sdB star candidates are present also in these fields. This work presents the procedure adopted (following the recipe as in Santos et al.~1995) to construct the integrated spectrum of the bulge region MW05. | 7 | 10 | 0710.1440 |
|
0710 | 0710.5780_arXiv.txt | We report high-resolution spectroscopy of 125 field stars previously observed as part of the Sloan Digital Sky Survey and its program for Galactic studies, the Sloan Extension for Galactic Understanding and Exploration (SEGUE). These spectra are used to measure radial velocities and to derive atmospheric parameters, which we compare with those reported by the SEGUE Stellar Parameter Pipeline (SSPP). The SSPP obtains estimates of these quantities based on SDSS $ugriz$ photometry and low-resolution ($R \sim 2000$) spectroscopy. For F- and G-type stars observed with high signal-to-noise ratios ($S/N$), we empirically determine the typical random uncertainties in the radial velocities, effective temperatures, surface gravities, and metallicities delivered by the SSPP to be 2.4 km s$^{-1}$, 130 K (2.2 \%), 0.21 dex, and 0.11 dex, respectively, with systematic uncertainties of a similar magnitude in the effective temperatures and metallicities. We estimate random errors for lower $S/N$ spectra based on numerical simulations. | Starting from the sixth public data release (DR-6; Adelman-McCarthy et al. 2007), the Sloan Digital Sky Survey (SDSS) provides estimates of the atmospheric parameters for a subset of the stars observed spectroscopically in the survey (those in the approximate range of temperature $4500 \le T_{\rm eff} \le 7500$~K). Following completion of the main survey (SDSS-I), the SDSS instrumentation has been devoted to several programs, including SEGUE: Sloan Extension for Galactic Understanding and Exploration, a massive survey of the stellar content of the Milky Way. Collectively, the suite of computer programs employed to determine atmospheric parameters from SEGUE data is known as the SEGUE Stellar Parameter Pipeline (SSPP). Because each of the public data releases of the SDSS includes and supersedes previous releases, DR-6 also includes atmospheric parameters for archival stellar observations in SDSS-I. These stellar parameters are derived by a series of methods, some of which consider purely spectroscopic information (continuum-normalized spectra), solely photometry (available in the survey's $ugriz$ system for all targets), or a combination of photometry and spectroscopy. Paper I in this series describes the SSPP in detail (Lee et al. 2007a). Paper II compares the predictions of the SSPP radial velocities and atmospheric parameters with likely members of Galactic globular and open clusters (Lee et al. 2007b). The SDSS uses a CCD camera (Gunn et al. 1998) on a dedicated 2.5m telescope (Gunn et al. 2006) at Apache Point Observatory, New Mexico, to obtain images in five broad optical bands ($ugriz$; Fukugita et al.~1996) over approximately 10,000~deg$^2$ of the high Galactic latitude sky. The survey data-processing software measures the properties of each detected object in the imaging data in all five bands, and determines and applies both astrometric and photometric calibrations (Lupton et al. 2001; Pier et al. 2003; Ivezi\'c et al.~2004). Photometric calibration is provided by simultaneous observations with a 20-inch telescope at the same site (Hogg et al.~2001; Smith et al.~2002; Stoughton et al.~2002; Tucker et al.~2006). A technical summary is provided by York et al. (2000). SDSS-I and the ongoing SEGUE survey have already built the largest-ever catalog of stars in the Milky Way. To date, this includes photometry in five bands for over 200 million stars and spectroscopy for nearly 300,000 stars (Adelman-McCarthy et al. 2007). The SDSS spectrographs deliver a resolving power $\lambda/$FWHM $\sim 2000$ over the wavelength range 380-900 nm. Data reduction is fully automated, and the SSPP employs the final products from the SDSS pipeline as input to produce atmospheric parameters (effective temperature, surface gravity, and metallicity) for stars with spectral types A, F, G, and K. The best results are obtained for F- and G-type stars spanning the effective temperature range $5000 < T_{\rm eff} < 7000$~K. The quality of the SSPP atmospheric parameters is evaluated using different approaches, as already described in Paper I: comparing with previously published spectral libraries, well-studied open and globular clusters, and with high-resolution observations of field stars. Existing spectral libraries are useful in order to evaluate and calibrate the SSPP methods that rely on spectroscopy alone. Allende Prieto et al. (2006) employed the low-resolution Indo-US library (Valdes et al. 2004), and high-resolution spectra from the Elodie library (Prugniel \& Soubiran 2001) and the S$^4$N archive (Allende Prieto et al. 2004). Because the $ugriz$ system was introduced with the SDSS, the stars included in existing spectral libraries lack photometry in this system. In addition, these are relatively bright stars, typically with $V<14$ mag, brighter than the bright magnitude limit of the SDSS imaging. The bright magnitude limit for the SDSS is set by the saturation threshold of the detectors at the sidereal driftscan rate of the survey. Obtaining data for these brighter stars would require special-purpose observations with a very different instrument configuration, which would call into question their value as calibration observations for the otherwise homogeneous imaging survey. Star clusters provide stringent tests of the SSPP, as the same metallicity should be derived for stars that explore wide ranges of masses and luminosities. Paper II in this series examines SSPP results for likely members of clusters included in DR-6. One cannot choose clusters with any given metallicity, but has to take what is provided by nature and accessible from Apache Point. Furthermore, the effective temperatures and surface gravities for the members of any given cluster are very strongly correlated, depending on age and chemical composition. This leads to a patchy coverage of the parameter space. Field stars, on the other hand, can be chosen to provide better coverage and, therefore, naturally complement the clusters. Among the stars spectroscopically observed with SDSS, those in the range $14 < V <16.5$ mag can be observed at high spectral resolution with large-aperture telescopes and modest integration times. Due to the vast size of the SDSS stellar sample, these stars can be selected to more uniformly cover the parameter space of stellar properties, and have the additional benefit that photometry is already available for them in the SDSS native system. This paper, the third in the SSPP series, is devoted to the analysis of 125 SDSS stars newly observed at high-resolution with the Hobby-Eberly, Keck, and Subaru telescopes. Section 2 describes the sample selection and the observations. The determination of radial velocities and atmospheric parameters, based on these observations, are discussed in \S 3 and \S 4, respectively. Section 5 describes the results for several well-known standard stars observed with the Hobby-Eberly Telescope. Section 6 compares the parameters derived from high-resolution spectroscopy with those from the SSPP. Section 7 describes numerical experiments that explore how the parameters degrade at lower signal-to-noise ratios. Our conclusions are summarized in \S 8. | We report on an analysis of high-resolution spectroscopic observations of a sample of stars previously observed with the SDSS instrumentation as part of SDSS-I or SEGUE. These new data are used to derive radial velocities and atmospheric parameters, and to scrutinize the performance of the SSPP Pipeline described in Paper I in this series. The sample we have examined includes 81 stars observed with the HET-HRS, 25 stars obtained with Keck-ESI, 11 stars observed with Keck-HIRES, and 9 stars from Subaru-HDS. Through a comparison with external spectroscopic libraries, and by employing multiple methods of analysis for the HET sample, we estimate that our reference radial velocities are accurate to 1.6 km s$^{-1}$. Our values for the stellar atmospheric parameters, effective temperature, surface gravity, and metallicity, are accurate to 1.5 \% ($\sim 90$~K), 0.13~dex and 0.05~dex, respectively. These figures are derived from the comparison with the parameters for nearby stars in the S$^4$N catalog, but we find they are still valid for the moderately high $S/N$ of the HET spectra. Using the HET sample to benchmark the SSPP, subtracting in quadrature the uncertainties in the results for the former, we conclude that the SDSS/SEGUE radial velocities are typically accurate to 2.4 km s$^{-1}$ for high signal-to-noise SDSS spectra ($S/N > 50/1$). A similar comparison of the atmospheric parameters returned by the SSPP with those obtained from HET spectra leads to the conclusion that the SSPP effective temperatures, surface gravities, and metallicities for bright targets show random errors of 2.2\% ($\sim 130$ K), 0.21 dex, and 0.11 dex, respectively. Systematic offsets of a similar size are detected for the effective temperatures and metallicities. We evaluate the expected random uncertainties as a function of $S/N$ by repeating the analysis after introducing noise in the SDSS spectra. More extended tests are underway and will be reported elsewhere. Our study also finds that the internal uncertainties delivered by the SSPP for both radial velocities and atmospheric parameters need to be systematically increased by a factor of $2-3$ in order to be consistent with our derived external errors. The uncertainties in the average SSPP atmospheric parameters are simply derived as the standard error of the mean for a Normal distribution from the multiple techniques applied to any particular target. The fact that many methods share the same spectroscopic indicators (e.g. Balmer lines or SDSS color indices to gauge $T_{\rm eff}$), and models (e.g. Kurucz's model atmospheres) may cause unaccounted correlations that result in underestimated uncertainties. The validation and calibration of the SSPP is an ongoing project. Several additional open and globular clusters have recently had data obtained with SDSS instrumentation, and will be considered in future papers. A sample of up to several hundred very low-metallicity stars from SDSS/SEGUE is presently being observed with the HET, which we will add to our calibration sample. Additional stars of intermediate metallicity, and with hotter and cooler temperatures than considered in the present work, will be added to our calibration sample based on observations with a number of large-aperture telescopes. Our goal is to produce an SSPP validation catalog for on the order of 500 stars, which will be used to refine and adjust the individual parameter estimation techniques employed by the SSPP, and thus establish a definitive atmospheric parameter estimation scale for application to the large (and growing) SDSS/SEGUE stellar samples, as well as to other future surveys. | 7 | 10 | 0710.5780 |
0710 | 0710.5922_arXiv.txt | \noindent We derive upper limits on the ratio $\fgrbccsn (z)\equiv \rgrb(z) /\rccsn(z)\, \equiv\fgrbccsn(0)(1+z)^\alpha$, the ratio of the rate, $\rgrb$, of long-duration Gamma Ray Bursts (GRBs) to the rate, $\rccsn$, of core-collapse supernovae (CCSNe) in the Universe ($z$ being the cosmological redshift and $\alpha\geq 0$), by using the upper limit on the diffuse TeV--PeV neutrino background given by the AMANDA-II experiment in the South Pole, under the assumption that GRBs are sources of TeV--PeV neutrinos produced from decay of charged pions produced in $p\gamma$ interaction of protons accelerated to ultrahigh energies at internal shocks within GRB jets. For the assumed ``concordance model'' of cosmic star formation rate, $\rsf$, with $\rccsn (z) \propto \rsf (z)$, our conservative upper limits are $\fgrbccsn(0)\leq 5.0\times10^{-3}$ for $\alpha=0$, and $\fgrbccsn(0)\leq 1.1\times10^{-3}$ for $\alpha=2$, for example. These limits are already comparable to (and, for $\alpha\geq 1$, already more restrictive than) the current upper limit on this ratio inferred from other astronomical considerations, thus providing a useful independent probe of and constraint on the CCSN-GRB connection. Non-detection of a diffuse TeV--PeV neutrino background by the up-coming IceCube detector in the South pole after three years of operation, for example, will bring down the upper limit on $\fgrbccsn (0)$ to below few $\times10^{-5}$ level, while a detection will confirm the hypothesis of proton acceleration to ultrahigh energies in GRBs and will potentially also yield the true rate of occurrence of these events in the Universe. | } Detection of supernova (SN) features in the afterglow spectra of several long duration (typically $>2\s$) Gamma Ray Bursts (GRBs) in the past one decade has provided strong support to the hypothesis that a significant fraction, if not all, of the long duration GRBs arise from collapse of massive stars; see, e.g., Refs.~\cite{Meszaros_araa02,Woosley-Bloom_araa06,DellaValle06} for recent reviews. The observed SN features in the GRB afterglow spectra are similar to those usually associated with core-collapse supernovae (CCSNe) of Type Ib/c (see, e.g., ~\cite{Woosley-Heger06,DellaValle06}). The total energy (corrected for beaming) in keV--MeV gamma rays emitted by typical long-duration GRBs is of order $10^{51}\erg$, which is roughly the same as the total explosion energy seen in typical CCSNe, although there exists considerable diversity in the energetics of both the SN and the GRB components in the SN-GRB associations observed so far. In particular, the estimated explosion energies of the SNe associated with the GRBs observed so far seem to be somewhat larger than those of normal SNe, leading to this ``special'' class of SNe being sometimes referred to as ``hypernovae''. The broad class of observational results on SN-GRB associations can be understood within the context of the ``collapsar'' model~\cite{collapsar_model_refs} in terms of a simple phenomenological picture (see, e.g., ~\cite{Woosley-Zhang_astroph_0701320}) in which the core-collapse of a massive Wolf-Rayet star gives rise to two kinds of outflows emanating from the central regions inside the collapsed star: (a) a narrowly collimated and highly relativistic jet that is responsible for the GRB activity, the jet being driven, for example, by a rapidly rotating and accreting black hole formed at the center in the core-collapse process, and (b) a more wide-angled, quasi-spherical and non-relativistic (or at best sub-relativistic) outflow that goes to blow up the star and gives rise to the supernova. The energies channeled into these two components may in general vary independently, which may explain the diversity of energetics in the observed SN-GRB associations. Actually, depending on the energy contained in it the ``GRB jet'' may or may not be able to penetrate through the stellar material and emerge outside. Indeed, the fact that the SN-GRB associations observed so far involve CCSNe of Type Ib/c, but not of Type II, may be due to the inability of the GRB-causing jet to penetrate through the relatively larger amount of outer stellar material in the case of Type II SN as compared to that in SNe of Type Ib/c~\cite{fn1}. Considering various factors that may govern the energy channeled into the GRB-causing jet, such as the mass and rotation rate of the black hole, accretion efficiency, efficiency of conversion of accretion energy into collimated relativistic outflow, and so on, Woosley and Zhang~\cite{Woosley-Zhang_astroph_0701320} have obtained a rough lower limit of $\sim10^{48}\erg/\s$ for the power required for the jet to be able to emerge from the star. This is consistent with the energetics of the GRB components of the SN-GRB associations observed so far. While SN-GRB associations strongly support the stellar core-collapse origin of most long-duration GRBs, clearly, not all core-collapse events may result in a GRB --- the latter depends on whether or not the core-collapse event actually results in a ``central engine'' (a rotating black hole fed by an accretion disk in the above mentioned phenomenological picture, for example) that is capable of driving the required collimated relativistic outflow. In other words, while every long-duration GRB would be expected to be accompanied by a core-collapse supernova~\cite{fn2}, the reverse is not true in general. What fraction of all stellar core-collapse events in the universe produce GRBs? Methods based on astronomical observations generally indicate the ratio between the cosmic GRB rate and the cosmic Type Ib/c SN rate, $\fgrbsnIbc$, to be in the range $\sim 10^{-3}$ -- $10^{-2}$ for a wide variety of different assumptions on various relevant parameters such as those that characterize the cosmic star formation rate (SFR), initial mass function (IMF) of stars, masses of Type Ib/c SN progenitors, the luminosity function of GRBs, the beaming factor of GRBs (associated with the fact that individual GRB emissions are highly non-isotropic and confined to narrowly collimated jets covering only a small fraction of the sky), and so on; see, for example, \cite{Guetta-DellaValle_06,Bissaldi_etal_07} and references therein. The dominant uncertainty in the estimate of $\fgrbsnIbc$ comes from the uncertainties in the estimates of the local GRB rate and the average GRB beaming factor. However, irrespective of the exact value of the ratio $\fgrbsnIbc$, it is clear that this ratio is significantly less than unity. This indicates that, apart from just being sufficiently massive stars, the GRB progenitors may need to satisfy additional special conditions. For example, it has been suggested~\cite{Woosley-Bloom_araa06} that the degree of rotation of the central iron core of the collapsing star and the metalicity of the progenitor star may play crucial roles in producing a GRB. In this paper, we discuss an alternative probe of the cosmic GRB rate that uses the predicted high energy (TeV--PeV) diffuse neutrino background produced by GRBs and the experimental upper limit on high energy diffuse neutrino background given by the AMANDA-II experiment in the South Pole~\cite{amanda_II_limit}. Existence of a high (TeV--PeV) energy diffuse GRB neutrino background (DGRBNuB) due to $p\gamma$ interactions of (ultra)high energy protons accelerated within GRB sources is a generic prediction~\cite{Waxman-Bahcall_grbnu} in most currently popular models of GRBs. This DGRBNuB is subject to being probed by the currently operating and up-coming large volume (kilometer scale) neutrino detectors such as IceCube~\cite{icecube}, ANITA~\cite{anita}, ANTARES~\cite{antares}, for example. Since neutrinos, unlike electromagnetic radiation, can travel un-hindered from the furthest cosmological distances, the DGRBNuB automatically includes the contributions from all GRBs in the Universe. Thus, an analysis of the DGRBNuB is likely to provide a good picture of the true rate of occurrence of these events in the Universe. Indeed, as we show in this paper, the upper limits on $\fgrbccsn (0)$, the ratio of the local (i.e., redshift $z=0$) GRB to CCSN rates, derived here from the consideration of DGRBNuB, are, for a wide range of values of the relevant parameters, already more restrictive than the current upper limit on this ratio ($\sim 2.5\times10^{-3}$) inferred from other astronomical considerations~\cite{Guetta-DellaValle_06,Bissaldi_etal_07}. Further, non-detection of a diffuse TeV--PeV neutrino background by the up-coming IceCube detector~\cite{icecube} in the South Pole after three years of operation, for example, will imply upper limits on $\fgrbccsn(0)$ at the level of few \ $\times10^{-5}$, while a detection of the DGRBNuB will provide strong support to the hypothesis of proton acceleration to ultrahigh energies within GRB jets. Our use of the DGRBNuB in constraining the cosmic GRB rate is in the same spirit as efforts to constrain the cosmic star formation rate (and thereby the cosmic CCSN rate) by using the experimental upper limit (set by the Super-Kamiokande (SK) detector)~\cite{SK_limit} on the predicted~\cite{dsnub_original} low (few MeV) energy Diffuse Supernova Neutrino Background (DSNuB); see, for example, Refs.~\cite{Ando-Sato_rev_04,Strigari_etal_05,Hopkins-Beacom_06}. Now that the cosmic SFR including its absolute normalization and thereby the cosmic CCSNe rate have got reasonably well determined by the recent high quality data from a variety of astronomical observations (see, e.g., ~\cite{Hopkins-Beacom_06}) (which, by the way, predicts a DSNuB flux that is close to the SK upper limit, implying that the DSNuB is probably close to being detected in the near future), one can begin to think of using this SFR to constrain the ratio of the cosmic GRB rate to CCSNe rate by using the predicted DGRBNuB flux together with the recent upper limits on the diffuse high energy neutrino flux from neutrino telescopes. We should emphasize here that the upper limits derived in this paper actually refer to the ratio of the rate of GRBs to that of {\it all} CCSNe including those of Type Ib/c and Type II, although SN-GRB associations observed so far involve SNe of Type Ib/c only. It is known, however, that Type II SNe probably constitute as much as $\sim$ 75\% of all CCSNe; see, e.g., \cite{Cappellaro_etal_07}. Thus, one can get the constraint on the GRB-to-SNIb/c ratio from the GRB-to-CCSNe ratio we obtain here by multiplying the latter by a factor of $\sim 4$. Conversely, for later comparison, we shall take the ``observed'' value of the ratio $\fgrbccsn(0)$ to be in the range $2.5\times (10^{-4}$ -- $10^{-3})$~\cite{Guetta-DellaValle_06,Bissaldi_etal_07}. Below, we first briefly review the calculation of the DGRBNuB spectrum in section \ref{sec:DGRBNuB_calc}. The resulting upper limits on $\fgrbccsn$ obtained by comparing the DGRBNuB with the current upper limit from AMANDA-II experiment are discussed in section \ref{sec:fgrbccsn_constraints} for various values of some of the relevant GRB parameters. Finally, in section \ref{sec:summary} we summarize the main results and conclude. | } In this paper we have attempted to derive upper limits on the fraction $\fgrbccsn$ of all stellar core-collapse events that give rise to GRBs, by using the current experimental upper limit on the high energy (TeV -- PeV) diffuse neutrino background given by the AMANDA-II experiment in the South Pole, under the assumption that GRBs are sources of such high energy neutrinos. High energy neutrinos are predicted to be produced within GRB jets through photopion production by protons and subsequent decay of the charged pions, provided protons are accelerated to ultrahigh energies at the internal shocks within GRB jets. In our calculation we have allowed for a possible evolution of the cosmic GRB rate relative to star formation rate. For a wide range of values of various parameters, the upper limits on $\fgrbccsn(0)$ derived here from the AMANDA-II results are already more restrictive than the upper limit on this ratio inferred from other astronomical considerations, thus providing a useful independent probe of and constraint on the CCSN-GRB connection. The closeness of the upper limits on $\fgrbccsn(0)$ derived here (in particular for the case of enhanced evolution of the GRB rate relative to the star formation rate at high redshifts) to the lower limit on this ratio inferred from various astronomical considerations seems to indicate that the predicted DGRBNuB flux should be detectable by the upcoming detectors such as IceCube which will have significantly improved sensitivity over that of AMANDA-II. On the other hand, non-detection of the DGRBNuB by the IceCube detector after three years of operation, for example, will give more stringent upper limits on $\fgrbccsn$, but at the same time will also imply that either the values of $\fgrbccsn$ inferred from direct astronomical observations have been significantly overestimated (which is possible, for example, due to incorrect estimates of the average GRB beaming factor) or that the assumption of proton acceleration to ultrahigh energies within GRB jets is invalid, or both of these. However, more precise determination of the distribution of some of the crucial GRB parameters such as the bulk flow Lorentz factor and variability timescale of the GRBs will be needed to reliably calculate the expected contribution of the GRBs to the high energy diffuse neutrino background, and thereby to determine the upper limits on $\fgrbccsn$ more reliably. To conclude, then, the up-coming large volume neutrino telescopes hold immense promise of yielding significant information both on the nature of the fundamental physical process of particle acceleration in GRB sources as well as on the rate of occurrence of these events in the Universe. One of us (PB) wishes to thank Nayantara Gupta for helpful clarifications. | 7 | 10 | 0710.5922 |
0710 | 0710.4029_arXiv.txt | The cold dark matter (CDM) scenario generically predicts the existence of triaxial dark matter haloes which contain notable amounts of substructure. However, analytical halo models with smooth, spherically symmetric density profiles are routinely adopted in the modelling of light propagation effects through such objects. In this paper, we address the biases introduced by this procedure by comparing the surface mass densities of actual N-body haloes against the widely used analytical model suggested by Navarro, Frenk and White (1996) (NFW). We conduct our analysis in the redshift range of 0.0 - 1.5. In cluster sized haloes, we find that triaxiality can cause scatter in the surface mass density of the haloes up to $\sigma_+ = +60 \%$ and $\sigma_- = -70 \%$, where the 1-$\sigma$ limits are relative to the analytical NFW model given value. Subhaloes can increase this scatter to $\sigma_+ = +70 \%$ and $\sigma_- = -80 \%$. In galaxy sized haloes, the triaxial scatter can be as high as $\sigma_+ = +80 \%$ and $\sigma_- = -70 \%$, and with subhaloes the values can change to $\sigma_+ = +40 \%$ and $\sigma_- = -80 \%$. We present an analytical model for the surface mass density scatter as a function of distance to the halo centre, halo redshift and halo mass. The analytical description enables one to investigate the reliability of results obtained with simplified halo models. Additionally, it provides the means to add simulated surface density scatter to analytical density profiles. As an example, we discuss the impact of our results on the calculation of microlensing optical depths for MACHOs in CDM haloes. | The cold dark matter model, in which the non-baryonic part of the dark matter is assumed to consist of particles that were non-relativistic already at the time of decoupling, and that interact predominantly through gravity, has been very successful in explaining the formation of large-scale structures in the Universe \citep[see e.g.][ for a review]{Primack}. In this scenario, both galaxies and galaxy clusters are hosted by CDM haloes, which formed hierarchically through mergers of smaller subunits. Even though N-body simulations generically predict CDM haloes to be triaxial \citep[e.g.][]{Jing & Suto} with substantial amounts of substructures left over from the merging process \citep[e.g.][]{Moore et al.}, simplified halo models are often adopted in the modelling of light propagation through such objects. The most common approach is to treat dark matter haloes as spherical objects with smooth density profiles, usually either of the NFW \citep{NFW} form, some generalization thereof \citep{Zhao}, or that of a cored or singular isothermal sphere. The light emitted from high-redshift objects such as quasars, supernovae, gamma-ray bursts, galaxies and galaxy clusters will typically have to pass through many dark matter haloes before reaching an observer on Earth. Several investigations have already indicated that smooth and/or spherical halo models may lead to incorrect results when treating the gravitational lensing effects associated with such foreground mass condensations \citep[e.g][]{Bartelmann & Weiss,Dalal et al.,Oguri & Keeton,Hennawi et al.} More realistic features like triaxiality and substructures can be included in gravitational lens calculations either by employing N-body simulations directly \citep[e.g.][]{Bartelmann & Weiss,Seljak & Holz,Holopainen et al.} or by using analytical expressions for the halo shapes \citep[e.g.][]{Kochanek,Golse & Kneib,Evans & Hunter,Chae} and subhalo properties \citep[e.g.][]{Oguri b,Zackrisson & Riehm}. While N-body simulations often represent the safest choice, the approach is computationally demanding and does not always allow one to identify the features of the mass distribution responsible for a specific lensing effect. Methods which bring simple, analytical halo models into contact with the full phenomenology of the N-body simulations are therefore highly desirable. In this paper, we focus on the projected mass density of CDM haloes as a function of distance from the halo centre. There are several situations in gravitational lensing when realistic estimates of the surface mass density (i.e. convergence) along a given line of sight through a dark halo may be important. Examples include the calculation of image separations in strong lensing by subhaloes located in the external potential of its host halo \citep{Oguri b}, attempts to correct the luminosities of supernovae type Ia for the magnification by foreground haloes \citep[e.g.][]{Gunnarsson} and estimates of the distribution of microlensing optical depths for high-redshift MACHOs \citep[e.g.][]{Wyithe & Turner,Zackrisson & Riehm}. Other applications include the assessments of light propagation effects in models with non-zero coupling betweeen dark matter particles and photons \citep[e.g.][]{Profumo & Sigurdson}. Here, we use high-resolution, dissipationless N-body simulations of CDM haloes to investigate the errors in surface mass density introduced by treating these objects as spherical with smooth density profiles of the NFW type. Simple relations for the surface mass density error as a function of halo redshift and distance to the halo centre are presented, making it easy to investigate the reliablitiy of results obtained with simplified halo models. On a related note, \citet{KnebeWiessner} recently investigated the error introduced by spherically averaging an elliptical mass distribution. They found that for axis ratios typical for cosmological dark matter haloes, the variance in the local density can be as large as 50\% in the outer parts. The current paper examines the problem of halo triaxiality from a slightly different point of view. The N-body simulations used are described in Section 2. In Section 3, we describe the methods for extracting the halo sample. In Section 4, we compare the CDM surface mass densities obtained along random sightlines through the N-body haloes to the corresponding results obtained from smooth and spherical NFW models fitted to the same haloes. Section 5 presents a set of simple relations for the surface mass density errors introduced by this procedure as a function of distance to the halo centre and halo redshift. Section 6 discusses how these relations may be used in the context of optical depth estimates for MACHO microlensing. A number of caveats are discussed in Section 7. Section 8 summarizes our findings. | The results presented here do suffer from a number of shortcomings which should be pointed out. The simulations used are dissipationless. In reality, dark matter haloes contain baryons, and the dissipation and feedback associated with these will inevitably affect the overall potential of the system, and thereby the spatial distribution of the CDM. According to current models, baryonic cooling will increase the central density of the CDM \citep[e.g.][]{Gnedin et al.} and also make the halo more spherical \citep{Kazantzidis et al.}. The significance of these effects are, however, still difficult to predict reliably, as the gas dynamical simulations involved still suffer from so-called ``overmerging'' problems \citep[e.g.][]{Balogh et al., Springel & Hernquist}. When calculating the surface mass density profiles, we have moreover considered only the matter present within $r_\mathrm{vir}$ of each halo, whereas simulations have shown that galaxy sized CDM haloes extend at least out to 2--3$r_\mathrm{vir}$ \citep{Prada et al.}. We restricted our analysis to $r_\mathrm{vir}$ because it becomes increasingly demanding to separate halo particles from the background the further one wants to extend the analysis. Even in the presented case, we need to separate particles out to $\sim 1.5 r_\mathrm{vir}$ because the smoothed particles extend their influence inside the virial radius region even though they are positioned outside it. We tested our method out to 3$r_\mathrm{vir}$, but the number counts of the halo particles at those distances are too low to produce reliable results. Our halo sample does not have the resolution needed for extending the analysis further than $r_\mathrm{vir}$ safely. The most significant limitation of this paper is the small number of haloes in our analysis. This is of course due to the limited resolution of the cosmological simulations we had access to. We would like to repeat our analysis with a more complete statistical sample of haloes, which would hopefully confirm our analytical description with smaller error bars. We also note that our analytical description of the surface mass density is more reliable in the case in which subhaloes are {\it excluded}. This is because subhaloes can introduce significant mass peaks to some radial bins. These peaks can lead to unwanted effects in the log-normal fitting procedure which is designed to handle relatively smooth and continuous mass distributions within a bin. Large subhalos can also disturb the NFW fits, at last in the low density regions. Thus, the use of the models which include subhaloes is discouraged. | 7 | 10 | 0710.4029 |
0710 | 0710.1110_arXiv.txt | We examine the effects Lorentz violation on observations of cosmic microwave background radiation. In particular, we focus on changes in polarization caused by vacuum birefringence. We place stringent constraints on previously untested violations. | 7 | 10 | 0710.1110 |
||
0710 | 0710.3862_arXiv.txt | We study a large-scale instability in a sheared nonhelical turbulence that causes generation of large-scale vorticity. Three types of the background large-scale flows are considered, i.e., the Couette and Poiseuille flows in a small-scale homogeneous turbulence, and the "log-linear" velocity shear in an inhomogeneous turbulence. It is known that laminar plane Couette flow and antisymmetric mode of laminar plane Poiseuille flow are stable with respect to small perturbations for any Reynolds numbers. We demonstrate that in a small-scale turbulence under certain conditions the large-scale Couette and Poiseuille flows are unstable due to the large-scale instability. This instability causes formation of large-scale vortical structures stretched along the mean sheared velocity. The growth rate of the large-scale instability for the "log-linear" velocity shear is much larger than that for the Couette and Poiseuille background flows. We have found a turbulent analogue of the Tollmien-Schlichting waves in a small-scale sheared turbulence. A mechanism of excitation of turbulent Tollmien-Schlichting waves is associated with a combined effect of the turbulent Reynolds stress-induced generation of perturbations of the mean vorticity and the background sheared motions. These waves can be excited even in a plane Couette flow imposed on a small-scale turbulence when perturbations of mean velocity depend on three spatial coordinates. The energy of these waves is supplied by the small-scale sheared turbulence. | Large-scale vortical structures are universal features observed in geophysical, astrophysical and laboratory flows (see, e.g., \cite{L83,P87,C94,GLM97,T98,RAO98}). Formation of vortical structures is related to the Prandtl secondary flows (see, e.g., \cite{P52,T56,P70,B87}). A lateral stretching (or ''skewing") by an existing shear generates streamwise vorticity that results in formation of the first kind of the Prandtl secondary flows. In turbulent flow the large-scale vorticity is generated by the divergence of the Reynolds stresses. This mechanism determines the second kind of the Prandtl turbulent secondary flows \cite{B87}. The generation of large-scale vorticity in a homogeneous nonhelical turbulence with an imposed large-scale linear velocity shear has been recently studied in \cite{EKR03}. Let us discuss a mechanism of this phenomenon. The equation for the mean vorticity ${\bf W} = \bec{\nabla} {\bf \times} {\bf U}$ read \begin{eqnarray} {\partial {\bf W} \over \partial t} = \bec{\nabla} {\bf \times} ({\bf U} {\bf \times} {\bf W} + {\bf F} - \nu \bec{\nabla} {\bf \times} {\bf W}) \;, \label{W10} \end{eqnarray} where ${\bf U}$ is the mean fluid velocity, ${\bf F}_i = - \nabla_j \, \langle u_i u_j \rangle$ is the effective force caused by velocity fluctuations, ${\bf u}$, and $ \nu$ is the kinematic viscosity. The first term, ${\bf U} {\bf \times} {\bf W}$, in Eq.~(\ref{W10}) determines laminar effects of the mean vorticity production caused by the sheared motions, while the effective force ${\bf F}$ determines the turbulent effects on the mean fluid flow. Let us consider a simple large-scale linear velocity shear ${\bf U}^{(s)} = (0, Sx, 0)$ imposed on the small-scale nonhelical turbulence. The equation for the perturbations of the mean vorticity, $\tilde{\bf W} = (\tilde{W}_x(z), \tilde{W}_y(z), 0)$, reads \begin{eqnarray} {\partial \tilde{W}_x \over \partial t} &=& S \, \tilde{W}_y + \nu_{_{T}} \tilde{W}''_x \;, \label{E2}\\ {\partial \tilde{W}_y \over \partial t} &=& - \beta_0 \, S \, l_0^2 \, \tilde{W}''_x + \nu_{_{T}} \tilde{W}''_y \;, \label{E3} \end{eqnarray} (see \cite{EKR03}), where $\tilde{W}'' = \partial^2 \tilde{W} /\partial z^2$, $\, \nu_{_{T}}$ is the turbulent viscosity, $l_0$ is the maximum scale of turbulent motions and the parameter $\beta_0$ is of the order of 1, and depends on the scaling exponent of the correlation time of the turbulent velocity field (see Sect. II). A solution of Eqs.~(\ref{E2}) and~(\ref{E3}) has the form $ \propto \exp(\gamma t + i K_z z)$, where the growth rate of the large-scale instability is given by $\gamma = \sqrt{\beta_0} \, S \, l_0 \, K_z - \nu_{_{T}} \, K_z^2$ and $K_z$ is the wave number. The maximum growth rate of perturbations of the mean vorticity, $ \gamma_{\rm max} = \beta_0 \, (S \, l_0)^2 / 4 \nu_{_{T}}$, is attained at $ K_z = K_m = \sqrt{\beta_0} \, S \, l_0 /2 \nu_{_{T}}$. This corresponds to the ratio $\tilde{W}_y / \tilde{W}_x = \sqrt{\beta_0} \, l_0 \, K_m \approx S \, \tau_0$, where the time $\tau_0 = l_{0} / u_0$ and $u_0$ is the characteristic turbulent velocity in the maximum scale $l_{0}$ of turbulent motions. Note that in a laminar flow this instability does not occur. The mechanism of this instability is as follows (see \cite{EKR03} for details). The first term, $S \tilde{W}_y = ({\bf W}^{(s)}\cdot\bec{\nabla})~\tilde{U}_x$, in Eq.~(\ref{E2}) determines a ''skew-induced" generation of perturbations of the mean vorticity $\tilde{W}_x$ by stretching of the equilibrium mean vorticity ${\bf W}^{(s)}= (0,0,S)$, where $\tilde{\bf U}$ are the perturbations of the mean velocity. In particular, the mean vorticity $\tilde{W}_x {\bf e}_x$ is generated from $\tilde{W}_y {\bf e}_y$ by equilibrium shear motions with the mean vorticity ${\bf W}^{(s)}$, whereby $\tilde{W}_x {\bf e}_x \propto ({\bf W}^{(s)} \cdot \bec{\nabla}) \tilde{U}_x {\bf e}_x \propto \tilde{W}_y {\bf e}_y \times {\bf W}^{(s)} $. Here ${\bf e}_x$, ${\bf e}_y$ and ${\bf e}_z$ are the unit vectors along $x$, $y$ and $z$ axes, respectively. On the other hand, the first term, $- \beta_0 \, S \, l_0^2 \, \tilde{W}''_x$, in Eq.~(\ref{E3}) determines a ''Reynolds stress-induced" generation of perturbations of the mean vorticity $\tilde{W}_y$ by the Reynolds stresses. In particular, this term is determined by $ (\bec{\nabla} {\bf \times} {\bf F})_y$. This implies that the component of the mean vorticity $\tilde{W}_y {\bf e}_y $ is generated by an effective anisotropic viscous term $ \propto - l_0^2 \, \Delta \, (\tilde{W}_x {\bf e}_x \cdot \bec{\nabla}) \, {U}^{(s)}(x) {\bf e}_y \propto - l_0^2 \, S \, \tilde{W}''_x {\bf e}_y .$ This instability is caused by a combined effect of the sheared motions (''skew-induced" generation) and the ''Reynolds stress-induced" generation of perturbations of the mean vorticity. The mechanism for this large-scale instability in a sheared nonhelical homogeneous turbulence is different from that discussed in \cite{MST83,KMT91,CMP94}, where the generation of large-scale vorticity in the helical turbulence occurs due to hydrodynamic alpha effect. The latter effect is associated with the hydrodynamic helicity of turbulent flow. In a nonhelical homogeneous turbulence this effect does not occur. The large-scale instability in a nonhelical homogeneous turbulence has been studied in \cite{EKR03} only for a simple case of unbounded turbulence with an imposed linear velocity shear and when the perturbations of the mean vorticity depend on one spatial variable $z$. In this study the theoretical approach proposed in \cite{EKR03} is further developed and applied for comprehensive investigation of the large-scale instability for different situations with nonuniform shear, inhomogeneous turbulence and a more general form of the perturbations of the mean vorticity $\tilde{\bf W}({\bf r})$ that depends on three spatial variables. In the present study we consider three types of the background large-scale flows, i.e., the Couette flow (linear velocity shear) and Poiseuille flow (quadratic velocity shear) in a small-scale homogeneous turbulence, and the "log-linear" velocity shear in an inhomogeneous turbulence. We have derived new mean-field equations for perturbations of large-scale velocity which depend on three spatial coordinates in a small-scale sheared turbulence, for a nonuniform background large-scale velocity shear and for an arbitrary scaling of the correlation time $\tau(k)$ of the turbulent velocity field. The stability of the laminar Couette and Poiseuille flows in a problem of transition to turbulence has been studied in a number of publications (see, e.g., \cite{DR81,SH01,CJJ03,BOH88,REM03,ESH07}, and references therein). It is known that laminar plane Couette flow and antisymmetric mode of laminar plane Poiseuille flow are stable with respect to small perturbations for any Reynolds numbers. A symmetric mode of laminar plane Poiseuille flow is stable when the Reynolds number is less than 5772 \cite{CJJ03}. In laminar flows the Tollmien-Schlichting waves can be excited. The molecular viscosity plays a destabilizing role in laminar flows which promotes the excitation of the Tollmien-Schlichting waves (see, e.g., \cite{SH01}). These waves are growing solutions of the Orr-Sommerfeld equation. In the present study we have found a turbulent analogue of the Tollmien-Schlichting waves. These waves are excited by a small-scale sheared turbulence, i.e., by a combined effect of the turbulent Reynolds stress-induced generation of perturbations of the mean vorticity and the background sheared motions. The energy of these waves is supplied by the small-scale sheared turbulence. We demonstrate that the off-diagonal terms in the turbulent viscosity tensor play a crucial role in the excitation of the turbulent Tollmien-Schlichting waves. These waves can be excited even in a plane Couette flow imposed on a small-scale turbulence when perturbations of velocity depend on three spatial coordinates. When perturbations of large-scale velocity depend on one or two spatial coordinates the turbulent Tollmien-Schlichting waves can not be excited in a sheared turbulence. In the present study we show that the large-scale Couette and Poiseuille flows imposed on a small-scale turbulence can be unstable with respect to small perturbations. The critical effective Reynolds number (based on turbulent viscosity) required for the excitation of this large-scale instability, is of the order of 200. This paper is organized as follows. In Sect. II the governing equations are formulated. In Sect. III we consider a homogeneous turbulence with a large-scale linear velocity shear (Couette flow), while in Sect. IV we study a homogeneous turbulence with a large-scale quadratic velocity shear (Poiseuille flow). In Sect. V we investigate formation of large-scale vortical structures in an inhomogeneous turbulence with an imposed nonuniform velocity shear. Finally, we draw conclusions in Sec.~VI. | In this study the theoretical approach proposed in \cite{EKR03} is further developed and applied to investigate the large-scale instability in a nonhelical turbulence with a nonuniform shear and a more general form of the perturbations of the mean vorticity. In particular, we consider three types of the background large-scale sheared flows imposed on small-scale turbulence: Couette flow (linear velocity shear) and Poiseuille flow (quadratic velocity shear) in a small-scale homogeneous turbulence, and a more complicated nonuniform velocity shear with the logarithmic velocity profile near the boundaries matched with the linear shear velocity for the central part of the background flow. This nonuniform velocity shear is imposed on an inhomogeneous turbulence. The latter flow is typical for the atmospheric boundary layer. We show that the large-scale Couette and Poiseuille flows imposed on a small-scale turbulence are unstable with respect to small perturbations due to the excitation of the large-scale instability. This instability causes generation of large-scale vorticity and formation of large-scale vortical structures. The size of the formed vortical structures in the direction of the background velocity shear is much larger than the sizes of the structures in the directions perpendicular to the velocity shear. Therefore, the large-scale structures formed during this instability are stretched along the mean sheared velocity. Increase of shear promotes the large-scale instability. The thresholds for the excitation of the large-scale instability in the value of shear and the aspect ratio of structures for Poiseuille background flow are larger than that for the Couette background flow. The growth rate of the large-scale instability for the inhomogeneous turbulence with the "log-linear" velocity shear is much larger than that for the Couette and Poiseuille background flows. The characteristic spatial and time scales for the instability are much larger than the characteristic turbulent scales. This justifies separation of scales which is required for the validity of the mean-field theory applied in the present study. The large-scale instability results in excitation of the turbulent Tollmien-Schlichting waves. The mechanism for the excitation of these waves is different from that for the Tollmien-Schlichting waves in laminar flows. In particular, the molecular viscosity plays a crucial role in the excitation of the Tollmien-Schlichting waves in laminar flows. Contrary, the turbulent Tollmien-Schlichting waves are excited by a combined effect of the turbulent Reynolds stress-induced generation of perturbations of the mean vorticity and the background sheared motions. The energy of these waves is supplied by the small-scale sheared turbulence, and the off-diagonal terms in the turbulent viscosity tensor play a crucial role in the excitation of the turbulent Tollmien-Schlichting waves. Note that this study is principally different from the problems of transition to turbulence whereby the stability of the laminar Couette and Poiseuille flows are investigated (see, e.g., \cite{DR81,SH01,CJJ03,BOH88,REM03,ESH07}, and references therein). Here we do not analyze a transition to turbulence. We study the large-scale instability caused by an effect of the small-scale anisotropic turbulence on the mean flow. This anisotropic turbulence is produced by an interaction of equilibrium large-scale Couette or Poiseuille flows with a small-scale isotropic background turbulence produced by, e.g., a steering force. The anisotropic velocity fluctuations are generated by tangling of the mean-velocity gradients with the velocity fluctuations of the background turbulence \cite{EKR03,EKRZ02}. The "tangling" mechanism is an universal phenomenon that was introduced in \cite{W57,BH59} for a passive scalar and in \cite{G60,M61} for a passive vector (magnetic field). The Reynolds stresses in a turbulent flow with a mean velocity shear is another example of tangling anisotropic fluctuations \cite{L67}. For instance, these velocity fluctuations are anisotropic in the presence of shear and have a steeper spectrum $\propto k^{-7/3}$ than, e.g., a Kolmogorov background turbulence (see, e.g., \cite{L67,WC72,SV94,IY02,EKRZ02}). The anisotropic velocity fluctuations determine the effective force and the Reynolds stresses in Eq.~(\ref{B15}). This is the reason for the new terms $\propto \beta_n \, l_0^2$ appearing in Eqs.~(\ref{AA1})-(\ref{A2}). The obtained results in this study may be of relevance in different turbulent astrophysical, geophysical and industrial flows. Turbulence with a large-scale velocity shear is a universal feature in astrophysics and geophysics. In particular, the analyzed effects may be important, e.g., in accretion disks, extragalactic clusters, merged protostellar and protogalactic clouds. Sheared motions between interacting clouds can cause an excitation of the large-scale instability which results in generation of the mean vorticity and formation of large-scale vortical structures (see, e.g., \cite{P80,ZN83,C93}). Dust particles can be trapped by the vortical structures to enhance agglomeration of material and formation of particle clusters \cite{BS95,BR98,EKR98,CH00,JAB04}. The suggested mechanism can be used in the analysis of the flows associated with Prandtl's turbulent secondary flows (see, e.g., \cite{P52,B87}). However, in this study we have investigated only simple physical mechanisms to describe an initial (linear) stage of the formation of vortical structures. The simple models considered in this study can only mimic the flows associated with turbulent secondary flows. Clearly, the comprehensive numerical simulations of the nonlinear problem are required for quantitative description of the turbulent secondary flows. | 7 | 10 | 0710.3862 |
0710 | 0710.2864_arXiv.txt | We present results from two high--contrast imaging surveys that exploit a novel technique, L--band angular differential imaging. Our first survey targeted 21 young stars in the $\beta$~Pic and Tuc--Hor moving groups with VLT/NACO reaching typical sensitivities of $<$1~M$_{\mathrm Jup}$ at $r>20$~AU. The statistical analysis of the null result demonstrates that the giant planet population is truncated at 30~AU or less (90\% confidence level). Our second, on--going MMT/Clio survey utilizes the unique sensitivity achieved in the L--band for old planets to probe all M--dwarf stars within 6~pc. The proximity of these targets enables direct imaging of planets in orbits like Jupiter for the first time --- a key step for directly imaging giant planets. | The study of exoplanetary systems is arguably the most rapidly developing field in modern astrophysics. Surprisingly, much progress has been made without directly imaging a single planet: radial velocity/microlensing and primary and secondary planet eclipses provide limited, but valuable insights. Direct imaging of planetary systems will have a fundamental impact on the field --- a single, low--resolution 3--5 $\mu$m spectrum of a planet may carry more information than all existing Spitzer transit photometry combined. As was the case in the search for the first brown dwarf, or for radial velocity and planet transit techniques, achieving the first firm detection is a very difficult and often frustrating challenge. But these investments paid off rapidly by opening whole new classes of objects for study. | We present results from a novel high--contrast imaging technique. Our NACO survey of 21 nearby young stars demonstrates that the giant planet population does not extend beyond 30 AU and suggests a cut-off at radius $<$ 15 AU. Most previous imaging surveys have not detected planets because they targeted young stars ($>$ 20 pc) forcing them to probe orbital radii $>$20~AU. Our ongoing MMT 6pc volume--limited survey is probing --- for the first time --- the massive giant planet population around the closest stars with orbital radii $>$3. | 7 | 10 | 0710.2864 |
0710 | 0710.5169_arXiv.txt | We construct new models of black hole-neutron star binaries in quasiequilibrium circular orbits by solving Einstein's constraint equations in the conformal thin-sandwich decomposition together with the relativistic equations of hydrostationary equilibrium. We adopt maximal slicing, assume spatial conformal flatness, and impose equilibrium boundary conditions on an excision surface (i.e., the apparent horizon) to model the black hole. In our previous treatment we adopted a ``leading-order" approximation for a parameter related to the black-hole spin in these boundary conditions to construct approximately nonspinning black holes. Here we improve on the models by computing the black hole's quasilocal spin angular momentum and setting it to zero. As before, we adopt a polytropic equation of state with adiabatic index $\Gamma=2$ and assume the neutron star to be irrotational. In addition to recomputing several sequences for comparison with our earlier results, we study a wider range of neutron star masses and binary mass ratios. To locate the innermost stable circular orbit we search for turning points along both the binding energy and total angular momentum curves for these sequences. Unlike for our previous approximate boundary condition, these two minima now coincide. We also identify the formation of cusps on the neutron star surface, indicating the onset of tidal disruption. Comparing these two critical binary separations for different mass ratios and neutron star compactions we distinguish those regions that will lead to a tidal disruption of the neutron star from those that will result in the plunge into the black hole of a neutron star more or less intact, albeit distorted by tidal forces. | Coalescing black hole-neutron star (BHNS) binaries, as well as other compact binaries composed of neutron stars and/or black holes, are among the most promising sources of gravitational waves for both ground-based \cite{LIGO,GEO,TAMA,VIRGO} and space-based laser interferometers \cite{LISA,DECIGO}. BHNS binary mergers are also candidate central engines of short-hard gamma-ray bursts (SGRBs) (see, e.g. \cite{LeeR07} and references cited therein). The remnants of both BHNS binary mergers \cite{FaberBST06,ShibaU0607} and binary neutron star mergers \cite{ShibaT06,PriceR06,OechsJ06,HMNS} are feasible progenitors for SGRBs because both may result in black holes surrounded by hot, massive accretion disks with very little, if any, baryon contamination along the polar symmetry axis. Motivated by these factors, considerable effort has gone into the study of BHNS binaries. Most approaches to date assume Newtonian gravity in either some or all aspects of the calculation (see, e.g.~\cite{Chand69,Fishb73,LaiRS93,LaiW96,TanigN96,Shiba96,UryuE99,WiggiL00,IshiiSM05,Mille05} for quasiequilibrium calculations and \cite{Mashh75,CarteL83,Marck83,LeeK99,Lee00,RosswSW04,KobayLPM04,RantsKLR07} for dynamical simulations). More recently, several groups have also studied BHNS binaries in a fully relativistic framework, both for quasiequilibrium models \cite{Mille01,BaumgSS04,TanigBFS05,TanigBFS06,Grand06,TanigBFS07} and dynamical simulations \cite{FaberBSTR06,FaberBST06,SopueSL06,LofflRA06,ShibaU0607}. Our group has pursued a systematic approach to developing increasingly realistic models of BHNS binaries in quasiequilibrium circular orbits. Our first studies \cite{BaumgSS04,TanigBFS05,FaberBSTR06,FaberBST06} assumed extreme mass ratios, i.e., black hole masses that are much greater than the neutron star mass. While this is a very natural first step from a computational point of view, binaries with comparable masses are much more interesting from the perspective of ground-based gravitational wave observations and for the launching of SGRBs. More recently we have therefore relaxed this assumption and have extended our results to the case of comparable-mass BHNS binaries \cite{TanigBFS06,TanigBFS07}. Specifically, in \cite{TanigBFS07} (hereafter Paper I) we constructed quasiequilibrium models by solving Einstein's constraint equations in the conformal thin-sandwich formalism, assuming conformal flatness and maximal slicing, together with the relativistic equations of hydrostationary equilibrium. We accounted for the black hole by excising a coordinate sphere and imposing the equilibrium black-hole boundary conditions of Cook and Pfeiffer \cite{CookP04}. This original version implemented a ``leading-order" approximation to nonspinning black holes, which equates an otherwise undetermined spin parameter $\Omega_r$ that appears in the boundary condition for the shift vector with the orbital angular velocity seen by an inertial observer at infinity, $\Omega$. As for the original irrotational binary black hole models of \cite{CookP04}, this condition does not lead to simultaneous turning points of the binding energy and the total angular momentum in constant-mass sequences in Paper I. Such simultaneous turning points are expected for those sequences if they are truly in quasiequilibrium \cite{dMOmegadJ}. An improvement over this condition, namely to iterate over $\Omega_r$ until the quasilocal spin angular momentum of the black hole vanishes, was suggested and implemented for binary black holes by \cite{CaudiCGP06}. In this paper we reconstruct quasiequilibrium models of BHNS binaries using the same techniques as in Paper I, but with the improved black hole spin angular velocity condition as suggested by \cite{CaudiCGP06}. We then compute sequences of BHNS binaries in quasicircular orbits for a wider range of neutron star masses and binary mass ratios than in Paper I, focusing our attention on irrotational neutron stars orbiting nonspinning black holes. Here we focus only on the irrotational state for the neutron star because it is astrophysically considered to be more realistic in a BHNS binary \cite{Kocha92,BildsC92,FaberBSTR06}. On the other hand, we will compute the case of spinning black holes in future work. As was the case for the irrotational black hole binaries constructed in \cite{CaudiCGP06}, we find that this improved condition for the spin parameter of the black hole $\Omega_r$ does lead to simultaneous turning points in the binding energy and the total angular momentum along constant-mass sequences. The paper is organized as follows. We briefly review the basic equations in Section II. We present numerical results in Section III, and outline some qualitative considerations concerning the fate of BHNS binaries in Section IV. In Section V we summarize our findings. Throughout this paper we adopt geometrized units with $G=c=1$, where $G$ denotes the gravitational constant and $c$ the speed of light. Latin and Greek indices denote purely spatial and spacetime components, respectively. | We have constructed new quasiequilibrium configurations of black hole-neutron star binaries in general relativity. We have solved the Einstein constraint equations in the conformal thin-sandwich formalism coupled with the equations of relativistic hydrostationary equilibrium. In Paper I, we set the spin angular velocity parameter of the black hole equal to that of the orbital angular velocity in order to produce a nonspinning black hole in the ``leading-order" approximation \cite{CookP04}, while in this paper we compute this parameter by requiring the quasilocal spin angular momentum of the black hole to be zero \cite{CaudiCGP06}. We have also improved the formulation of the gravitational field equations and obtained more accurate results than in Paper I. As an indication of the improvements in these calculations, a post-Newtonian analysis predicts smaller binary eccentricities for these new BHNS models than for those computed in Paper I (\cite{Will07}, compare \cite{BertiIW07}). In \cite{BertiIW07}, Berti {\it et al.} fit numerical results for the binding energy and angular momentum of binaries in circular orbit to post-Newtonian expressions for binaries that are not necessarily in circular orbit. Deviations between the the two approaches then lead to non-zero eccentricities in the post-Newtonian expressions. These eccentricities are smaller for our new results than for those of Paper I. We also remark on another finding of \cite{BertiIW07}, namely that for a given neutron star mass $\bar{M}_{\rm ADM,0}^{\rm NS}$ and a given value of $\Omega M_0$, the eccentricities in BHNS models, though small, are found to be larger than in binary neutron star models \cite{TanigG0203}. This suggests a larger deviation from quasiequilibrium for BHNS binaries than binary neutron stars. But, for BHNS binaries with a mass ratio of $\hat{q} = 5$, these parameters correspond to a larger binary separation than for binary neutron stars with a mass ratio of $\hat{q} = 1$ (compare Eq.~(\ref{eq:Kepler})). For similar numerical resources, this larger binary separation leads to a larger numerical error, which may explain the larger eccentricity found by \cite{BertiIW07}, at least in part. In addition to recomputing several sequences we presented in Paper I, we have constructed sequences for a wider range of neutron star masses and binary mass ratios, employing a $\Gamma=2$ polytropic neutron-star equation of state throughout. We computed several constant-mass sequences, for various mass ratios and neutron star compactions, and searched for the appearance of a cusp at the neutron star surface -- indicating the onset of tidal disruption -- and turning points on the binding energy and angular momentum curves -- identifying the ISCO. We also included some qualitative fits that allow for a simple prediction of those binary parameters separating these two different outcomes of binary coalescence. Unlike in our earlier findings, we found simultaneous turning points along the binding energy and angular momentum quasiequilibrium curves. | 7 | 10 | 0710.5169 |
0710 | 0710.0785_arXiv.txt | {\it NeXT} (New X-ray Telescope) is the next Japanese X-ray astronomical satellite mission after the {\it Suzaku} satellite. {\it NeXT} aims to perform wide band imaging spectroscopy. Due to the successful development of a multilayer coated mirror, called a supermirror, {\it NeXT} can focus X-rays in the energy range from 0.1~keV up to 80~keV. To cover this wide energy range, we are in the process of developing a hybrid X-ray camera, Wideband X-ray Imager (WXI) as a focal plane detector of the supermirror. The WXI consists of X-ray CCDs (SXI) and CdTe pixelized detectors (HXI), which cover the lower and higher X-ray energy bands of 0.1--80~keV, respectively. The X-ray CCDs of the SXI are stacked above the CdTe pixelized detectors of the HXI. The X-ray CCDs of the SXI detect soft X-rays below $\sim 10$~keV and allow hard X-rays pass into the CdTe detectors of the HXI without loss. Thus, we have been developing a ``back-supportless CCD'' with a thick depletion layer, a thinned silicon wafer, and a back-supportless structure. In this paper, we report the development and performances of an evaluation model of CCD for the SXI, ``CCD-NeXT1''. We successfully fabricated two types of CCD-NeXT1, unthinned CCDs with 625-$\mu$m thick wafer and 150-$\mu \rm m$ thick thinned CCDs. By omitting the polishing process when making the thinned CCDs, we confirmed that the polishing process does not impact the X-ray performance. In addition, we did not find significant differences in the X-ray performance between the two types of CCDs. The energy resolution and readout noise are $\sim 140$~eV (FWHM) at 5.9~keV and $\sim 5$ electrons (RMS), respectively. The estimated thickness of the depletion layer is $\sim 80 ~\mu \rm m$. The performances almost satisfy the requirements of the baseline plan of the SXI. | {\it NeXT} (New X-ray Telescope) is the sixth Japanese X-ray astronomical satellite mission, which is proposed to be launched around 2012. A Hard X-ray Telescope (HXT), supermirror onboard {\it NeXT} (see Tawara {\it et~al.} (2003) \cite{tawara03} and references therein), has a large collecting area for X-rays in the energy from 0.1 to 80~keV. In particular, the HXT has a high reflectivity even in the hard X-ray band above 10~keV. Previous satellite missions have not had X-ray focusing optics capable of observations in this band. {\it NeXT} is designed to be the first to perform imaging and spectroscopic observations in the energy band above 10~keV. In order to meet the energy range covered by the HXT, we have been developing a Wideband X-ray Imager (WXI). The first successful space flight use of X-ray CCDs as photon counting and spectroscopic imagers was the SIS aboard $ASCA$ \cite{byrke91}. Since then, X-ray CCDs have become standard focal plane detectors for X-ray telescopes in the X-ray energy band of 0.1--10~keV, and have been adopted as the principal detectors of recent X-ray observatories such as the ACIS of {\it Chandra} \cite{garmire03}, the EPIC of {\it XMM-Newton} \cite{struder01,turner01}, and the XIS of {\it Suzaku} \cite{koyama07} because X-ray CCDs have well balanced spectroscopic, imaging, and time resolution performances. However, to achieve a quantum efficiency of 10\% for X-rays with an energy of 40~keV, a $\sim 1000~{\rm \mu m}$ depletion layer is required, which is nearly impossible. A detector with a high-Z material is essential for observations in the hard X-ray band above 10--20~keV. On the other hand, the performances of imaging and spectroscopy below 10~keV of high-Z solid detectors such as CdTe detectors are poorer than those of X-ray CCDs, suggesting that a single detector cannot cover the entire 0.1--80~keV band with the best X-ray performance. Thus, we have been developing a hybrid camera, the Wide band X-ray Imager (WXI), which combines an X-ray CCD and a CdTe pixelized detector \cite{Takahashi1999,tsuru01,tsuru04,takahashi04}. Holland (2003) \cite{adholland03} has also reported the first laboratory demonstration of such a hybrid detector with a thinned X-ray CCD, which is operated in front of CZT detector. The WXI consists of two sub-instruments; the Soft X-ray Imager (SXI) and the Hard X-ray Imager (HXI); overviews can be found in Tsuru {\it et~al.} (2005) \cite{tsuru05} and Takahashi {\it et~al.} (2004) \cite{takahashi04s}, respectively. The SXI and HXI are the upper and lower parts of the WXI, respectively. The SXI consists of X-ray CCDs with a thick depletion layer for the lower energy band below 10--20~keV. The HXI is based on CdTe pixelized detectors, which cover the hard X-rays above 10--20~keV. We have developed a new type of CCD for the SXI, a ``Back-Supportless CCD'' (BS-CCD), in which the back supporting package under the imaging area of the CCD is removed. Most X-rays absorbed in the field-free region of the CCD are undetected and lost. Hence, we also removed the field-free region as much as possible. The BS-CCD of the SXI is placed over the CdTe pixelized detectors of the HXI. Soft X-rays are detected in the BS-CCD, while hard X-rays penetrate through the BS-CCD and are detected by the CdTe pixelized detectors. Thus, both soft and hard X-rays are detected without loss. As previously reported by Tsuru {\it et~al.} (2005) \cite{tsuru05} in detail, we have been developing a BS-CCD for the SXI following two plans of a rather conservative ``baseline plan'' and an innovative ``goal plan'' in parallel. Table~\ref{tab:SXIgoala} shows the specifications of the BS-CCD of the two plans along with those of the XIS \cite{koyama07}, which is one of the most excellent X-ray CCDs currently in orbit. In the goal plan, we realize a fully-depleted back-illuminated type of BS-CCD with a very thick depletion layer of $\sim 200~{\rm \mu m}$ by adopting a p-channel device. We have already successfully fabricated test devices with a full depletion layer, which was $\sim$ 200-$\mu{\rm m}$ thick. The details and status of the developments are reported elsewhere \cite{kamata04,takagi05,kamata06,matuura06,takagi06,tsuru06}. In the baseline plan, we developed BS-CCDs based on a natural extension of our successful developments of CCD-CREST/CREST2 and MAXI-CCD in order to minimize the risk involved with their development \cite{bamba01,tsunemi05,tomida00,miyata02}. We adopted a front-illuminated type of CCD mainly due to the manufacturing process \cite{takagi05}. The thickness of the depletion layer is designed to be 70--80$~{\rm \mu m}$ or more. We have already successfully developed a small test model of BS-CCD and confirmed that the thinning processes of the wafer and the back-supportless structure do not degrade the performance \cite{takagi05}. After the successfully developing the small test model, we constructed an evaluation model, ``CCD-NeXT1''. Following CCD-NeXT1, we will develop a flight model, ``CCD-NeXT2'', which matches the specifications of the SXI shown in Table~1. In this paper, we report the development and the performance of CCD-NeXT1\footnote{Note that part of the results reported herein have already been reported as a contributing paper of a SPIE conference \cite{OzawaSPIE06_CCD-NeXT1}.}. | \begin{itemize} \item Following the successful development of the test models of BS-CCD for SXI onboard the {\it NeXT} satellite, we developed an evaluation model, CCD-NeXT1. We fabricated two types of CCD-NeXT1. One was a type of un-thinned CCD with 625-$\mu$m thick wafers. The other type was a thinned CCD with 150-$\mu \rm m$ thickness. \item We processed thinned CCD-NeXT1 devices by omitting the polishing process in the thinning process. The evaluation of the devices confirmed that omitting the polishing process did not impact on the X-ray performance. \item We did not observe a significant difference in the X-ray performance of the unthinned and the thinned devices. The energy resolution and the readout noise were $\sim$140~eV (FWHM) at 5.9~keV and $\sim$5 electrons (RMS), respectively. The detection efficiency of the $\rm ^{109}Cd$ photons indicates that the depletion layer is $\sim$ 80-${\rm \mu m}$ thick. This performance meets the requirements for the baseline plan of the SXI. \end{itemize} | 7 | 10 | 0710.0785 |
0710 | 0710.2780_arXiv.txt | }[2]{{\footnotesize\begin{center}ABSTRACT\end{center} \vspace{1mm}\par#1\par \noindent {~}{\it #2}}} \newcommand{\TabCap}[2]{\begin{center}\parbox[t]{#1}{\begin{center} \small {\spaceskip 2pt plus 1pt minus 1pt T a b l e} \refstepcounter{table}\thetable \\[2mm] \footnotesize #2 \end{center}}\end{center}} \newcommand{\TableSep}[2]{\begin{table}[p]\vspace{#1} \TabCap{#2}\end{table}} \newcommand{\FigCap}[1]{\footnotesize\par\noindent Fig.\ % \refstepcounter{figure}\thefigure. #1\par} \newcommand{\TableFont}{\footnotesize} \newcommand{\TableFontIt}{\ttit} \newcommand{\SetTableFont}[1]{\renewcommand{\TableFont}{#1}} \newcommand{\MakeTable}[4]{\begin{table}[p]\TabCap{#2}{#3} \begin{center} \TableFont \begin{tabular}{#1} #4 \end{tabular}\end{center}\end{table}} \newcommand{\MakeTableTop}[4]{\begin{table}[t]\TabCap{#2}{#3} \begin{center} \TableFont \begin{tabular}{#1} #4 \end{tabular}\end{center}\end{table}} \newcommand{\MakeTableSep}[4]{\begin{table}[p]\TabCap{#2}{#3} \begin{center} \TableFont \begin{tabular}{#1} #4 \end{tabular}\end{center}\end{table}} \newcommand{\TabCapp}[2]{\begin{center}\parbox[t]{#1}{\centerline{ \small {\spaceskip 2pt plus 1pt minus 1pt T a b l e} \refstepcounter{table}\thetable} \vskip2mm \centerline{\footnotesize #2}} \vskip3mm \end{center}} \newcommand{\MakeTableSepp}[4]{\begin{table}[p]\TabCapp{#2}{#3}\vspace*{-.7cm} \begin{center} \TableFont \begin{tabular}{#1} #4 \end{tabular}\end{center}\end{table}} \newfont{\bb}{ptmbi8t at 12pt} \newfont{\bbb}{cmbxti10} \newfont{\bbbb}{cmbxti10 at 9pt} \newcommand{\uprule}{\rule{0pt}{2.5ex}} \newcommand{\douprule}{\rule[-2ex]{0pt}{4.5ex}} \newcommand{\dorule}{\rule[-2ex]{0pt}{2ex}} \def\thefootnote{\fnsymbol{footnote}} \newenvironment{references}% { \footnotesize \frenchspacing \renewcommand{\thesection}{} \renewcommand{\in}{{\rm in }} \renewcommand{\AA}{Astron.\ Astrophys.} \newcommand{\AAS}{Astron.~Astrophys.~Suppl.~Ser.} \newcommand{\ApJ}{Astrophys.\ J.} \newcommand{\ApJS}{Astrophys.\ J.~Suppl.~Ser.} \newcommand{\ApJL}{Astrophys.\ J.~Letters} \newcommand{\AJ}{Astron.\ J.} \newcommand{\IBVS}{IBVS} \newcommand{\PASP}{P.A.S.P.} \newcommand{\Acta}{Acta Astron.} \newcommand{\MNRAS}{MNRAS} \renewcommand{\and}{{\rm and }} {Period--luminosity (PL) relations of variable red giants in the Large (LMC) and Small Magellanic Clouds (SMC) are presented. The PL diagrams are plotted in three planes: $\log P$--$K_S$, $\log P$--$W_{JK}$, and $\log P$--$W_I$, where $W_{JK}$ and $W_I$ are reddening free Wesenheit indices. Fourteen {\it PL} sequences are distinguishable, and some of them consist of three closely spaced ridges. Each of the sequences is fitted with a linear or quadratic function. The similarities and differences between the {\it PL} relations in both galaxies are discussed for four types of red giant variability: OGLE Small Amplitude Red Giants (OSARGs), Miras and Semiregular Variables (SRVs), Long Secondary Periods (LSPs) and ellipsoidal variables. We propose a new method of separating OSARGs from non-variable stars and SRVs. The method employs the position in the reddening-free {\it PL} diagrams and the characteristic period ratios of these multiperiodic variables. The {\it PL} relations for the LMC OSARG are compared with the calculated relations for RGB models along isochrones of relevant ages and metallicities. We also compare measured periods and amplitudes of the OSARGs with predictions based on the relations valid for less luminous solar-like pulsators. Miras and SRVs seem to follow {\it PL} relation of the same slopes in the LMC and SMC, while for LSP and ellipsoidal variables slopes in both galaxies are different. The {\it PL} sequences defined by LSP variables and binary systems overlap in the whole range of analyzed wavebands. We put forward new arguments for the binary star scenario as an explanation of the LSP variability and elaborate on it further. The measured pulsation to orbital period ratio implies nearly constant ratio of the star radius to orbital distance, $R/A\approx0.4$, as we find. Combined effect of tidal friction and mass loss enhanced by the low-mass companion may explain why such a value is preferred.}{Stars: AGB and post-AGB -- Stars: late-type -- Stars: oscillations -- Magellanic Clouds} | The first attempt to determine period--luminosity (PL) relation for long period variables (LPVs) was made by Gerasimovi\v{c} (1928), who noticed that Mira stars with longer periods are on average fainter at visual wavelengths. This result has been confirmed by subsequent studies (\eg Wilson and Merrill 1942, Osvalds and Risley 1961, Clayton and Feast 1969), however the scatter of this period--luminosity dependence turned out to be very large. First tight {\it PL} relation for LPVs was discovered for Mira stars at near infrared (NIR) wavebands (Glass and Lloyd Evans 1981). This {\it PL} law, based on only 11 Miras in the LMC, was refined by extensive studies of Feast \etal (1989) and Hughes and Wood (1990). The second, parallel {\it PL} sequence, occupied by semiregular variables (SRVs), was identified by Wood and Sebo (1996). This sequence was shifted relative to the Miras' ridge toward shorter periods by a factor of two. However, the subject of {\it PL} distribution of LPVs has progressed rapidly over recent years when large microlensing surveys (MACHO, OGLE, EROS, MOA) published long-term photometry of huge number of stars. Complex structure of the {\it PL} distribution was demonstrated for the first time by Cook \etal (1997), who published {\it PL} diagram for variable stars detected during the MACHO survey in the LMC. A series of three or four {\it PL} sequences defined by LPVs can be distinguished in that diagram. Sharper picture was presented by Wood \etal (1999), who distinguished and described five {\it PL} sequences (denoted as A--E) in the period--Wesenheit index plane. Wood (2000) showed similar distribution in the $\log P$--$K$ diagram. These results were then confirmed by many studies based on observations originated in various sources (Cioni \etal 2001, 2003, Noda \etal 2002, Lebzelter \etal 2002, Ita \etal 2004, Groenewegen 2004, Fraser \etal 2005). Kiss and Bedding (2003, 2004) used OGLE data to reveal new features in the {\it PL} distribution. They noticed that sequence B consists of two closely spaced parallel ridges (Ita \etal 2004 denoted the additional sequence as C$'$). Below the tip of the red giant branch (TRGB) Kiss and Bedding (2003) found three sequences shifted in $\log P$ relative to stars brighter than TRGB. It was the definitive proof that stars in the first ascent Red Giant Branch (RGB) pulsate similarly to objects being in the Asymptotic Giant Branch (AGB) phase. The Optical Gravitational Lensing Experiment (OGLE) collected unprecedented amount of photometric data of stars in the Large and Small Magellanic Clouds. Both galaxies have been constantly monitored since 1997 and at present time this is the best and longest available photometric dataset for analyzing huge number of variable red giants. Our studies on LPVs resulted in many discoveries, regarding also the {\it PL} relations. Soszy{\'n}ski \etal (2004a) showed that OGLE Small Amplitude Red Giants (OS\-ARGs) constitute separate class of variable stars, with different structure in the {\it PL} plane than ``classical'' SRVs and Miras. We indicated two previously overlooked {\it PL} relations -- the longest (${\rm a}_1$) and the shortest (${\rm a}_4$) period sequences followed by AGB OSARGs. We also suggested a method of empirical division between RGB and AGB OSARGs fainter than TRGB. Red giants revealing ellipsoidal modulation caused by binarity were analyzed by Soszy{\'n}ski \etal (2004b). It was shown that, if true orbital periods are considered, the {\it PL} relation of ellipsoidal variables (sequence~E) is a direct continuation of sequence~D occupied by mysterious Long Secondary Period (LSP) variables. This is a hint that the LSP phenomenon may be related to binarity, but taking into account available radial velocity measurements, the secondary component usually must be a low mass object, possibly former planet. This idea was supported by Soszy{\'n}ski (2007) who discovered in some LSP variables ellipsoidal-like and eclipsing-like modulations with periods equal to LSPs. In Soszy{\'n}ski \etal (2005) we again increased complexity of the PL distribution of LPVs. Each of the NIR {\it PL} sequences C$'$, C and D in the LMC (occupied by SRV, Miras and LSP variables) split into two separate ridges in the period -- optical Wesenheit index plane, what corresponds to the spectral division into oxygen-rich (O-rich) and carbon-rich (C-rich) AGB stars. Thus, we found a new photometric method of distinguishing between these two populations. In this paper we describe in details the {\it PL} relations of variable red giants in both Magellanic Clouds. We show new details in the {\it PL} plane and compare {\it PL} distribution in the LMC and SMC. The paper is organized as follows. Section~2 gives details of the observations and data reduction. In Section~3 the {\it PL} relations are presented with a description of their derivation. A discussion about four types of red giant variability -- OSARGs, Miras/SRVs, LSPs and ellipsoidal modulation -- is given in Sections~4--7. Section~8 summarizes and concludes the paper. | In this paper we showed the most complex structure of the {\it PL} distribution presented so far. The Wood's five ridges turn out to be an overlap of fourteen sequences (if consider closely spaced {\it PL} relations of OSARGs, the number of sequences exceeds twenty). In order to help recognizing {\it PL} relations with published {\it PL} laws we provide Table~3 with appropriate identifications. Diagrams employing period ratios (similar to the Petersen diagram) were used as a tool for discriminating OGLE Small Amplitude Red Giants (OSARGs). We compared three sequences of the {\it PL} relations for the LMC OSARGs in the RGB phase with the calculated relation for first three radial modes using isochrone calculations. We found an essential agreement with our knowledge about metallicities and ages of red giants population in the LMC. However, there are also discrepancies which may suggest the necessity of refinement of the knowledge and/or stellar models. This is a potential application of the OSARGs. \MakeTableTop{l@{\hspace{4pt}}c@{\hspace{4pt}} l@{\hspace{8pt}} c@{\hspace{8pt}} c@{\hspace{8pt}} c@{\hspace{8pt}}}{12.5cm} {Labels of the {\it PL} relations in this and previous papers} {\hline \noalign{\vskip3pt} \multicolumn{3}{c}{this paper} & Wood \etal 1999 & Kiss and Bedding 2003 & Ita \etal 2004 \\ \noalign{\vskip3pt} \hline \noalign{\vskip3pt} & & ${\rm b}_1$ & & $R_1$ & \\ & RGB & ${\rm b}_2$ & B & $R_2$ & $B^-$ \\ & & ${\rm b}_3$ & A & $R_3$ & $A^-$ \\ OSARGs & & ${\rm a}_1$ & & & \\ & & ${\rm a}_2$ & B & 2O & $B^+$ \\[-1ex] & \raisebox{1.5ex}{AGB} & ${\rm a}_3$ & A & 3O & $A^+$ \\ & & ${\rm a}_4$ & & & \\ \noalign{\vskip3pt} \hline \noalign{\vskip3pt} & & C$_{\rm O}$ & C & F & C \\[-1ex] Miras & \raisebox{1.5ex}{O-rich} & C$'_{\rm O}$ & B & 1O & C$'$ \\ and SRVs & & C$_{\rm C}$ & C & F & C \\[-1ex] & \raisebox{1.5ex}{C-rich} & C$'_{\rm C}$ & B & 1O & C$'$ \\ \noalign{\vskip3pt} \hline \noalign{\vskip3pt} & O-rich & D$_{\rm O}$ & D & $L_2$ & D \\[-1ex] \raisebox{1.5ex}{LSPs} & C-rich & D$_{\rm C}$ & D & $L_2$ & D \\ \noalign{\vskip3pt} \hline \noalign{\vskip3pt} \multicolumn{2}{l}{Ell. and Ecl.} & E & E$^*$ & $L_1^*$ & E$^*$ \\ \noalign{\vskip3pt} \hline \multicolumn{6}{l}{$^*$ -- the sequence is shifted due to halving the orbital periods.} } Most likely mechanism responsible for pulsation in the OSARGs is the stochastic excitation. Therefore we looked at these objects as solar-like pulsators. The range of periods agrees with predictions based on extrapolation of the relations found for much less luminous stars. The amplitudes are lower than predicted by the Kjeldsen and Bedding (1995) formula, where the the amplitude rises linearly with the luminosity-to-mass ratio. However, there are calculations predicting a slower amplitude rise. Testing theory of stochastic excitation is another possible application of OSARGs. We would like to bring the reader's attention to NIR Wesenheit index ($W_{JK}$). The sequences in the period--$W_{JK}$ plane are generally better defined than those at $K_S$ magnitudes. Wesenheit index is a reddening independent quantity, so even heavy reddened Miras fall on the {\it PL} sequence. Therefore, the period--$W_{JK}$ relations can be used as a distance indicators without correcting them for reddening. Moreover, O- and C-rich Miras and SRV obey very similar relations in the period--$W_{JK}$ plane, while at $K_S$ the {\it PL} relations are significantly different for both populations. We conclude that $W_{JK}$ index can be a useful tool for studying LPVs. Comparison of the {\it PL} relations in NIR supports the binary star scenario as the explanation of the LSP variability. We further develop this scenario taking into account data on the ratios of the short period (pulsational) and LSP variability. For OSARGs these ratio is nearly constant which translates into nearly constant ratio of the stellar to orbital radius the value of about 0.4. We proposed that the small mass companion position is determined by the balance between mass loss and tidal effects. This scenario requires that the proximity of the companion enhances mass loss. Moreover, it requires that a substantial fraction (majority) of red giants have such companions and that their mass cannot be too small. \vspace*{9pt} \Acknow{The paper was supported by the Foundation for Polish Science through the Homing Program and by MNiSW grants: 1P03D01130 and N20303032/4275. This publication makes use of data products from the Two Micron All Sky Survey, 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.} \vspace*{9pt} | 7 | 10 | 0710.2780 |
0710 | 0710.5505_arXiv.txt | We analyze a toy swiss-cheese cosmological model to study the averaging problem. In our swiss-cheese model, the cheese is a spatially flat, matter only, Friedmann-Robertson-Walker solution (\textit{i.e.,} the Einstein--de Sitter model), and the holes are constructed from a Lema\^{\i}tre-Tolman-Bondi solution of Einstein's equations. We study the propagation of photons in the swiss-cheese model, and find a phenomenological homogeneous model to describe observables. Following a fitting procedure based on light-cone averages, we find that the the expansion scalar is unaffected by the inhomogeneities (\textit{i.e.,} the phenomenological homogeneous model is the cheese model). This is because of the spherical symmetry of the model; it is unclear whether the expansion scalar will be affected by non-spherical voids. However, the light-cone average of the density as a function of redshift is affected by inhomogeneities. The effect arises because, as the universe evolves, a photon spends more and more time in the (large) voids than in the (thin) high-density structures. The phenomenological homogeneous model describing the light-cone average of the density is similar to the $\Lambda$CDM concordance model. It is interesting that although the sole source in the swiss-cheese model is matter, the phenomenological homogeneous model behaves as if it has a dark-energy component. Finally, we study how the equation of state of the phenomenological homogeneous model depends on the size of the inhomogeneities, and find that the equation-of-state parameters $w_{0}$ and $w_{a}$ follow a power-law dependence with a scaling exponent equal to unity. That is, the equation of state depends linearly on the distance the photon travels through voids. We conclude that within our toy model, the holes must have a present size of about $250$ Mpc to be able to mimic the concordance model. | Most, if not all, observations are consistent with the cosmic concordance model, according to which one-fourth of the present mass-energy of the universe is clustered and dominated by cold dark matter (CDM). The remaining three-quarters is uniform and dominated by a fluid with negative pressure (dark energy, or $\Lambda$). While the standard $\Lambda$CDM model seems capable of accounting for the observations, 95\% of the mass-energy of the present universe is unknown. This is either a feature, and we are presented with the opportunity of discovering the nature of dark matter and dark energy, or it is a bug, and nature might be different than described by the $\Lambda$CDM model. Regardless, until such time as dark matter and dark energy are completely understood, it is useful to look for alternative cosmological models that fit the data. One non-standard possibility is that there are large effects on the {\it observed} expansion rate (and hence on other observables) due to the back-reaction of inhomogeneities in the universe (see, \textit{e.g.,} Ref.\ \cite{kmr,notari,rasa, buchert} and references therein). The basic idea is that all evidence for dark energy comes from the observational determinations of the expansion history of the universe. Anything that affects the observed expansion history of the universe alters the determination of the parameters of dark energy; in the extreme it may remove the need for dark energy. The ``safe'' consequence of the success of the concordance model is that the isotropic and homogeneous $\Lambda$CDM model is a good {\it phenomenological} fit to the real inhomogeneous universe. And this is, in some sense, a verification of the cosmological principle: the inhomogeneous universe can be described by means of an isotropic and homogeneous solution. However, this does not imply that a primary source of dark energy exists, but only that it exists as far as the phenomenological fit is concerned. For example, it is not straightforward that the universe is accelerating. If dark energy does not exist at a fundamental level, its presence in the concordance model would tell us that the pure-matter inhomogeneous model has been renormalized, from the phenomenological point of view (luminosity-distance and redshift of photons), into a homogeneous $\Lambda$CDM model. The issue is the observational significance of the back-reaction of inhomogeneities. Our point of view is tied to our past light cone: we focus on the effects of large-scale nonlinear inhomogeneities on observables such as the luminosity-distance--redshift relation. We will not discuss averaged domain dynamics, even though if we think it is a crucial step in understanding how General Relativity effectively works in a lumpy universe \cite{ellis, buchert_new}. Following this approach, we built in Ref.\ \cite{marra-sc} a particular swiss-cheese model, where the cheese consists of a spatially flat, matter only Friedmann-Robertson-Walker (FRW) solution and the holes are constructed out of a Lema\^{i}tre-Tolman-Bondi (LTB) solution of Einstein's equations. We attempted to find a model that was solvable and ``realistic'' (even if still toy), rather than finding a model with interesting volume-averaged dynamics. The model, however, will turn out to be useful to investigate light-cone averages. It has been indeed shown that the LTB solution can be used to fit the observed $d_{L}(z)$ without the need of dark energy (for example, see Ref.\ \cite{alnes0602}). To achieve this result, however, it is necessary to place the observer at the center of a rather large-scale underdensity. To overcome this fine-tuning problem we built a swiss-cheese model with the observer in the cheese looking through a series of holes. In Ref.\ \cite{marra-sc} we studied this model in detail and discussed the effects of large-scale nonlinear inhomogeneities on observables such as the luminosity-distance--redshift relation. We found that inhomogeneities are able (at least partly) to mimic the effects of dark energy. In this paper we will analyze the same swiss-cheese model through the fitting scheme developed by Ellis and Stoeger \cite{ellis-f} in order to better understand how inhomogeneities renormalize the (matter only) swiss-cheese model allowing us to avoid a physical dark-energy component. We think that this model fits well in that context and therefore we might be able to shed some light on the important topics discussed there. We will propose a fitting procedure based on light-cone averages. The paper is organized as follows: In Sec.\ \ref{model} we will specify the parameters of our swiss-cheese model and summarize the main results obtained in Ref.\ \cite{marra-sc}. In Sec.\ \ref{fitti}, we develop our fitting procedure, and in Sec.\ \ref{disco} we discuss our results. Then, in Sec.\ \ref{dressing} we study the dependence of the best-fit parameters on the size of the holes. Conclusions are given in Sec.\ \ref{conclusions}. | \label{conclusions} The aim of this investigation was to understand the role of large-scale non-linear cosmic inhomogeneities in the interpretation of observational data. We focused on an exact (if toy) solution, based on the Lema\^{\i}tre-Tolman-Bondi (LTB) model. This solution has been studied extensively in the literature \cite{alnes0607, notari-mansouri, alnes0602, celerier, mansouri, flanagan, rasanen, tomita, chung, nambu}. It has been shown that it can be used to fit the observed $d_{L}(z)$ without the need of dark energy (for example in Ref.\ \cite{alnes0602}). To achieve this result, however, it is necessary to place the observer at the center of a rather large-scale underdensity. To overcome this fine-tuning problem we built a swiss-cheese model, placing the observer in the cheese and having the observer look through the holes in the swiss-cheese as pictured in Fig.\ \ref{schizzo}. In Sec.\ \ref{model} we defined the model and described its dynamics: it is a swiss-cheese model where the cheese is made of the usual FRW solution and the holes are made of a LTB solution. The voids inside the holes are expanding faster than the cheese. We reported also the results for $d_{L}(z)$ obtained in Ref.\ \cite{marra-sc}, to which we refer the reader for a more thorough analysis. We found that redshift effects are suppressed because of a compensation effect due to spherical symmetry. However, we found interesting effects in the calculation of the angular distance: the evolution of the inhomogeneities bends the photon path compared to the FRW case. Therefore, inhomogeneities will be able (at least partly) to mimic the effects of dark energy. After having analyzed the model from the observational point of view, we set up in Section \ref{fitti} the fitting problem in order to better understand how inhomogeneities renormalize the matter swiss-cheese model allowing us to eschew a primary dark energy. We followed the scheme developed in Ref.\ \cite{ellis-f}, but modified in the way to fit the phenomenological model to the swiss-cheese one. We chose a method that is intermediate between the fitting approach and the averaging one: we fitted with respect to light-cone averages. In particular, we focused on the expansion and the density. While the expansion behaved as in the FRW case because of the compensation effect mentioned above, we found that the density behaved differently thanks to its intensiveness to that compensation effect: a photon is spending more and more time in the (large) voids than in the (thin) high density structures. This effect is not directly linked to the one giving us an interesting $d_{A}$. The best fit we found for holes of $r_{h}=250$ Mpc is $w_{0}=-1.03$ and $w_{a}=2.19$; qualitatively similar to the concordance model. The flow chart of Fig.\ \ref{schema} summarizes the results obtained. The insensitivity to the compensation effect made us think that a swiss cheese made of spherical symmetric holes and a swiss cheese without an exact spherical symmetry would share the same light-cone averaged density. Knowing the behavior of the density we are therefore able to know the one of the Hubble parameter that will be the one of the FRW solution with an phenomenological source characterized by the fit equation of state. In this way we can think to go beyond the main limitation of this model, that is, the assumption of spherical symmetry. From this point of view, the light-cone averaged density can be seen as a tool in performing this step. Summarizing: \begin{itemize} \item We started with a swiss-cheese model based on spherically symmetric holes. A photon, during its journey through the swiss cheese, undergoes a redshift which is not affected by inhomogeneities. However the photon is spending more and more time in the voids than in the structures. The lack of an effect is due to the the assumption of spherical symmetry. We focus on this because a photon spending most of its time in voids should have a different redshift history than a photon propagating in a homogeneous background. \item Assuming that the density is a quantity that does not heavily depend on the assumption of spherical symmetry, we tried to resolve the issue by focusing on the density alone and getting from it the expansion (and therefore the redshift history). \item This resulted in a swiss-cheese model with holes that effectively are not perfectly spherical. In this model the redshift history of a photon depends on the time passed inside the voids. \item In practice this means that we will use the phenomenological best-fit model found, that is, we will use a model that behaves similarly to the concordance model. \end{itemize} Then, in Section \ref{dressing} we studied how the equation of state of a phenomenological model with only one effective source depends on the size of the inhomogeneity. We found that $w^{R}_{0}$ and $w^{R}_{a}$ follow a power-law dependence with the same scaling exponent which is equal to unity. That is, the equation of state depends linearly on the distance the photon travels through voids. We finally asked which size of the holes will give us a phenomenological model able to mimic the concordance model. We found that for $n=1$, that is for a holes of radius $r_{h}=250$ Mpc, we have $w_{0}=-1.03$ and $w_{a}=2.19$. | 7 | 10 | 0710.5505 |
0710 | 0710.5219_arXiv.txt | {} {The analysis of near infrared spectropolarimetric data at the internetwork at different regions on the solar surface could offer constraints to reject current modeling of these quiet areas.} {We present spectro-polarimetric observations of very quiet regions for different values of the heliocentric angle for the Fe\,{\sc i} lines at $1.56$ $\mu$m, from disc centre to positions close to the limb. The spatial resolution of the data is $0.7-1''$. We analyze direct observable properties of the Stokes profiles as the amplitude of circular and linear polarization as well as the total degree of polarization. Also the area and amplitude asymmetries are studied.} {We do not find any significant variation of the properties of the polarimetric signals with the heliocentric angle. This means that the magnetism of the solar internetwork remains the same regardless of the position on the solar disc. This observational fact discards the possibility of modeling the internetwork as a Network-like scenario. The magnetic elements of internetwork areas seem to be isotropically distributed when observed at our spatial resolution.} {} | The presence of magnetic fields in the solar internetwork was discovered more than 30 years ago \citep{livingston_75,smithson_75}. Since that work, improvements on the instrumentation have allowed the observation of the full Stokes vector in those regions with a signal-to-noise ratio good enough to retrieve the magnetic field vector from the observational data. Nevertheless, the internetwork magnetic topology remains yet very indistinct and shows a growing complexity as we improve the quality of the data. \cite{martin_87} observed that in her videomagnetographs at a spatial resolution of $3''$ the longitudinal component of internetwork magnetic fields was present everywhere in the solar disc. She immediately concluded that these magnetic structures should have very small scales and a very tangled geometry, giving as an example a scenario in which the magnetic fields in the internetwork would consist of a maze of small loops. As spectro-polarimeters have made possible the precise detection of these signals, the efforts have concentrated on the study of the distribution of magnetic fields at disc centre. Different methodologies (mainly based upon the use of the Zeeman and Hanle effects) shed light on different aspects of the internetwork magnetism. Concerning the Zeeman effect, the Stokes $Q$, $U$ and $V$ signals detected on the most widely used spectral lines (Fe\,{\sc i} at $1.5$ $\mu$m and 630 nm) are very weak at the best spatial resolutions of 0.5-1$''$. Moreover, the Stokes $I$ profile seems to come from a field free atmosphere, as expected from weakly polarized media \citep{jorge_99}. The filling factor of the magnetic elements retrieved from the analysis of these signals is always around 2 \% \citep{khomenko_03, jorge_ita_03, marian_spw4}. The rest of the resolution element would be filled with very weak magnetic fields. Another possibility is that the real element is filled with mixed polarity magnetic fields which would partially cancel out due to the lack of spatial resolution. The magnetic field strength distributions recovered for this 2 \% of the resolution element appear to show a preference for magnetic fields around the equipartition field at photospheric heights and even weaker \citep{khomenko_03, marian_spw4, julio07}. Additionally, \cite{andres_07} have shown the first direct observational evidence of flux cancellation in the internetwork (note that their internetwork region is surrounded by a very enhanced network structure), showing that more than 95 \% of the magnetic flux is cancelled in the resolution element ($\sim 1''$). This means that the above-mentioned distributions could not give a complete vision of the internetwork magnetism. \cite{andres_07} give an amount of $250$ G for the mean magnetic field in the resolution element (note that the values of the magnetic flux density obtained by means of the Zeeman effect are below 10 Mx/cm$^2$). This would mean that the magnetism of the internetwork could play an important role on the solar global magnetism. This has also been pointed out by works using the Hanle effect \citep{javier_04}. The study of regions in different positions of the solar disc could represent a strong constraint to reject models for the solar internetwork. Using the 630 nm lines with a spatial resolution of $\sim 1''$, \cite{lites_02} built a histogram of both circular and linear polarization signals in two quiet regions, one at disc centre and another at $\mu=0.82$ (being $\mu$ the cosine of the heliocentric angle). He did not observe significant linear polarization signals in neither of the two regions. However, Figure 9 in \cite{lites_02} shows that the histograms of circular polarization do not present any variation for those signals whose integrated signal is below 0.005. \cite{meunier_98}, studying integrated polarization, and \cite{harvey_07} using magnetograms, show the presence of a horizontal component of the magnetic field everywhere in the solar disc. In this paper we present the first study of the internetwork at several positions on the solar disc using high quality $0.8''$ spectro-polarimetric data. | We have analyzed high quality spectro-polarimetric data of the Fe\,{\sc i} lines at $1.56$ $\mu$m in order to infer information about the internetwork magnetism at different positions on the Sun's surface. The whole field of view presents significant signal, meaning that the magnetic fields pervade the observed areas. We have found that the circular and linear polarization amplitudes do not have any clear dependence on the heliocentric angle. This fact goes against a Network-like scenario for the internetwork: quasi--vertical flux tubes cannot explain this observational result, nor in fact any field topology with a preferred orientation within the field-of-view. An isotropical distribution of magnetic fields, oriented in all directions in the whole field of view, is on the other hand expected to show this behaviour. \cite{marian_07} found that at least 10-20 \% of the magnetic flux in the internetwork is connected by low-lying loops. Consequently, the scenario proposed by \cite{martin_87} of an internetwork characterized by a myriad of small loops is a very reasonably idea that is compatible with all the observational constraints presented in this work. Of course one can think of other scenarios that are compatible with the observations. \citep{stenflo_87, rafa_04, javier_04} adopt turbulent internetwork magnetism, \cite{jorge_00} proposes a micro-structuration of the atmosphere (MISMA) to explain all the magnetic phenomena on the solar surface. All of them are compatible with the presented results and our efforts should be headed towards finding more constraints to reject some of them and strengthen others. The size of the magnetic structures can also be constrained by the information presented in this study. First, improving the spatial resolution, we do not see a global increase in the signals. \cite{lites_04} found no increment of the magnetic flux density from $1"$ to $0.6"$. Recently \cite{lites_07} computed a magnetic flux density of about $11$ Mx/cm$^2$ at 0.3$''$ using HINODE's data, which is compatible with the value of $10$ Mx/cm$^2$ found by \cite{martin_87} at $3''$. This means that, either the magnetic field structures are already resolved at $\sim 0.5''$ or we are very far from resolving them. The fact that the polarimetric signals do not vary along the solar surface would point towards very small structures as the responsibles for the internetwork magnetism. The size of these magnetic structures is something not yet constrained by the observations. We have presented a study of high quality spectro-polarimetric data in different positions of the solar surface, from the disc centre towards $\mu=0.28$. This is the first step of the study of the variation on the magnetism of the internetwork with the heliocentric angle. Much more work has to be done by retrieving physical parameters as the magnetic field strength vector, magnetic flux, etc. to really constrain the modeling of the internetwork and reject models that are not compatible with the results. | 7 | 10 | 0710.5219 |
0710 | 0710.4134_arXiv.txt | Since 1999, a radial velocity survey of 179 red giant stars is ongoing at Lick Observatory with a one month cadence. At present $\sim$20$-$100 measurements have been collected per star with an accuracy of 5 to 8 m\,s$^{-1}$. Of the stars monitored, 145 (80\%) show radial velocity (RV) variations at a level $>$20 m\,s$^{-1}$, of which 43 exhibit significant periodicities. Here, we investigate the mechanism causing the observed radial velocity variations. Firstly, we search for a correlation between the radial velocity amplitude and an intrinsic parameter of the star, in this case surface gravity ($\log g$). Secondly, we investigate line profile variations and compare these with theoretical predictions. | Since 1999, a radial velocity survey of 179 red giant stars is ongoing at Lick Observatory, using the 60 cm Coud\'e Auxiliary Telescope (CAT) in conjunction with the Hamilton echelle spectrograph (R $\approx$ 60\,000). These stars have been selected from the Hipparcos catalogue \cite{esa1997}, based on the criteria described by \cite{frink2001}. The selected stars are all brighter than 6~mag, are presumably single and have photometric variations $< 0.06$~mag in V. The system with an iodine cell in the light path has been developed as described by \cite{marcy1992} and \cite{valenti1995}. With integration times of up to thirty minutes for the faintest stars ($m_{v}$ = 6 mag) we reach a signal to noise ratio of about $80-100$ at $\lambda = 5500$ \AA , yielding a radial velocity precision of $5-8$ m\,s$^{-1}$. The initial aim of the survey was to check whether red giants would be stable enough to serve as reference stars for astrometric observations with SIM/PlanetQuest \cite{frink2001}. In \cite{hekker2006} it is shown that a large fraction of the red giants in a specific part of the absolute magnitude vs. B-V colour diagram are stable to a level of 20 m\,s$^{-1}$ and could be effectively searched for long period companions, as is required for astrometric reference stars. For other stars in the sample the radial velocity variations are larger, and for 43 stars these show significant periodicities. So far, sub-stellar companions have been announced for two stars from the present sample ($\iota$ Dra \cite{frink2002} and $\beta$ Gem \cite{reffert2006}). Here, we investigate which physical mechanism causes the observed radial velocity variations. In cases for which we do not find a significant periodicity in the observed radial velocity variations, an intrinsic mechanism such as spots or pulsations, possibly multi-periodic, seems most likely. On the other hand, the periodic radial velocity variations can be caused by sub-stellar companion, an intrinsic mechanism, or by both these mechanisms simultaneously. In Section 2 we search for a relation between the amplitude of the radial velocity variations and an intrinsic parameter, i.e.~$\log g$. In Section 3 we investigate line shape variations and compare these with theoretical predictions. Our conclusions are presented in Section 4. A more extended paper on this subject is submitted \cite{hekker2007b}. | There exists a clear correlation between the half peak-to-peak values of the radial velocity variations and the surface gravity of the red giant stars. This is a strong indication that the observed radial velocity variations are caused by a mechanism intrinsic to the star. Companions and an intrinsic mechanism might be present in stars with periodic radial velocity variations and $\log g > 1.6$ dex, while for stars with $\log g<1.6$ dex solely an intrinsic mechanism seems most likely. We have investigated whether there is a relation between the radial velocity amplitude and the amplitude of the line depth residuals. From theory we find that the rotational velocity, intrinsic and equivalent line width largely influence the line depth residual and no clear correlation could be identified. In the theoretical models temperature variations due to the pulsations is ignored, but these temperature variations influence the intrinsic and equivalent line width of the spectral line, which influences the line depth residuals. This might be an explanation why the theoretical models do not overlap with the observations. | 7 | 10 | 0710.4134 |
0710 | 0710.4072_arXiv.txt | {We investigate correlations between the optical linear polarization position angle and the orientation of the host galaxy/extended emission of Type~1 and Type~2 Radio-Loud (RL) and Radio-Quiet (RQ) quasars. We have used high resolution Hubble Space Telescope ({\it HST}) data and deconvolution process to obtain a good determination of the host galaxy orientation. With these new measurements and a compilation of data from the literature, we find a significant correlation between the polarization position angle and the position angle of the major axis of the host galaxy/extended emission. The correlation appears different for Type~1 and Type~2 objects and depends on the redshift of the source. Interpretations in the framework of the unification model are discussed.} \addkeyword{Quasars : general} \addkeyword{Polarization} \begin{document} | We can summarize our results as follow : \begin{asparaitem} \item{} While Type~2 RL and RQ quasars are known to exhibit an anti-alignment between the major axis of hteir host/extended emission and their optical polarization, we find that an alignment is mostly observed for Type~1 quasars. \item{} The redshift dependence of the alignment effect, and lack of correlation with the near-IR $PA_{host}$, suggest that it might be related to the rest-frame extended UV/blue emission of quasars. \item{}We show that these observations can be interpreted in the framework of the unification model + a two component scattering model. \end{asparaitem} | 7 | 10 | 0710.4072 |
|
0710 | 0710.3378_arXiv.txt | In order to prepare the analysis of COROT data, it has been decided to build a simple tool to simulate the expected light curves. This simulation tools takes into account both instrumental constraints and astrophysical inputs for the COROT targets. For example, granulation and magnetic activity signatures are simulated, as well as p modes, with the expected photon noise. However, the simulations rely sometimes on simple approach of these phenomenons, as the main goal of this tool is to prepare the analysis in the case of COROT data and not to perform the most realistic simulations of the different phenomenons. | Simulating the data that a space instrument like COROT will provide might look presomptuous. Indeed, it is certainly, when comparing to previous comparable instruments like IPHIR or GOLF. These two examples show that the nominal behaviour of the instrument is not always reached, but this does not prevent this instrument to provide very interesting data. However, despite some technical problems, IPHIR and GOLF yielded a wealth of scientific results. Thus, what is the interest of simulating COROT data? How close to reality these simualtions will get? This might not be the most important fact as the preparation of these simulations will help us to prepare the analysis of real data and to be ready in case of unexpected technical behaviour of the instrument perturbating the data, or unexpected physical behaviour of the targets of the instrument. A consequence of that is that the simulation tool must include technical and physical aspects, making the task even more difficult. These aspects cover: photon noise, p modes excitation, granulation signal, stellar activity signal, orbital perturbations, stellar rotation... The software presented here is freely available at:\\ {\tt www.lesia.obspm.fr/$\sim$corotswg/simulightcurve.html} | This simulator software will continue to evolve with time. As indicated above, intensity modulation due to starspots will be included, as well as other stellar or instrumental signals, as for example instrumental perturbations due to orbital vraiations. Moreover, this effort of simulation will not end with the delivery of first data but will be continued after that. The comparison with real data will allow to check for the validity of physical hypothesis used to simulate the different signals of astrophysics origin in the data. This shoud bring a great amount of information on our knowmedge of these often not well known phenomena, which stellar simulation is often derived from the solar case. In parallel, the simulation of instrumental components of the signal will be improved to help the interpretation of real data. All these reasons justify in our opinion the need for the simulation tool presented here. | 7 | 10 | 0710.3378 |
0710 | 0710.4244_arXiv.txt | We summarise the mathematical foundation of the holographic method of measuring the reflector profile of an antenna or radio telescope. In particular, we treat the case, where the signal source is located at a finite distance from the antenna under test, necessitating the inclusion of the so-called Fresnel field terms in the radiation integrals. We assume a ``full phase'' system with reference receiver to provide the reference phase. We describe in some detail the hardware and software implementation of the system used for the holographic measurement of the 12m ALMA prototype submillimeter antennas. We include a description of the practicalities of a measurement and surface setting. The results for both the VertexRSI and AEC (Alcatel-EIE-Consortium) prototype ALMA antennas are presented. | \label{intro} \PARstart{L}{arge} reflector antennas, as those used in radio astronomy and deep-space communication, generally are composed of a set of surface panels, supported on three or more points by a support structure, often called the backup structure. After assembly of the reflector it is necessary to accurately locate the panels onto the prescribed paraboloidal surface in order to obtain the maximum antenna gain. The fact that some antennas have a "shaped"contour is irrelevant for the purpose of our discussion. We are concerned with describing a method which allows us to derive the position of the individual panels in space and compute the necessary adjustments of their support points to obtain a continuous surface of a certain prescribed shape. The analysis by Ruze \cite{Ruze1966} of the influence of random errors in the reflector contour on the antenna gain indicates that the RMS error should be less than about one-sixteenth of the wavelength for acceptable performance. Under the assumption that the errors are small compared to a wavelength, randomly distributed with RMS value \(\epsilon \), have a correlation length {\bfseries c} which is much larger than the wavelength \(\lambda \), and much smaller than the reflector diameter D, the relative decrease in aperture efficiency (or gain) can be expressed by the simple formula \begin{equation} \frac{\eta_A}{\eta_{A0}} = \exp\left\{-{\left(\frac{4\pi\epsilon}{\lambda}\right)}^2\right\}, \label{eq:etaa} \end{equation} \noindent{where} \({{\eta }_{A0}}\) is the aperture efficiency of the perfect reflector. An error $\epsilon$ of \(\lambda \)/40 is required to limit the gain loss to 10 percent; with an error of \(\lambda \)/16 the gain is decreased to about half of the maximum achievable. The setting of the reflector panels at accuracies better than 100 \(\mu \)m has required the development of measuring methods of hitherto unsurpassed accuracy. It should be noted that these measurements need to be done ``in the field'', which in the case of millimeter radio telescopes generally means the hostile environment of a high mountain site. One versatile, and by now widely used method is normally called ``radio holography''. The method makes use of a well-known relationship in antenna theory: the far-field radiation pattern of a reflector antenna is the Fourier Transformation of the field distribution in the aperture plane of the antenna. Note that this relationship applies to the amplitude/phase distributions, not to the power pattern. Thus, if we can measure the radiation pattern, in amplitude and phase, we can derive by Fourier Transformation the amplitude and phase distribution in the antenna aperture plane with an acceptable spatial resolution. Bennett \etal \cite{Bennett1976} presented a sufficiently detailed analysis of this method to draw the attention of radio astronomers. Scott \& Ryle \cite{Scott1977} used the new Cambridge 5 km array to measure the shape of four of the eight antennas, using a celestial radio point source and the remaining antennas to provide the reference signal. Simulation algorithms were developed by Rahmat-Samii \cite{RahmatSamii1985} and others, adding to the practicability of the method. Using the giant water vapour maser at 22 GHz in Orion as a source Morris \etal \cite{Morris1988} achieved a measurement accuracy of 30 $\mu$m and were able to set the surface of the IRAM 30-m millimeter telescope to an accuracy of better than 100 $\mu$m. \pubidadjcol Artificial satellites, radiating a beacon signal at a fixed frequency have also been used as farfield (${R_f} = \frac{2 {D^2}}{\lambda}$) signal sources. Extensive use has been made of synchronous communication satellites in the 11 GHz band \cite{Godwin1986}, \cite{Rochblatt1992}. These transmitters of course do not provide the range of elevation angles accessible with cosmic sources. Some satellites, notably the LES (Lincoln Experimental Satellite) 8 and 9, have been used for radio holography of millimeter telescopes \cite{Baars1999}. They provided a signal at the high frequency of 37 GHz and with their geo-synchronous orbit moved over some 60 degrees in elevation angle. Unfortunately, both satellites are no longer available. Lacking a sufficiently strong source in the farfield, we have to take recourse to using an earth-bound transmitter. In practice these will be located at a distance of several hundreds of meters to a few kilometers and be at an elevation angle of less than 10 degrees. Clearly, these are in the nearfield of the antenna, requiring significant corrections to the received signals. In particular, the phase front of the incoming waves will not be plane and it contains higher order terms in the radial coordinate of the antenna aperture. These must be corrected before the Fourier transformation can be applied. We treat these corrections in detail in this paper. Successful measurements on short ranges have been reported for the University of Texas millimeter telescope \cite{Mayer1983}, the IRAM 30-m telecope \cite{Morris1988b}, the JCMT \cite{Hills2002} and the ASTE antenna of NAOJ \cite{Ezawa2000}. ALMA (Atacama Large Millimeter Array) is a new large aperture synthesis array for submillimeter astronomy consisting of 50 high accuracy antennas of 12 m diameter. The instrument is under construction at 5000 m altitude in the Atacama desert of northern Chile. ALMA is a collaboration of North America and Europe with participation of Japan. Two prototype antennas were procured and erected at the site of the Very Large Array of NRAO in New Mexico. The results of an extensive evaluation program of these antennas has been presented by Mangum \etal \cite{Mangum2006}. The reflector surface accuracy was specified at 20-25 $\mu$m, requiring a measurement method with an accuracy of 10 $\mu$m or better. This was achieved with a near-field holographic system using a transmitter at a wavelength of 3 mm and at a distance of only 315 m from the antennas at an elevation angle of 9 degrees. Here we describe these measurements in some detail. | \label{conclusions} We have successfully performed a holographic measurement and consecutive panel setting of the reflectors of the two ALMA prototype antennas to an accuracy of better than 20 $\mu$m. Our estimated measurement accuracy is approximately 5 $\mu $m. The data collection and analysis software packages are easy to use and provide quick results of the measurements, directly usable for a panel adjustment setting. We consider this system suitable for the routine setting of the ALMA production antennas to the goal of 20 $\mu$m accuracy in an acceptable time span. Modern survey equipment enables contractors to deliver reflectors with an accuracy of 50-60 $\mu$m without undue cost. Although the holography system can easily start with a much larger error, in the former case it is feasible to reach the specification with only one panel setting based on holography. We note that these measurements, being performed at one elevation angle only, do not provide information on the gravitationally induced deformation as function of elevation angle. In summary: \begin{enumerate} \item The holography system has functioned according to specification and has enabled us to measure the surface of the antenna reflector with a repeatability of better than $10\mu$m. \item As shown in Figs.~\ref{fig:holosVertex} and \ref{fig:holosAEC}, we have set both antenna surfaces to and accuracy of 16-17 $\mu$m RMS. This will provide an aperture efficiency of about 65 percent of that of a perfect reflector at the highest observing frequency of 950 GHz. \item The small differences in the surface maps obtained over several days of measurement are consistent with the measurement repeatability and at best marginally significant. If taken at face value, they indicate that the deformations of the reflector under varying wind and temperature influence are fully consistent with, and probably well within, the specification. This excellent behaviour over time is more important than the actual achieved surface setting. We stopped iteration of the settings after having achieved the goal of less than 20 $\mu$m. \item Further information on the performance of the ALMA Prototype Antennas can be found in \cite{Mangum2006}. \end{enumerate} \appendices | 7 | 10 | 0710.4244 |
0710 | 0710.2654_arXiv.txt | In this letter we briefly describe the first results of our numerical study on the possibility of magnetic origin of relativistic jets of long duration gamma ray bursters within the collapsar scenario. We track the collapse of massive rotating stars onto a rotating central black hole using axisymmetric general relativistic magnetohydrodynamic code that utilizes a realistic equation of state of stellar matter, takes into account the cooling associated with emission of neutrinos, and the energy losses due to dissociation of nuclei. The neutrino heating is not included. We describe the solution for one particular model where the progenitor star has magnetic field $B=3\times10^{10}$G. The solution exhibits strong explosion driven by the Poynting-dominated jets whose power exceeds $2\times10^{51}\,\mbox{erg/s}$. The jets originate mainly from the black hole and they are powered via the Blandford-Znajek mechanism. | \label{introduction} The phenomenon of Gamma Ray Burst (GRB) has been puzzling astrophysicists for many years since its discovery in 1970s~\cite{KSO73,MGI74}. The recent identification of long duration GRBs with supernovae (see Della Valle 2006, and Woosley \& Bloom 2006 for full review) means that we are dealing with enormous amount of energy, $10^{51}-10^{52}\mbox{erg}$, released within a very short time, 2-100 seconds, in the form of highly relativistic collimated outflow \cite{P05}. Most of the current GRB studies are focused on the physics associated with production of gamma rays in such flows and their interaction with the interstellar medium or the stellar wind of the supernova progenitor. However, the central question in the problem of GRBs is undoubtedly the nature of their central engines. These powerful jets have to be produced as a result of stellar collapse, most likely by the relativistic object, neutron star or black hole (BH), formed in the center, and make their way through the massive star unscathed, remaining well collimated and highly relativistic. The most popular model of central engine is based on the ``failed supernova'' scenario of stellar collapse, or ``collapsar'', where the iron core of progenitor star forms a BH \cite{W93}. If the progenitor is non-rotating then its collapse is likely to continue in a ``silent'' manner until the whole star is swallowed by the BH. If, however, the specific angular momentum in the equatorial part of stellar envelope exceeds that of the last stable orbit of the BH then the collapse becomes highly anisotropic. While in the polar region it may proceed more or less uninhibited, for a while, the equatorial layers form dense and massive accretion disk. The gravitational energy released in the disk can be very large, more then sufficient to stop the collapse of outer layers and drive GRB outflows, presumably in the polar direction where density is much lower \cite{MW99}. In addition, there is plenty of rotational energy in the BH itself \begin{equation} E\sub{rot} = \frac{M\sub{bh}c^2}{2} \left\{ 2-\left[ \left(1+\sqrt{1-a^2}\right)^2+a^2 \right]^{1/2} \right\}, \end{equation} where $M\sub{bh}$ is the BH mass and $a\in(-1,1)$ is its dimensionless rotation parameter. For $M\sub{bh}=3M_{\sun}$ and $a=0.9$ this gives the enormous value of $E\sub{rot} \simeq8\times10^{53}$erg. The three currently actively discussed mechanisms of powering GRB jets in the collapsar scenario are the heating via annihilation of neutrinos produced in the disk \cite{MW99}, the magnetic braking of the disk \cite{BP82,UM06}, and the magnetic braking of the BH \cite{BZ77}. The potential role of neutrino mechanism is rather difficult to assess as this requires accurate treatment of neutrino transport in a complex dynamic environment of collapsar. The long and complicated history of numerical studies of neutrino-driven supernova explosions teaches us to be cautious. Numerical simulations by MacFadyen \& Woosley\shortcite{MW99} and Aloy et al.\shortcite{AIMGM00} have demonstrated that sufficiently large energy deposition in the polar region above the disk may indeed result in fast collimated jets. However, the neutrino transport has not been implemented in these simulations and the energy deposition was based simply on expectations. When Nagataki et al.\shortcite{NTMT07} utilized a simple prescription for neutrino transport in their code they found that neutrino heating was insufficient to drive polar jets. A number of groups have studied the collapsar scenario using Newtonian MHD codes and implementing the Paczynski-Witta potential in order to approximate the gravitational field of central BH \cite{PMAB03,FKYHS06,NTMT07}. In this approach it is impossible to capture the Blandford-Znajek effect and only the magnetic braking of the accretion disk can be investigated. The general conclusion of these studies is that the accretion disk can launch magnetically-driven jets provided the magnetic field in the progenitor core is sufficiently strong. Unfortunately, the jet power has not been given in most of these papers and is difficult to evaluate from the published numbers. In the simulations of Proga et al.\shortcite{PMAB03} the jet power at $t\simeq 0.25$s is $\simeq 10^{50}\mbox{erg}/$s. The initial magnetic field in these simulations is monopole with $B\simeq 2\times 10^{14}$G at $r=3r_g$, where $r_g=GM\sub{bh}/c^2$ (private communication). \begin{figure*} \includegraphics[width=57mm]{figures/exp2.png} \includegraphics[width=57mm]{figures/exp4.png} \includegraphics[width=57mm]{figures/exp3.png} \caption{Solution immediately before the explosion (t=0.24s). Left panel: the baryonic rest mass density, $log_{10}\rho$, in $g/cm^3$ and the magnetic field lines; Middle panel: the ratio of gas and magnetic pressures, $log_{10}P/P_m$, and velocity direction vectors; Right panel: the ratio of azimuthal and poloidal magnetic field strengths, $log_{10}B^\phi/B_p$, and the magnetic field lines. } \label{f0} \end{figure*} The study of collapsars in full GRMHD is still in its infancy. Sekiguchi \& Shibata\shortcite{SS07} studied the collapse of rotating stellar cores and formation of BH in the collapsar scenario. Their results show powerful explosions soon after the accretion disk is formed around the BH and the free falling plasma of polar regions collides with this disk. These explosions are driven by the heat generated as a result of such collision. However, the authors have not accounted for the neutrino cooling and the energy losses due to photo-dissociation of atomic nuclei. and the explosions could be similar in nature to the ``successful'' prompt explosions of early supernova simulations \cite{bethe}. Mizuno et al.\shortcite{MYKS04a,MYKS04b} carried out GRMHD simulations in the time-independent space-time of a central BH. The computational domain did not include the BH ergosphere and thus they could not study the role of the Blandford-Znajek effect~\cite{K04a}. The energy losses have not been included and the equation of state (EOS) was a simple polytrope. These simulations were run for a rather short time, $\simeq 280 r_g/c$ where $r_g=GM/c^2$, and jets were formed almost immediately due to unrealistically strong initial magnetic field. In this letter we describe the first results of axisymmetric GRMHD simulations of collapsars where we use realistic EOS~\cite{TS00}, include the energy losses due to neutrino emission (assuming optically thin regime) and photo-dissociation of nuclei (see the details of micro-physics in Komissarov \& Barkov 2007 ), use the computational domain that includes the BH horizon and its ergosphere, and run simulations for a relatively long physical time, up to 0.5s. The neutrino heating is not included. \begin{figure*} \includegraphics[width=57mm]{figures/rho1.png} \includegraphics[width=57mm]{figures/rho2.png} \includegraphics[width=57mm]{figures/rho3.png} \caption{Solution on different scales at $t=0.45$s. The colour images show the baryonic rest mass density, $log_{10}\rho$ in g/cm$^3$, the contours show the magnetic field lines, and the arrows show the velocity field.} \label{f1} \end{figure*} \begin{figure*} \includegraphics[width=57mm]{figures/b1.png} \includegraphics[width=57mm]{figures/b2.png} \includegraphics[width=57mm]{figures/b3.png} \caption{The inner region at $t=0.45$s. Left panel: the magnetization parameter, $log_{10}P/P_m$, and the magnetic field lines; Middle panel: the ratio of azimuthal and poloidal magnetic field strengths, $log_{10}B^\phi/B_p$, and the magnetic field lines; Right panel: the magnetic field strength, $log_{10}(B)$, and the magnetic field lines. } \label{f2} \end{figure*} | Our results provide strong support to the idea that magnetic fields can play a crucial role in driving powerful GRB jets and associated stellar explosions not only in the magnetar model but also in the collapsar model. The main energy source for the jets and explosions in our simulations is the rotational energy of black hole and it is released via the Blandford-Znajek mechanism. The measured rate of energy release, $\dot{E} \ge 2\times10^{51}\mbox{erg}\,\mbox{s}^{-1}$, can explain the energetics of even the shortest of long duration GRBs. The fact that the rotational energy of black hole, $E\sub{bh}\simeq\mbox{few}\times10^{53}\mbox{erg}$, exceeds the typical explosion values derived from observations, $E\simeq 10^{52}\mbox{erg}$, suggests a self-regulating process in which the black hole activity ceases when the blast wave terminates further mass supply to the accretion disk. The full details of the simulations together with the results of parameter study will be presented elsewhere. | 7 | 10 | 0710.2654 |
0710 | 0710.5418_arXiv.txt | By applying recent results for the slab correlation time scale onto cosmic ray scattering theory, we compute cosmic ray parallel mean free paths within the quasilinear limit. By employing these results onto charged particle transport in the solar system, we demonstrate that much larger parallel mean free paths can be obtained in comparison to previous results. A comparison with solar wind observations is also presented to show that the new theoretical results are much closer to the observations than the previous results. | Cosmic rays (CRs) interacting with turbulent magnetic fields get scattered and accelerated (see Melrose 1968, Schlickeiser 2002). The theoretical description of these scattering and acceleration processes are essential for understanding the penetration and modulation of low-energy cosmic rays in the heliosphere, the confinement and escape of galactic cosmic rays from the Galaxy, and the efficiency of diffusive shock acceleration mechanisms. A key factor in CR scattering are the properties of the magnetic fields. A standard approach is the assumption of a superposition of a mean magnetic field $\vec{B}_0 = B_0 \vec{e}_z$ and a turbulent component $\delta \vec{B} (\vec{x})$. Whereas the mean field can easily be meassured in the solar system (here we find approximatelly $B_0 \approx 4-5 nT$), the turbulent component has to be emulated by turbulence models. In the literature there is no consensus available about the true turbulence properties (see Cho \& Lazarian 2005 for a review). In the solar system, however, some turbulence properties such as the wave spectrum can be obtained from meassurements (see e.g. Denskat \& Neubauer 1983, Bruno \& Carbone 2005). More unclear are the orientation of the turbulence wave vectors (also refered to as turbulence geometry) and the dynamical decorrelation of the magnetic fields. In a recent CR diffusion study (Shalchi et al. 2006) a slab/2D composite model was combined with a nonlinear anisotropic dynamical turbulence (NADT) model. This model can be used to reproduce meassured CR mean free paths parallel and perpendicular to the mean field $\vec{B}_0$. The authors of this article assumed that the slab correlation time scale is independent of the wave vector $\vec{k}$. In a recent study (Lazarian \& Beresnyak 2006), however, it was shown that the slab time scale is indeed $\vec{k}-$dependent.\footnote{That study put to the test the idea of the damping of slab perturbations by the ambient turbulence in Yan \& Lazarian (2002), Farmer \& Goldreich (2004).} More precisely it was found that $t_c^{-1} = \gamma_c = v_A \sqrt{k_{\parallel} / L}$; here we used the correlation time $t_c$, the correlation rate $\gamma_c$, the Alfv\'en speed $v_A$, and the outer scale of the turbulence $L$. It is the purpose of this article to apply this new result of the slab correlation time scale onto cosmic ray parallel diffusion. A comparison with solar wind observations of the parallel mean free path is also presented. It is demonstrated that we can find a much larger parallel mean free path if we employ the correlation time scale of Lazarian \& Beresnyak (2006). In Section 2 we explain the turbulence model that is used in this article. In Section 3 a quasilinear description of cosmic ray scattering is combined with this turbulence model to derive analytic forms of the pitch-angle diffusion coefficient and the parallel mean free path. In Section 4 we evaluate these formulas numerically to compute diffusion coefficients and we also provide a comparison with previous results and solar wind observations. In the closing Section 5 our results are summerized. | The theoretical explanation of measured parallel mean free paths in the solar system is a fundamental problem of space science. In a recent article (Shalchi et al. 2006) it has been demonstrated the these observations can indeed be reproduced theoretically. By using recent results of turbulence theory (Lazarian \& Beresnyak 2006) we further improved the dynamical correlation function which is a key input in transport theory considerations. It is demonstrated in this article that the improved slab correlation time scale (see Eq. (\ref{Alexmodel})) leads to a much larger parallel mean free path (see Fig. \ref{nadtf4}). This effect is important since it was argued in several previous articles that the theoretical parallel mean free path is too small (Palmer 1982, Bieber et al. 1994) in comparison with solar wind observations. Another problem of cosmic ray scattering theory is the importance of nonlinear effects. Whereas we have applied QLT in the current article it was argued in other papers (e.g. Shalchi et al. 2004) that nonlinear effects are important for parallel diffusion. However, these nonlinear effects are directly related to the interaction between charged particles and 2D modes. These 2D modes were neglected since we assumed pure slab fluctuations. Therefore, QLT can be applied and the results presented in this article should be valid. For non-slab models, where 2D modes are present, however, the applicability of QLT is questionable. It has to be subject of future work to explore the validity of QLT for realistic turbulence models such as dynamical turbulence models in non-slab geometry. | 7 | 10 | 0710.5418 |
0710 | 0710.4522_arXiv.txt | We directly measure the evolution of the intergalactic Lyman-$\alpha$ effective optical depth, $\tau_{\rm eff}$, over the redshift range $2 \leq z \leq 4.2$ from a sample of 86 high-resolution, high-signal-to-noise quasar spectra obtained with Keck/ESI, Keck/HIRES, and Magellan/MIKE. We find that our estimates of the quasar continuum levels in the \Lya~forest obtained by spline fitting are systematically biased low, but that this bias can be accounted for using mock spectra. The mean fractional error $\langle \Delta C/C_{\rm true} \rangle$ is $<1\%$ at $z=2$, 4\% at $z=3$, and 12\% at $z=4$. We provide estimates of the level of absorption arising from metals in the \Lya~forest based on both direct and statistical metal removal results in the literature, finding that this contribution is $\approx6-9\%$ at $z=3$ and decreases monotonically with redshift. The high precision of our measurement indicates significant departures from the best-fit power-law redshift evolution, particularly near $z=3.2$. | The evolution of the intergalactic medium (IGM) as traced by the \Lya~forest provides a powerful record of the thermal and radiative history of the Universe. This power owes to our ability to measure the \Lya~opacity of the IGM as a function of redshift, as well as to the relatively simple physics of the \Lya~forest. In fact, cosmological simulations in which the forest arises from absorption by smooth density fluctuations imposed on the warm photoionized IGM as a natural consequence of hierarchical structure formation within cold dark matter models \citep[e.g.,][]{1996ApJ...457L..51H}, have been remarkably successful at reproducing the properties of the absorption observed in actual quasar spectra. This synergy between theory and observations make the \Lya~forest a particularly compelling probe of the diffuse Universe.\\ \\ In this work, we present a direct precision measurement of the effective \Lya~optical depth and its evolution over the redshift range $2\leq z \leq4.2$ from a sample of 86 high-resolution, high signal-to-noise quasar spectra obtained obtained with Keck/ESI (16), Keck/HIRES (44), and with Magellan/MIKE (26). The full details have been reported in \citep{2007arXiv0709.2382F}. We assume a cosmology with $(\Omega_{m},~\Omega_{b},~\Omega_{\Lambda},~h,~\sigma_{8})=(0.27,~0.046,~0.73,~0.7,~0.8)$ \citep[][]{2007ApJS..170..377S}. | 7 | 10 | 0710.4522 |
|
0710 | 0710.3591_arXiv.txt | The near-infrared spectrum of (50000) Quaoar obtained at the Keck Observatory shows distinct absorption features of crystalline water ice, solid methane and ethane, and possibly other higher order hydrocarbons. Quaoar is only the fifth Kuiper belt object on which volatile ices have been detected. The small amount of methane on an otherwise water ice dominated surface suggests that Quaoar is a transition object between the dominant volatile-poor small Kuiper belt objects (KBOs) and the few volatile-rich large KBOs such as Pluto and Eris. | While once Pluto and Triton were the only objects in the outer solar system known to contain volatile ices on their surfaces, the recent discoveries of frozen methane on the large Kuiper belt objects (KBOs) Eris, Sedna, and 2005 FY9 have shown that these objects are part of a new class of surface volatile rich bodies in the outer solar system \citep{2005ApJ...635L..97B, 2006A&A...445L..35L,2005A&A...439L...1B, ebspec}. In contrast to these bodies with detectable volatiles, spectral observations of small KBOs over the past decade have found that most of these objects either contain varying amounts of involatile water ice on their surfaces or have flat spectra with no identifiable features \citep{Kris}. To understand the dichotomy between volatile rich and volatile free surfaces in the outer solar system, \citet{2007ApJ...659L..61S} constructed a simple model of atmospheric escape of volatile ices over the age of the solar system. They found that while most KBOs are too small and hot to retain their initial volatile ices to the present day, a small number are large and cold enough to retain these ices on their surfaces. As anticipated, the model suggests that the largest KBOs, Eris, Pluto, and Sedna are all expected to retain surface volatiles, while the vast majority of the other known objects in the Kuiper belt are expected to have lost all surface volatiles. Two known intermediate-sized KBOs are predicted to be in the transition region where they may have differentially lost some volatile ices (N$_2$) but retained others (CH$_4$). One of these transition objects, 2005 FY9, with a diameter of $\sim$1450 km \citep{2007astro.ph..2538S} does indeed appear to contain abundant CH$_4$ but be depleted in N$_2$. The other object that appears to be in the volatile non-volatile transition region is Quaoar, with a diameter of 1260$\pm 190$ km\citep{2004AJ....127.2413B}. The infrared spectrum of Quaoar does not resemble that of 2005 FY9, however. Quaoar's spectrum is dominated by absorptions due to involatile water ice, which is not detected at all on 2005 FY9. In addition, \citet{2004Natur.432..731J} reported the detection of an absorption feature near 2.2 $\mu m$ that they attributed to ammonia hydrate. They also detected the presence of crystalline water ice which, at the $\sim$ 40 K radiative equilibrium temperature of Quaoar, is thought to be converted to amorphous water ice on a relatively short ($\sim 10$ Myr) timescale by cosmic ray bombardment. The crystallinity of the water ice and the detection of the 2.2 $\mu m$ feature that they attributed to ammonia hydrate led \citet{2004Natur.432..731J} to suggest that Quaoar may have experienced relatively recent cryovolcanic activity. In this paper, we present a new infrared spectrum of Quaoar with a signal-to-noise in the K-band six times greater than that of \citet{2004Natur.432..731J} and model the ices present on the surface. | With significantly higher signal-to-noise in the 2.0 - 2.4 $\mu m$ region, the 2.2 $\mu m$ absorption feature on Quaoar previously identified as ammonia hydrate \citep{2004Natur.432..731J} is clearly seen to be due to methane ice. No compelling evidence is seen for the presence of ammonia. The presence of crystalline water ice on the surface of Quaoar still remains unexplained because it is expected that ice should currently exist in the amorphous form on the $\sim$ 40 K surface of Quaoar. However, the presence of the 1.65 $\mu m$ absorption feature due to crystalline water ice in the spectrum of every well observed water ice rich KBO (even down to diameters of only a few hundred kilometers) \citep{Kris} suggests that exotic processes such as cryovolcanism are unlikely to be required. The presence of crystalline water ice on so many small outer solar system bodies may indicate that our current understanding of the physics of the crystalline/amorphous phase transition may not be complete. The spectrum of Quaoar is consistent with that of a cold geologically dead object slowly losing the last of its volatile ices by escape in a tenuous, perhaps patchy, atmosphere. Ethane is an expected by-product of irradiation of methane ice \citep{2003NIMPB.209..283B}. The presence of ethane on Quaoar and on 2005 FY9 supports the suggestion of \citet{ebspec} that these irradiation products are preferentially seen on bodies with large abundances of pure methane rather than on the bodies where the methane is diluted in nitrogen. Quaoar also appears to be rich in more complex irradiation products. Quaoar is the only water ice rich KBO which has a red color in the visible. Other water ice rich KBOs like Orcus, Charon, and 2003 EL61 and its family of collisional fragments are all blue in the visible \citep{Kris}. Quaoar's red surface is likely due to the continued irradiation of methane, ethane, and their products on the surface \citep{2006ApJ...644..646B}. While methane on Quaoar is sufficiently volatile that it is likely to seasonally migrate if Quaoar has a moderate obliquity, ethane and the other irradiation products are essentially involatile at Quaoar's temperature. Quaoar is therefore likely to have an irregular covering of irradiation products, perhaps leading to rotational variability in its visible color and in the abundance of ethane. Continued observations of this object will provide insight into the nature of the volatile non-volatile transition and atmospheric escape in the outer solar system. {\it Acknowledgments:} We thank an anonymous referee for a helpful review. E.L.S. is supported by a NASA Graduate Student Research Fellowship. The data presented herein were obtained at the W.M. Keck Observatory, which is operated as a scientific partnership among the California Institute of Technology, The University of California and the National Aeronautics and Space Administration. The observatory was made possible by the generous financial support of the W.M. Keck Foundation. | 7 | 10 | 0710.3591 |
0710 | 0710.1846_arXiv.txt | {The theory of stellar evolution can be more closely tested if we have the opportunity to measure new quantities. Nowadays, observations of galactic RR Lyr stars are available on a time baseline exceeding 100 years. Therefore, we can exploit the possibility of investigating period changes, continuing the pioneering work started by V.~P.~Tsesevich in 1969.} {We collected the available times of maximum brightness of the galactic RR Lyr stars in the GEOS RR Lyr database. Moreover, we also started new observational projects, including surveys with automated telescopes, to characterise the O--C diagrams better.} {The database we built has proved to be a very powerful tool for tracing the period variations through the ages. We analyzed 123 stars showing a clear O--C pattern (constant, parabolic or erratic) by means of different least--squares methods.} {Clear evidence of period increases or decreases at constant rates has been found, suggesting evolutionary effects. The median values are $\beta$=+0.14~d~Myr$^{-1}$ for the 27 stars showing a period increase and $\beta$=--0.20~d~Myr$^{-1}$ for the 21 stars showing a period decrease. The large number of RR Lyr stars showing a period decrease (i.e., blueward evolution) is a new and intriguing result. There is an excess of RR Lyr stars showing large, positive $\beta$ values. Moreover, the observed $\beta$ values are slightly larger than those predicted by theoretical models.} {} | RR Lyr variables are low--mass stars in a core helium burning phase; they fill the part of the HR diagram where the horizontal branch intersects the classical instability strip. The crossing of the instability strip can take place in both directions; as a consequence, the periods will be either increasing, if the stars evolve from blue to red, or decreasing, if they evolve from red to blue. Despite its importance as a test for the stellar evolution theory, the {\it observed} rate of the period changes is still an unknown quantity. It has been measured in globular clusters (e.g., Smith \& Sandage \cite{mfifteen}, Lee \cite{lee}, Jurcsik et al. \cite{omega}), comparing periods determined in different epochs. However, this task has not been undertaken yet on the wide population of galactic RR Lyr stars, though most of them have been studied for tens of years, several of them since the end of the XIX$^{\rm th}$ century, and therefore the available data are almost continuous. One of the critical points is to determine the importance of the negative rates, i.e., the period decreases. Such rates have been observed in some cluster stars (Jurcsik et al. \cite{omega}), but it is unclear in which evolutionary phase they occur (Sweigart \& Renzini \cite{random2}). In the present paper, we derive the period variation rate of the best observed RR Lyr stars belonging to the RRab sub-class. For this purpose we use the {\it GEOS RR Lyr database} which is described in Sect.~\ref{sect_grrdb}. The observations of the TAROT telescopes coordinated in the {\it GEOS RR Lyr Survey} give a strong impulse to this analysis, providing a large number of times of maximum brightness in the recent years. They complete the effort of the amateur observers, in particular in the European associations BAV (Bundesdeutsche Arbeitsgemeinschaft f\"ur Ver\"anderliche Sterne) and GEOS (Groupe Europ\'een d'Observations Stellaires), which have been surveying these stars for years. These new observations are described in Sect.~\ref{sect_grrs}. Next, Sect.~\ref{sect_datan} shows how the data were analysed. In Sect.~\ref{sect_cstrate}, we analyse the stars with constant period variation. Section~\ref{sect_BLZH} is devoted to the study of some particularities encountered during the present work: Blazhko and light--time effects. Finally, a discussion of the results is developed in Sect.~\ref{sect_discuss} in the light of stellar evolution. | The analysis of O--C variations over a timescale of 100 years or more has proved to be a powerful tool for providing quantitative tests of the stellar evolution of the horizontal branch stars. We can stress some well--established observational facts: \begin{enumerate} \item RRab stars experiencing blueward evolution (i.e., period decreases) are quite common, only slightly less so than RRab stars experiencing redward evolution (i.e., period increases). The period ranges covered by the two groups are very similar and the mean and median period values are nearly coincident; \item the absolute values of period changes are larger than expected. Also in the extreme case of the rapid--evolving star \object{SV Eri} the rate is much larger than expected; \item the O--C behavior can be very complicated in some cases, with abrupt or irregular period variations, rather than monotonic. The regular variations caused by the light--time effect (and hence duplicity), often invoked to explain large O--C excursions, are not convincingly observed in our extensive sample. The physical explanations should be searched in the stellar structure; \item Blazhko effect is often superimposed on secular changes, but the monotonic trend due to evolutionary variations still remains visible. \end{enumerate} As a general conclusion about the comparison between our observational results and the theoretical predictions, we claim that there is a very powerful feedback between the two approaches. In particular, theoretical investigations should take into account that we have observational evidence of many RR Lyr stars showing blueward evolution. The theoretical models should also match the observed $\beta$ values in a more satisfactory way, as these seem to be higher than expected, both for redward and blueward evolutions. On the other hand, the observational effort to monitor RR Lyr stars should be continued, possibly extended to stars at fainter magnitudes, and the accuracy of the maximum time determinations should be improved (by using automated telescopes), thus obtaining the same information on a shorter time baseline. | 7 | 10 | 0710.1846 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.