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
|
|---|---|---|---|---|---|---|---|
9803 | astro-ph9803186_arXiv.txt | Observations of elemental abundances in metal-poor halo stars provide important evidence regarding the early history, evolution and age of the Galaxy. Spectroscopic studies over a number of years have indicated the presence of neutron-capture, specifically rapid neutron-capture ({\it i.e.} {\it r}-process), elements in a number of these metal-poor halo stars (see {\it e.g.} Spite and Spite 1978, Sneden and Parthasarathy 1983, Sneden and Pilachowski 1985, Gilroy et al. 1988, Gratton and Sneden 1991, 1994, Sneden et al. 1994, McWilliam et al. 1995a, 1995b, Cowan et al. 1996, Sneden et al. 1996, Burris et al. 1998). In addition, abundance comparisons of the {\it r}-process and {\it s}-process ({\it i.e.} slow neutron-capture) elements between the oldest metal-poor halo stars and more metal-rich halo and disk stars provide direct evidence about the nature of the chemical evolution of the Galaxy. | Ground-based observations of the halo stars have indicated the presence of a number of neutron-capture elements. The availability of the HST has allowed for other spectral regions to be studied, and as a result more elements have been detected in these stars. For example, Figures 2 and 3 illustrate (see also Sneden et al. 1998) that we have now detected the element Ge in (three) halo stars. In addition, the 3$^{rd}$ {\it r}-process peak elements, Os-Pt, as well as Pb have also now been detected in two stars using the HST (Cowan et al. 1996, Sneden et al. 1998). Including the ground-based detections of Th in CS~22892--052 (see Figure 1) and in HD 115444 (see Cowan et al. 1998), {\it r}-process elements from proton numbers of Z = 32 to 90 have now been observed in metal-poor stars. This is a much wider range in proton (and mass) number than ever seen before and now includes the important 3$^{rd}$ {\it r}-process peak. These detections further demonstrate that the {\it r}-process, ranging up to the formation of the element Th, was in operation early in the history of the Galaxy. The observations also provide important information about the nature of the progenitors of the halo stars. The {\it r}-process elements cannot be internally synthesized in the halo stars. Instead they must be produced in a previous generation (or generations) in an {\it r}-process site. Since the metal-poor halo stars were formed early in the history of the Galaxy, presumably shortly after formation, the presence of {\it r}-process elements in the halo stars requires very short evolutionary timescales for their progenitors. This further implies massive stars. While there is some uncertainty about the exact nature of the astrophysical site for the {\it r}-process, it has long been suspected to be in supernovae, particularly core collapse supernovae from massive stars (see Cowan et al. 1991a). The abundance observations of the metal-poor halo stars appear to support that suspicion. It has also become clear with the accumulating data that the neutron-capture elements in the metal-poor stars have an abundance pattern that appears to be the same as the solar {\it r}-process distribution. This is illustrated vividly in Figure 1, where all of the elements from Ba to Os in CS~22892--052 have relative solar abundances. While this correlation has been noted in the past (see {\it e.g.} Gilroy et al. 1988), the high-resolution data in this star, covering a wide range of elements including Os in the 3$^{rd}$ {\it r}-process peak, makes the argument much stronger. This same solar {\it r}-process pattern also appears in other stars, as shown in Figure 2. Both ground-based and HST observations of HD~115444 ([Fe/H] = --2.7) show that the elemental abundance from Ba to Pt are consistent with a scaled solar {\it r}-process curve. Further evidence of this is seen in the metal-poor halo star HD~122563 (Sneden et al. 1998). Elements such as Ba and La, which today are made predominantly in the {\it s}-process, appear to have been made exclusively in the {\it r}-process early in the history of the Galaxy. While this has been suggested previously (see {\it e.g.} Truran 1981), the new observational data strongly support the contention that most (or all) of the elements were made in the {\it r}-process at the earliest times in the Galaxy. The observations indicate the same relative {\it r}-process abundance pattern in the oldest galactic stars and in the solar material, at least for elements with Z $\ge$ 56. Therefore, the data indicate that the {\it solar system {\it r}-process abundances are not the result of global averages over different types of stars and epochs}. Instead, the stellar data suggest that the conditions that produced the {\it r}-process elements are narrowly confined, perhaps both in terms of temperature and density of the nucleosynthesis and in terms of the mass range of the astrophysical sites (see Wheeler et al. 1998, Freiburghaus et al. 1998). The apparent lack of mixing early in the history of the Galaxy, when the relative abundance pattern is already apparent, demonstrated by Figure 6 also makes it less likely that the solar system {\it r}-process distribution is the result of averages. We note in Figure 1, however, that while the elements in CS~22892--052 from Ba and above (Z $\ge$ 56) are well-fit by the solar {\it r}-process distribution, the extrapolation to the lower mass elements does not entirely fit the abundance data. In particular, while Sr and Zr do appear to be solar, the Y abundance is far below the solar curve. It is difficult to explain why two but not three of these neighboring elements in this star are solar. We note, further, that the abundance data for HD~115444, shown in Figure 2, show a similar separation between the lighter and heavier n-capture elemental abundances. In this star, again the abundances from Ba and above appear solar, but Sr-Zr do not. There may be several possible explanations. The weak {\it s}-process, expected to occur during helium core burning in massive stars, is expected to contribute to the abundances of the elements from Sr-Zr. The data may be showing such a contribution and Cowan et al. (1995) even suggested some combination of the weak {\it s}-process and the {\it r}-process might be needed to explain the abundances of Sr-Zr in CS~22892--052. We note, however, one problem in this scenario is the apparent difficulty of producing {\it s}-process elements in stars of extremely low metallicity. An alternative explanation may be that the more massive {\it r}-process elements are synthesized in one site and the lower mass elements in another. Based upon meteorite data, Wasserburg et al. (1996) have suggested the existence of two {\it r}-process sites with the separation in production occurring near mass number 140, {\it i.e.} near Ba. Possible alternative {\it r}-process sites have been discussed by Wheeler et al. (1998) and Baron et al. (1998). Further observations and analyses will be needed to understand the formation history of the lower mass {\it r}-process elements. The most metal-rich of the halo stars studied here is HD~126238, with a metallicity of [Fe/H] = --1.7. We see the same basic trends for this star in Figure 3 that are seen in the more metal-poor stars CS~22892--052 and HD~115444. We note, however, that the abundance of Ba seems to lie above the solar {\it r}-process curve. It was suggested by Cowan et al. (1996) that this might indicate some {\it s}-process contribution to the Ba abundance. In other words, by the time that HD~126238 formed at a metallicity of --1.7, presumably more recently than the other two previously mentioned stars, some galactic {\it s}-processing had occurred. It is seen in Figure 3 that the stellar Ba abundance is still below the total solar Ba abundance leading Cowan et al. to suggest that only the most massive stellar contributors to the {\it s}-process had evolved at that point in time. Some support for their contention is given by Figure 4, which indicates the galactic chemical evolutionary trends for Ba as a function of metallicity. In this case metallicity is indicated by $\alpha$, which may be a more reliable metallicity indicator than Fe, which is formed in both Type II and Type~I supernovae. These data spanning a wide range in metallicty might be explained by an evolutionary delay in the production of {\it s}-process material. As demonstrated earlier in this paper, at early times in the Galaxy Ba apparently is produced from the {\it r}-process. We note the clear change in slope in Figure 4 at a metallicity [$\alpha$/H] $\simeq$ --2 that appears to indicate the onset of the main {\it s}-process nucleosynthesis production for Ba (and presumably other n-capture elements) in the Galaxy. Further evidence of this change in production mechanism, as a function of metallicity (and presumably time), for Ba is given in Figure 5. At very low metallicities (and early times) the {\it s}-process element Ba and the {\it r}-process element Eu appear to be synthesized solely in the {\it r}-process. While there is scatter in the available data, we see that at the lowest values of [$\alpha$/H] the [Ba/Eu] value in most of the stars is consistent with a pure {\it r}-process origin. The long-lived radioactive nuclei (known as chronometers) in the uranium-thorium region are formed exclusively by the {\it r}-process and can be used to determine the ages of stars and the Galaxy. One such chronometer, Th, has been detected in CS~22892--052 (Sneden et al. 1996, Cowan et al. 1997) (see Figure 1). Comparison between the initial abundance value produced in an {\it r}-process site (often known as the production value) and the observed abundance value leads to a direct estimate of stellar ages. The abundances of the stable elements in the 3$^{rd}$ {\it r}-process peak, a nuclear region nearby to the U-Th region, can be used to help constrain the predictions of the initial values of the long-lived radioactive chronometers, independent of knowing the site for the {\it r}-process. Comparing the solar and the observed Th/Eu ratio in CS 22892--052, Cowan et al. (1997) found an age estimate of 15 $\pm$ 4 Gyr for this star. They noted that consideration of galactic chemical evolution could lead to an older age of 17 $\pm$ 4 Gyr. Pfeiffer et al. (1997) employed newer and more accurate nuclear data in the context of a waiting point approximation {\it r}-process model. Using the stable stellar and solar data to constrain the predicted abundances of the radioactive {\it r}-process nuclei, they compared the initial (as opposed to solar) value of Th/Eu with the stellar value and found a best estimate for this star of 13.5 Gyr. Their result for CS~22892--052 was not only consistent with Cowan et al. (1997), but is also consistent with recent globular cluster age determinations based upon Hipparcos data (see Pont et al. 1998). (See also Cowan et al. 1991a,b for a discussion of galactic and cosmological age determinations.) This technique, based upon predicted and observed radioactive chronometers, has been extended to an additional star with an age result approximately the same as that for CS~22892--052 (Cowan et al. 1998). We caution, however, that there are still many uncertainties, and to improve the accuracy of the chronometric estimates will require more observational and theoretical studies. It is encouraging to note, though, that the detection of thorium in CS 22892--052, and other halo stars, offers promise as an independent technique for determining stellar ages, and thus putting limits on galactic and cosmological age estimates. | 98 | 3 | astro-ph9803186_arXiv.txt |
|
9803 | gr-qc9803087_arXiv.txt | Despite the fact that the Schwarzschild and Kerr solutions for the Einstein equations, when written in standard Schwarzschild and Boyer-Lindquist coordinates, present coordinate singularities, all numerical studies of accretion flows onto collapsed objects have been widely using them over the years. This approach introduces conceptual and practical complications in places where a smooth solution should be guaranteed, i.e., at the gravitational radius. In the present paper, we propose an alternative way of solving the general relativistic hydrodynamic equations in background (fixed) black hole spacetimes. We identify classes of coordinates in which the (possibly rotating) black hole metric is free of coordinate singularities at the horizon, independent of time, and admits a spacelike decomposition. In the spherically symmetric, non-rotating case, we re-derive exact solutions for dust and perfect fluid accretion in Eddington-Finkelstein coordinates, and compare with numerical hydrodynamic integrations. We perform representative axisymmetric computations. These demonstrations suggest that the use of those coordinate systems carries significant improvements over the standard approach, especially for higher dimensional studies. | \label{sec:introduction} It is well known that the black hole solutions of the field equations of general relativity, with the Schwarzschild and Kerr solutions being astrophysically the most relevant, exhibit coordinate singularities when written in coordinates adapted to the exterior region. The very notion of a black hole was greatly clarified with the discovery, in the early sixties, of coordinate systems that remove those singularities, and indeed cover the whole spacetime~\cite{Kruskal60} (for a general overview see~\cite{MTW73},\cite{Hawking}). The (rotating) black hole solution for the metric tensor, expressed in standard Boyer-Lindquist coordinates $(t,r,\theta,\phi)$ is singular at the roots of the equation $\Delta=r^2-2Mr+a^2=0$ (where $M$ is the mass and $a$ the angular momentum aspect of the hole~\cite{units}). In the spherically symmetric Schwarzschild case, the singularity at $r=2M$ is removable with the use of appropriate transformations of the radial and time coordinates. Slightly more complicated transformations, involving also the azimuthal coordinate, can remove the singularities at $r=r_{\pm}=M \pm (M^2-a^2)^{1/2}$ of a rotating black hole. The location of the coordinate singularity coincides with the event horizon of the black hole. Asymptotic observers lose causal contact with events near the black hole at precisely this location. Hence, despite the singular appearance of the metric, in simulations of matter flows in the background gravitational field of a black hole, one is in principle allowed to consider the open interval extending from the event horizon to some ``far zone" at a large, but finite, distance from the hole. This approach has been widely used over the years in the numerical simulations of flows around black holes~\cite{Wi72} -\cite{Font98a}. The blow-up of the metric components at the horizon has implications on the behavior of hydrodynamical quantities. The coordinate flow velocity becomes ultra-relativistic and reaches the speed of light at the horizon. As a consequence, the Lorentz factor becomes infinite causing any numerical code to crash. Placing the inner boundary close to the horizon, required for capturing the effects of the relativistic potential, introduces large gradients in all hydrodynamical variables. The steep radial behavior makes the task of numerical evolution with a reasonable degree of accuracy challenging. Besides practical considerations, the approach of evolving only the exterior domain lacks of a well defined location for the inner boundary: the computation cannot include the horizon surface, but must commence at a location ``sufficiently close'' to it. Physically, the influence of the horizon region on the solution will progressively red-shift away the closer one gets to the horizon. The validity of a certain choice though, must be continuously reasserted, as new flows or effects are being investigated, with careful tests of convergence, as the inner boundary is progressively moved inwards. Such tests are complicated by the fact that, as mentioned in the previous paragraph, the solutions appear singular at the horizon and hence demand increasingly more resolution. Ameliorating those problems has motivated the use of a logarithmic radial coordinate (the so-called {\it tortoise} coordinate~\cite{RW57}). This technique relegates the event horizon to a infinite, negative, coordinate distance. An equidistant grid in the tortoise $r_{*}$ coordinate maps into an increasingly dense grid in the Schwarzschild $r$ coordinate, with infinite density at the horizon. This approach has proven successful in the extensive semi-analytic studies of black hole perturbation theory, and recently also for the axisymmetric integration of curvature perturbations as an initial value problem~\cite{Krivanetal}. In wave systems, the ambiguity of the location of the inner boundary is addressed by the simple limiting form of the governing equations near the horizon. This is the well known fact that black holes act as finite potential barriers to electromagnetic and gravitational perturbations~\cite{chandra}. The use of a tortoise coordinate does not bring similarly extensive benefits to the study of the hydrodynamical equations. The issue of the inner boundary location is less transparent for those equations. The steepness of the solution and the ``artificially'' high coordinate velocities persist, although they are now more treatable due to the substantial increase in resolution. In three dimensional simulations using Cartesian coordinates, the tortoise technique is not possible at all. It has been shown recently~\cite{Papadopoulos98a} that {\em adaptive mesh refinement} can indeed provide, even for 3D systems, the required resolution close to the event horizon. However, as alluded to above, these undesired pathologies can be eliminated in the case of fixed given black hole backgrounds with rather simple coordinate transformations. This reserves the power of adaptive mesh refinement for the more physically interesting features of the solution. There is considerable freedom in choosing coordinates regular at the horizon, which can be productively reduced by imposing criteria that can enhance their suitability for numerical applications. The obvious first criterion is of course the {\em regularity of the metric}, in particular at the horizon. A second condition is that the constant time surfaces are {\em everywhere spacelike}, as this is, currently, a pre-requisite for the implementation of modern numerical methods for relativistic hydrodynamic flows. An important third criterion is the {\em time-independence} of the metric components. This leads to constant in time coefficients in the equations and simplifies disentangling the true hydrodynamical evolution from coordinate effects in the black hole background. Interestingly, we will see that those conditions still do not fix the coordinate system uniquely. We show that the number of available choices greatly reduces, but is still infinite, in the rotating black hole case. We give several examples of such systems, which we collectively call {\em horizon adapted coordinate systems}. Such coordinate systems address the issues raised above in a straight-forward way: any radius {\em inside} the horizon is equally appropriate (in the idealized continuum limit) for the imposition of a boundary condition, as the domain is causally disconnected from the exterior. Importantly, the irrelevance of the inner boundary location will persist even after the inclusion of other possible local physical processes that may be considered in conjuction with the hydrodynamical flow, e.g., radiative processes. The coordinate velocities of the flow will be bounded at the horizon, as they represent projections of the (finite) fluid four velocity onto a regular coordinate system. Hence the hydrodynamical nature of the flow becomes considerably less demanding on the integration algorithm. Gradients in the solution for {\em scalar variables} such as pressure and enthalpy will of course persist. Those are physical and due to the curvature of the black hole, which requires a significant dynamic range for its resolution. The organization of this paper is as follows. In section II we introduce a class of horizon adapted coordinate systems for a non-rotating black hole. The rotating black hole and the more restricted class of coordinates available in that case are discussed in the Appendix. We outline the numerical hydrodynamical framework of our computations in section III. Exact solutions for spherical (Bondi) accretion are presented in section IV for both dust and perfect fluids. Section V describes the numerical results. In our numerical simulations we focus mainly on the spherically symmetric case, which captures the essential nature of the problem. Some axisymmetric computations are also briefly considered. The coordinate system on which we base our computations is the celebrated Eddington-Finkelstein coordinate system~\cite{Eddington24},\cite{Finkel58}. Our main aim in this report is to show the functionality of this class of coordinate systems as frameworks for the integration of the equations of relativistic hydrodynamics in black hole spacetimes. In three-dimensions (and rotating holes) computations of accretion onto black holes are likely to benefit significantly by the adoption of a horizon adapted coordinate system. | We have presented a family of {\it horizon adapted coordinate systems} for the numerical study of accretion flows around black holes. In the rotating case we identify a discrete but infinite family, of which the first simple members are well known stationary coordinate systems. In the non-rotating case the freedom in building the regular stationary foliation is quantified by (at least) the space of bounded positive functions of one variable. We have shown how these systems allow for a better numerical treatment of accretion scenarios. Existing numerical studies of accretion flows onto black holes have been performed in the original, singular systems, i.e., Schwarzschild coordinates for the Schwarzschild solution and Boyer-Lindquist coordinates for the rotating (Kerr) solution. Although it is possible to solve the problem in these pathological coordinates, one is introducing artificial complexity, being forced to use very high resolution to deal with the unphysically large gradients that develop in the vicinity of the horizon. This may prevent the accurate solution of three dimensional problems. At the same time, the ambiguity regarding the position of the inner boundary of the domain (which should be the horizon) introduces a convergence criterion that must be enforced at all times if the solutions are to be trusted. We have focussed on the particular case of the Eddington-Finkelstein form of the Schwarzschild metric. The general relativistic hydrodynamic equations are now regular at the horizon, which permits an accurate description of accretion flows. We have demonstrated the feasibility of this approach with the numerical study of the spherical accretion (Bondi accretion) of dust and perfect fluid, and the comparison with the exact solutions which we re-derived in this coordinate system. We have also shown the functionality of the new coordinates in axisymmetric computations of relativistic Bondi-Hoyle accretion flows. In a forthcoming paper we plan to extend this approach to the rotating case, considering horizon adapted coordinate systems to study accretion flows in stationary Kerr spacetimes. Three dimensional accretion flows onto black holes are interesting both from an astrophysical and geometrical point of view, as they are thought to correspond to observable electromagnetic emission, and hence may help map the relativistic rotating black hole potential. The framework proposed here will help detailed numerical studies of such systems in the near future. | 98 | 3 | gr-qc9803087_arXiv.txt |
9803 | astro-ph9803073_arXiv.txt | The depletion of lithium during the pre-main sequence and main sequence phases of stellar evolution plays a crucial role in the comparison of the predictions of big bang nucleosynthesis with the abundances observed in halo stars. Previous work has indicated a wide range of possible depletion factors, ranging from minimal in standard (non-rotating) stellar models to as much as an order of magnitude in models which include rotational mixing. Recent progress in the study of the angular momentum evolution of low mass stars (Krishnamurthi \etal 1997a) permits the construction of theoretical models which reproduce the angular momentum evolution of low mass open cluster stars. The distribution of initial angular momenta can be inferred from stellar rotation data in young open clusters. In this paper we report on the application of these models to the study of lithium depletion in main sequence halo stars. A range of initial angular momenta produces a range of lithium depletion factors on the main sequence. Using the distribution of initial conditions inferred from young open clusters leads to a well-defined halo lithium plateau with modest scatter and a small population of outliers. The mass dependent angular momentum loss law inferred from open cluster studies produces a nearly flat plateau, unlike previous models which exhibited a downwards curvature for hotter temperatures in the \7li - T$_{\rm eff}$ plane. The overall depletion factor for the plateau stars is sensitive primarily to the solar initial angular momentum used in the calibration for the mixing diffusion coefficients. The \6li/\7li depletion ratio is also examined. We find that the dispersion in the plateau and the \6li/\7li depletion ratio scale with the absolute \7li depletion in the plateau and we use observational data to set bounds on the \7li depletion in main sequence halo stars. A maximum of 0.4 dex depletion is set by the observed dispersion and \6li/\7li depletion ratio and a minimum of 0.2 dex depletion is required by both the presence of highly overdepleted halo stars and consistency with the solar and open cluster \7li data. The cosmological implications of these bounds on the primordial abundance of \7li are discussed. | The status of big bang nucleosynthesis (BBN) as a cornerstone of the hot big bang cosmology rests on the agreement between the theoretical predictions and the primordial abundances of the light elements deuterium (D), helium-3 (\3he), helium-4 (\4he), and lithium-7 (\7li) inferred from observational data (\cite{YTSSO}; \cite{WSSOK}; \cite{CSTI}). Comparisons of this type often rely on models of galactic chemical and/or stellar evolution in order to associate the observed abundances of these light elements at the present epoch with their predicted primordial values. Recently the confrontation between prediction and observation has come under stricter scrutiny as it appeared that the primordial abundance of deuterium inferred from ISM/Solar data was smaller than its predicted abundance - the value of which follows from requiring that the standard big bang model predictions for \4he are in good agreement with the abundance inferred from observations of metal-poor extragalactic \hii regions (\cite{Hata95}; \cite{CSTII}; \cite{Hata97}). Very recent measurements of the deuterium abundance along the lines-of-sight to high-redshift QSOs(\cite{BT97}), along with a reanalysis of the \4he data in light of new observations(\cite{OSS96}), increase the tension between prediction and observation. The role of \7li in these comparisons has been minor due to the large uncertainty in the estimate of the amount of lithium destruction during the lifetimes of \popii (\ie, metal poor) halo stars which could accomodate primordial lithium abundances ranging from the observed plateau value\footnote{The lithium abundances of \popii stars with $T_{\rm eff} > 5800$K and [Fe/H] $< -1.3$ is nearly independent of metallicity ($[Li] = 2.25\pm 0.1$, where $[X] = 12 + \log y_X$ with $y_X$ the number ratio of X to hydrogen) and, hence, is referred to as a ``plateau".} (no depletion) up to a factor of ten larger. This range in possible depletion factors is equally compatible with primordial lithium abundances which correspond to either ``low deuterium'' (which favors a primordial lithium abundance a factor of 3 higher than the plateau value), or the observed \4he (which favors the plateau value). The arguments in favor of minimal lithium depletion are the flatness of the \popii lithium abundance plateau (at low metallicity and high temperature) with respect to both metallicity and temperature and the low dispersion in the lithium abundance at fixed T$_{\rm eff}$. This was generally consistent with ``standard'' stellar models (\ie, models without rotation which burn lithium via convection during pre-main sequence evolution) which predicted little depletion (\cite{DDKKR}). There are however some specific areas of disagreement in the comparison of the halo star data with standard models. The observed dispersion may be greater than that predicted (\cite{DPD93}; \cite{Thorburn94}; \cite{Ryan96}) or not (\cite{MPB95}; \cite{Spite96}). There is a small population of highly overdepleted stars which appear normal except for \7li (\cite{Thorburn94}; \cite{NRBD}), and the trends with metal abundance appear to conflict with expectations from the models (Thorburn 1994, \cite{CD94}). However, the overall agreement between standard stellar evolution theory and observations in halo stars is good enough that nonstandard models would probably not be invoked to explain this data set. Therefore many investigators have adopted the reasonable assumption that the observed \popii lithium abundances are close to the primordial value. The overall properties of \7li depletion in standard models have been extensively studied (for a review see \cite{MP97}) and there are some model independent predictions: There is partial \7li depletion in the pre-main sequence (pre-MS), little or no main sequence (MS) depletion for stars hotter than about 5500 K, and there should be little or no dispersion in the \7li abundance at a given T$_{\rm eff}$ in clusters. Unfortunately, the observational data obtained from halo stars does not serve as a good diagnostic for these models since the initial abundance is not known and we have no information on the history of the \7li abundance. Instead, one can look to the Sun and open clusters, systems with a nearly uniform initial lithium abundance ([Li] = 3.2 to 3.4) where detailed abundance data is available as a function of mass, age, and composition. These \popi data provide stringent tests of theoretical models and it has been known for quite some time (\eg, see \cite{WS65} and \cite{Z72}) that standard models fail these tests. The disagreement between the data and standard models has increased as both the observational data and the standard theoretical models have improved. The open cluster data provide clear evidence that the \7li abundance decreases with increasing age on the MS, contrary to the standard model prediction that the convective zones of these hot ($\geq$ 5500K) stars are not deep enough to destroy \7li on the MS. In addition, there is strong evidence for a dispersion in abundance at fixed T$_{\rm eff}$ and unexpected mass-dependent effects, such as the strong depletion seen in mid-F stars relative to both hotter and cooler stars (the ``Li dip'' of \cite{BT86} - for extensive reviews of this issue, see \cite{PH88}, \cite{MC91}, \cite{S95}, \cite{Bal95}, and \cite{MP97}). For standard models, these features cannot be explained by variations in the input physics (\eg, opacities); rather, they indicate the operation of physical processes not usually included in standard models. By extension, these processes could also be operating in the \popii stars. Another potentially useful diagnostic of lithium depletion can be found in globular clusters. Clusters provide samples which are homogeneous in age and composition, so one would expect a smaller dispersion in a cluster sample than in a field star sample. This is certainly the case for \popi stars: \popi field stars do show a larger dispersion than open cluster stars (\cite{LHE91}; \cite{FMS96}). However, recent observations of Li in globular cluster subgiant and turnoff stars show the opposite trend and create some serious difficulties for the minimal \7li depletion scenario for metal poor stars (\cite{DBK95}, \cite{BDSK97}; \cite{TDRRO97}). There are clear star-to-star differences in M92; out of 3 stars observed with very similar color, 2 have lithium abundances well below the plateau value. In NGC 6397, 20 stars near the turnoff were observed. For 7 with identical B-V color, there is a scatter of a factor of 2-3 in lithium abundance. Therefore a standard model treatment of the halo field stars must explain why the lithium abundances in globular clusters, but not in field stars, are anomalous. The disagreement between the \popi lithium abundances and the predictions of standard stellar models has stimulated investigation of nonstandard stellar models. The most prominent explanations are mixing (either from rotation or waves), microscopic diffusion, and mass loss (\cite{MP97}; \cite{MC91}). For \popi G and K stars, microscopic diffusion and mass loss are not likely explanations (\cite{MC91}, \cite{SF92}), which leaves mixing as a logical candidate. Mixing induced by rotation can explain many of the overall properties of the \popi data. Rotational mixing can, on relatively long timescales, mix material from the base of the convective zone to interior regions of the star where lithium can be burned; both the rotation rate and the \7li depletion rate decrease with increased age; and a distribution of initial rotation rates can produce a distribution of \7li depletions and thus a dispersion in abundance at fixed T$_{\rm eff}$. Although models with rotational mixing have the qualitative properties needed to explain the Pop I and Pop II data, two classes of objections have been raised: 1) the uniqueness of the solutions and adequacy of the physical model; and 2) discrepancies in the quantitative comparison of observation and theory. Rotational models require an understanding of the angular momentum as a function of radius and as a function of time. There have been persistent difficulties in reproducing the surface rotation rates as a function of mass and time in open clusters (\cite{CDP95a},b; \cite{KMC95}; \cite{Bou95}). In addition, models with internal angular momentum transport from hydrodynamic mechanisms, such as the ones used in this paper, predict more rapid core rotation in the Sun than is compatible with helioseismology data (\cite{CDP95a}, Krishnamurthi et al. 1997a (KPBS), \cite{TST95}). Either of these difficulties could potentially affect the degree of mixing in the models. Models with rotational mixing predict that a range of initial angular momenta will generate a range of lithium depletion rates and therefore a dispersion in abundance among stars of the same mass, composition, and age. However, the difficulty in reproducing the angular momentum evolution of low mass stars has made quantifying the predicted dispersion difficult, and therefore only qualitative estimates have been made (\eg, see \cite{PKD}(PKD), \cite{CD94}, \cite{CDP95b}, and \cite{CVZ92}). There have also been mismatches between the mean trend inferred from the data and from all classes of theoretical models. Chaboyer \& Demarque (1994), for example, concluded that the observed mean trend of lithium abundance with T$_{\rm eff}$ in metal-poor halo stars was not reproduced in {\it any} class of models, be they standard, rotational, or with diffusion. There are also cases, such as the cool stars in young open clusters, where the observed dispersion is large but the rotational models predict little, if any, dispersion. The case for or against standard model lithium depletion is not as clear cut when we consider halo stars. It is argued that models with significant rotationally induced depletion could not produce a flat plateau with limited dispersion about the mean plateau abundance (\eg, \cite{BM97}). Furthermore, it was expected that such models would destroy far too much \6li (\cite{SFOSW}) in conflict with the observation of \6li in HD 84937 (see section 4.2). On the other hand, standard (convective burning) models show lithium depletion trends contrary to the \popi data. At minimum the mechanisms responsible for the \popi \7li pattern need to be identified and shown to not affect \popii stars. It is difficult to construct a model where the nonstandard effects completely cancel for \popii stars while still retaining consistency with the \popi data. Models with microscopic diffusion predict modest \7li depletion in plateau halo stars. In general these same models predict that the timescale for changes in the surface \7li abundance decreases with increased mass; this produces a downward curvature in the \7li-T$_{\rm eff}$ relationship at high temperatures which is not observed. Mass loss can counteract this trend (\cite{Sw95}; \cite{VC95}). In either case, the net effect is modest depletion at the 0.2 dex level. Rotational mixing can also suppress diffusion (Chaboyer \& Demarque 1994, \cite{VC95}). The cancellation of different effects still requires at least some mean \7li depletion in the halo stars. In a recent preprint \cite{VC98} note that although the surface \7li abundance varies strongly with mass in diffusive models, the peak subsurface abundance does not (note that the height of the peak is not preserved in models which include mixing). They then use the height of that peak to constrain the absolute \7li abundance, arguing that the appropriate mass loss strips each star down to the region that contains this peak \7li abundance. The survival of \6li is also problematic in unmixed models with sufficient mass loss to expose the peak abundance; in a model of ours which can be compared to Figure 1 in \cite{VC98}, \6li drops to half its surface value in the outer 0.01 solar masses, well below the comparable mass content in the \7li preserving region of 0.02 solar masses and also the peak in the \cite{VC98} model. Further, it is disturbing to use a class of models which do not include a fundamental characteristic of stars, namely, rotation, when that property has been shown to be capable of affecting the issue being studied. In our view the most serious source of uncertainty in models with rotational mixing has been the understanding of the angular momentum evolution of low mass stars. We will show that many of the difficulties in reconciling observation and theory are resolved when models that are consistent with the rotation data in open clusters are used. In order to make definitive predictions of the amount of rotational mixing, it is necessary to follow the stellar angular momentum histories which requires knowledge of the initial distribution of rotation velocities along with an angular momentum loss law. There has been significant recent progress in constructing theoretical models consistent with the rotation data in young open clusters (KPBS; \cite{CL94}; \cite{KMC95}; \cite{BFA97}; \cite{A97}). With the latest generation of models, we can both infer the distribution of initial conditions and place strong constraints on the surface rotation as a function of time. This enables us to quantify the expected dispersion in lithium abundance and greatly reduces the sensitivity of the model predictions to uncertainties in the input physics. Those models which accurately reproduce the lithium abundances observed in open clusters can then be used to predict lithium depletion in halo stars. The general trend is that these ``open cluster normalized'' rotation models for halo stars predict more lithium depletion than do the standard models but less depletion than that predicted by earlier studies of rotational mixing with a less sophisticated treatment of angular momentum evolution. Our goal in this paper is to constrain the amount of lithium depletion in \popii stars using several observables: (1) an estimate of the halo star initial rotation rate distribution as derived from open clusters, (2) the absence of large dispersion in the observed lithium abundances of the \popii ``plateau'' stars, (3) the \6li abundance and/or the \6li/\7li abundance ratio in HD 84937. We argue that, individually and in combination, these observables point to \popii halo star lithium depletion of at least 0.2 dex (following from the open cluster data) but no more than 0.4 dex (a consequence of the narrowness of the plateau and of the \6li considerations). Based on the observational data we can then infer the primordial lithium abundance and compare and contrast it with that predicted by standard BBN for consistency with the inferred primordial abundances of D and/or \4he. \section {Method and Comparison with Previous Models} \subsection { Angular Momentum Evolution and Rotational Mixing} In standard stellar models, lithium depletion is a strong function of mass and composition; it also depends on the input physics, particularly the opacities and model atmospheres used to specify the surface boundary condition. In our models the standard model physics is the same as described in KPBS (\cite{KPBS}): namely, interior opacities from OPAL (\cite{RI92}), low-T opacities and model atmospheres at solar abundance from \cite{Ku91a},\cite{Ku91b}); nuclear reaction rates from \cite{BPW95}, including the \cite{CF88} \7li(p, $\alpha$) cross-section; Yale EOS; a solar calibrated mixing length, and $Y=0.235$. Chaboyer \& Demarque (1994) noted that unusual Li-T$_{\rm eff}$ trends for cool metal-poor stars occurred in models using the Kurucz atmospheres and could be lessened by using a grey atmosphere. We therefore ran halo star models with grey atmospheres and a mixing length of 1.25 as obtained for solar models constructed under the same assumptions. Stellar models which include rotational mixing require additional input physics beyond standard stellar models. The important new ingredients include: \begin{enumerate} \item a distribution of initial angular momenta \item a prescription for angular momentum loss \item a prescription for the internal transport of angular momentum and the associated mixing in radiative regions \item the impact of rotation on the structure of the model. \end{enumerate} There have been important changes in the treatment of the first three ingredients since the study of rotational mixing in halo dwarfs by \cite{PDD}; in this section we discuss the ingredients of the current set of models and compare them with previous work. The treatment of angular momentum evolution in this set of models is the same as KPBS. The structural effects of rotation have been computed using the method of \cite{KT70}, and are small for low mass stars which experience angular momentum loss. \subsubsection {Initial Angular Momentum} The studies of lithium depletion in the presence of rotationally induced mixing by \cite{PKSD} (the Sun), PKD (open cluster stars), and PDD (halo stars) all used a range of initial rotation rates in the pre-MS from the measured rotation velocities of young T Tauri stars (10-60 km/s) at a typical reference age of 1 Myr. No interaction between accretion disks and the protostar was included; the entire range of initial conditions was generated from the range in the initial rotation velocity. This caused some difficulty in reproducing the observed range in rotation (a factor of 20 in young open clusters), especially since the higher angular momentum loss rate in rapid rotators acted to reduce the range in initial rotation rates. This approach was also used by Chaboyer \etal (1995a,b) for the Sun and open cluster stars respectively and by Chaboyer \& Demarque (1994) for halo stars, although these papers used a different angular momentum loss law than the earlier work. As protostars evolve their moment of inertia decreases; this would produce an increase in rotation rate with decreased luminosity. In contrast, the rotation periods of classical T Tauri stars appear to be nearly uniform - they do not scale with luminosity as they would if the stars were conserving angular momentum as they contracted towards the main sequence (\cite{B93}, \cite{E93}). Weak-lined T Tauri stars, which are thought to be protostars without significant accretion disks, have much shorter rotation periods. This behavior would be expected if there were an accretion disk regulating the rotation of the central object and spinup only occurred when the disk is no longer magnetically coupled to the protostar (\cite{K91}; \cite{CCQ95}; \cite{KMC95}). In this revised picture there is a constant surface rotation rate while there is a sufficiently massive disk around the central object; stars which detach from their disks earlier experience a larger change in angular velocity as they contract to the main sequence than those which detach from their disk later (and are only free to begin to spin up when their moment of inertia is smaller). This disk-locking hypothesis leads to a larger range in rotation rates for models of young open cluster stars, in better agreement with the data, and explains the very slow rotation of the majority of young stars (\cite{BFA97}; \cite{CCQ95}; \cite{KMC95}; KPBS). These differences in the initial conditions lead to interesting consequences for the lithium depletion pattern. Although the initial rotation periods are comparable to those used in the earlier studies, the inclusion of star-disk coupling implies that the MS angular momentum is much lower for models with long disk lifetimes than it would be if the disks were absent. Disk lifetimes of 3 Myr are needed to match the rotation rates of the slow rotators in young clusters. The mean depletion for the majority of stars is therefore significantly lower than in previous studies because most young stars are slow rotators and the degree of mixing decreases for models with lower MS angular momentum. \subsubsection {Angular momentum loss.} PKSD, PKD, and PDD all used a loss law of the form ${dJ}/{dt} \propto \omega^3$, where $\omega$ is the angular velocity (\cite{K88}). This implies greater angular momentum loss for rapid rotators than for slow rotators. However, models of this type do not reproduce the rapid rotators seen in young open clusters; the spin down of high angular momentum objects is too severe (PKD). On both theoretical and observational grounds, a more realistic functional form for the loss law is \begin {equation} \frac{dJ}{dt} \propto \omega^3\ \ \ \ \ \ (\omega < \omega_{\rm crit}), \end {equation} and \begin {equation} \frac{dJ}{dt} \propto \omega \omega_{\rm crit}^2 \ \ \ \ \ \ (\omega > \omega_{\rm crit}) \end {equation} (\cite{MB91}; \cite{CCQ95}; \cite{M84}; \cite{BS96}). Here $\omega_{\rm crit}$ is the angular velocity at which the angular momentum loss rate saturates; $\omega_{\rm crit}$ can be estimated from several observable quantities, thought to be correlated with surface magnetic field strength, which saturate at 5-20 times the solar rotation period (see \cite{PS96}, \cite{Kr97b} for reviews and discussions of recent work). A loss law which saturates at high rotation rates allows rapid rotation to survive into the early main sequence phase; a loss law of this form was adopted by Chaboyer \etal (1995a,b) and Chaboyer \& Demarque (1994). The observational data indicate that the duration of the rapid rotator phase is a function of mass in the sense that rapid rotation survives longer in lower mass stars. There is also indirect observational evidence for a mass-dependent saturation threshold (\cite{PS96}; \cite{Kr97b}). Models where the saturation threshold scales with the convective overturn time scale can reproduce the observed mass dependent spin down pattern. We adopt the mass-dependent saturation threshold of KPBS, \begin {equation} \omega_{\rm crit} = \omega_{\rm crit}(\odot) \frac {\tau_{\rm conv}(\odot)} {\tau_{\rm conv} (*)}, \end {equation} where the convective overturn time scale $\tau_{\rm conv}$ as a function of ZAMS T$_{\rm eff}$ was taken from the 200 Myr isochrones of Kim \& Demarque (1996; KD). \cite{KD96} only considered models at solar abundance, and we would need to recalibrate the KPBS models if we used a time and abundance dependent overturn timescale for the models. A comparison of the convection zone depth (as a function of T$_{\rm eff}$, Z, and age) in the tables of PKD and PDD indicates that at young ages the convection zone depth at a given T$_{\rm eff}$, and therefore the overturn time scale, depends only weakly on metal abundance. The value of $\omega_{\rm crit}$ is important primarily in the early main sequence. We therefore evolved standard halo star models to an age of 200 Myr and used the same $\tau_{\rm conv}$ as a function of T$_{\rm eff}$ for the metal-poor models as reported by KD for the solar abundance models. We will consider models with a mass, composition, and time dependent overturn time scale in a future paper in preparation (\cite{N98}). A loss law with a mass-dependent saturation threshold acts preferentially to suppress rapid rotation in the more massive stars, unlike the constant saturation threshold adopted by Chaboyer \& Demarque (1994). This reduces both the absolute depletion and the dispersion of lithium abundances in hotter halo plateau stars relative to cooler plateau stars. \subsubsection {Angular momentum transport.} The treatment of angular momentum transport is the same as in KPBS; a detailed review of the time scale estimates can be found in Pinsonneault (1997). We consider internal angular momentum transport by hydrodynamic mechanisms alone, and do not include potentially important mechanisms such as gravity waves (\cite{KQ97}; \cite{ZTM97}) and magnetic fields (\cite{CM92}, \cite{CM93}; \cite{BCM98}). In previous work the degree of mixing was found to be insensitive to the assumptions about internal angular momentum transport (PKD; PDD; \cite{CDP95a},b). \cite{CVZ92} and \cite{R96} obtained similar depletion patterns in models with solid body rotation, and \cite{BCM98} reported similar results in models with magnetic fields and hydrodynamic mechanisms. This can be understood because the diffusion coefficients for angular momentum transport are actually largely determined by the balance between the surface boundary condition (the applied torque) and the flux of angular momentum from below a given shell. Theory gives an estimate of the ratio of the diffusion coefficients for mixing to those for angular momentum, which is related to the existence of anisotropic turbulence in stellar interiors (\cite{CZ92}). The diffusion coefficients for mixing are calibrated on the Sun, but our lack of information on the rotational history of the Sun requires an exploration of models with varying solar initial conditions. | The lithium abundances of stars have interesting implications for cosmology and for the theory of stellar structure and evolution. The conclusions of this paper impact both of these areas. Standard stellar models implicitly assume that a variety of physical processes known to occur in real stars can be neglected; for many purposes this is a good approximation. There is now extensive evidence in \popi stars that the predictions of standard models are not in agreement with all the data. Furthermore, work from a variety of investigators on physical effects not included in the standard models has concluded that there are physically well-motivated processes which can affect the surface lithium abundances of stars. Standard models and the previous generation of nonstandard models predicted initial \7li abundances that differ by up to a factor of ten from the same set of current \popii data, with very different implications for cosmology. We believe that mild envelope mixing driven by rotation is the most promising candidate for explaining the {\it complete} observational picture. An improved treatment of angular momentum evolution in low mass stars now allows us to generate stellar models with rotation which are consistent with the rotation rates observed as a function of mass and age. We have inferred the distribution of initial angular momenta in young clusters and computed the distribution of lithium depletion factors as a function of mass, composition, and age. The resulting lithium depletion pattern is in good agreement with both \popi and \popii data, and the distribution of theoretical depletion factors is consistent with the distribution of abundances. For \popii stars in particular, the observed slope of the [Li]-T$_{\rm eff}$ relationship, the observed dispersion in abundance at fixed T$_{\rm eff}$, the existence of a small population of overdepleted stars, and the simultaneous detection of \6li and \7li in one halo star are all consistent with the predictions of the theoretical models. The primary uncertainty in the absolute depletion of \7li is the initial angular momentum of the Sun, which is used to calibrate the mixing diffusion coefficients. We have used the observed properties mentioned above to set bounds on \7li depletion for \popii stars. Those models which fit the \popi data best predict a small but real depletion of lithium in the warm (T$_{\rm eff} \geq 5800$K), metal-poor ([Fe/H] $\leq -1.3$) \popii stars: 0.2 $\leq$ logD$_{7} \leq$ 0.4. These same models are consistent with the very small observed dispersion in abundances about the plateau value and with the survival of some \6li. These depletion factors are significantly less than in previous models with rotational mixing (PDD, Chaboyer \& Demarque 1994), and this difference can be directly attributed to the adoption of an improved treatment of angular momentum evolution. The uncertainties in the \7li depletion factor can be reduced by a combination of new data and improved stellar evolution models. A large set of accurate lithium abundances in old open clusters would enable us to calibrate the diffusion coefficients for mixing without relying on the solar calibration. Microscopic diffusion and rotational mixing should be considered together, although the work of Chaboyer \& Demarque (1994) indicated that the predictions of such models are similar to those which include rotation alone. The observed \7li depletion pattern is relatively insensitive to the treatment of internal angular momentum transport, with a factor of ten in the time scale corresponding to less than a 10\% change in the logarithmic lithium depletion factor in the work of PDD. However, the solar rotation curve does provide evidence for angular momentum transport from mechanisms not considered in the current paper, such as magnetic fields and/or internal gravity waves. Lithium depletion in models which include these effects and hydrodynamic mechanisms should be explored. Intermediate metallicity stars could also provide a valuable test of the dependence of lithium depletion on metal abundance. When our constraints on lithium-depletion are combined with the abundances inferred from observations of the Spite plateau halo stars ([Li]$_{\rm OBS} = 2.25 \pm$ 0.10) we find for the range in the abundance of primordial lithium: 2.35 $\leq$ [Li]$_{\rm P} \leq$ 2.75 (2.2 $\leq 10^{10}$(Li/H)$_{\rm P} \leq$ 5.6). Although lithium is not the ideal baryometer, a comparison with the predictions of standard BBN identifies two options: low-$\eta$ and high-$\eta$. The low-$\eta$ branch suggests a very low baryon density, open Universe which may be in conflict with the baryon density inferred from observations of the Lyman-alpha forest. Although the helium abundance predicted for this low-$\eta$ option agrees with that inferred from \hii region data, the very high predicted deuterium abundance may not be consistent with observations. In contrast, the high-$\eta$ branch corresponds to a baryon density consistent with the Ly-$\alpha$ data and permits a higher density Universe. In this case the predicted primordial deuterium abundance is in excellent agreement with the low deuterium QSO data but a considerably higher primordial helium abundance is predicted than is inferred from some of the observational data. | 98 | 3 | astro-ph9803073_arXiv.txt |
9803 | astro-ph9803245_arXiv.txt | The low-redshift evolution of the intergalactic medium is investigated using hydrodynamic cosmological simulations. The assumed cosmological model is a critical density cold dark matter universe. The imposed uniform background of ionizing radiation has the amplitude, shape and redshift evolution as computed from the observed quasar luminosity function by Haardt \& Madau. We have analysed simulated \lya spectra using Voigt-profile fitting, mimicking the procedure with which quasar spectra are analysed. Our simulations reproduce the observed evolution of the number of \lya absorption lines over the whole observed interval of $z=0.5$ to $z=4$. In particular, our simulations show that the decrease in the rate of evolution of \lya absorption lines at $z\le 2$, as observed by {\em Hubble Space Telescope}, can be explained by the steep decline in the photo-ionizing background resulting from the rapid decline in quasar numbers at low redshift. | Neutral hydrogen in the intergalactic medium produces a forest of \lya absorption lines blueward of the \lya emission line in quasar spectra (Lynds 1971). Observations of quasars at redshifts $z>2$ show that there is strong cosmological evolution in the number of \lya lines, which can be characterised by a power law $dN/dz \propto (1+z)^\gamma$ (Sargent \etal 1980), where $N$ is the number of lines above a threshold rest-frame equivalent width $W$ (typically $W>0.32\AA$). Studies at high resolution using the Keck telescope find $\gamma=2.78\pm 0.71$ for $2<z<3.5$ (\eg Kim \etal 1997). At even higher $z$ the evolution is still stronger: Williger \etal (1994) find $\gamma>4$ for $z>4$ using the CTIO 4m telescope. In contrast at low redshifts, observations using the {\em Hubble Space Telescope} find much less evolution, $\gamma=0.48\pm0.62$ for $z<1$ (Morris \etal 1991, Bahcall \etal 1991, 1993, Impey \etal 1996). Recently, hydrodynamic simulations of hierarchical structure formation in a cold dark matter (CDM) dominated universe have been shown to be remarkably successful in reproducing this \lya forest in the redshift range $z=4\rightarrow 2$ (Cen \etal 1994, Zhang, Anninos \& Norman 1995, Miralda-Escud\'e \etal 1996, Hernquist \etal 1996, Wadsley \& Bond 1996, Zhang \etal 1997, Theuns \etal 1998). These simulations show that the weaker \lya lines (neutral hydrogen column density $N_\H\le 10^{14}$ cm$^{-2}$) are predominantly produced in the filamentary and sheet-like structures that form naturally in this structure formation scenario. Velocity structure in these lines is often due to residual Hubble flow since many of the absorbing structures are expanding. In contrast, the stronger lines ($N_\H\ge 10^{16}$ cm$^{-2}$) tend to occur when the line of sight passes near a dense virialised halo. Over the redshift range investigated in these simulations a photo-ionizing background close to that inferred from quasars (Haardt \& Madau 1996) is required to explain the properties of the \lya forest. In fact, although most simulations have assumed a critical density, scale-invariant CDM universe, other variants of the CDM model provide acceptable fits with relatively small changes to the ionizing background (Cen \etal 1994, Miralda-Escud\'e \etal 1996). The general success of CDM-like models in explaining the high redshift ($z\ge 2$) properties of the \lya forest is impressive. In this {\em Letter} we investigate using hydrodynamic simulations whether a CDM universe with a photo-ionizing background dominated by quasars can explain the observed transition in the cosmological evolution of the number of \lya lines at $z\le 2$. | \begin{figure*} \setlength{\unitlength}{1cm} \centering \begin{picture}(17,12) \put(1, -4){\special{psfile=fig2.ps hscale=65 vscale=65}} \end{picture} \caption{Evolution of the number of lines with given range in column density with redshift from simulations compared against observed evolution. Column density cut $10^{13.1}$cm$^{-2}\le N_\H\le 10^{14}$ cm$^{-2}$-- simulations: circles connected with dotted line; data: open triangles (Kim \etal 1997). Column density cut $10^{13.77}$ cm$^{-2}\le N_\H\le 10^{16}$ cm$^{-2}$-- simulations: circles connected with solid line; data: filled squares (Bahcall \etal 1993), open squares (Impey \etal 1996), filled triangles (Kim \etal 1997), filled pentagon (Lu \etal 1996). Column density cut $10^{14.5}$ cm$^{-2}\le N_\H\le 10^{16}$ cm$^{-2}$ -- simulation: circles connected with long dashed line; data: long dashed line shows evolution from Williger \etal (1994). Large filled circles are simulation results for the low resolution, large box size, run, open circles are for a higher resolution, smaller box size, run. Large open pentagon: re-analysis of simulation at $z=0.5$, but imposing the ionizing background appropriate to $z=2$. Data were taken from figure~2 in Kim \etal (1997), except for the Wiliger \etal (1994) data. } \label{fig:fig2} \end{figure*} We show in figure~\ref{fig:spectra} examples of simulated spectra at $z=3$, 2, 1 \& 0.5 for the $L=22.22$ Mpc lower resolution simulation. Fluctuations in neutral hydrogen density, caused by gas tracing dark matter potential wells, produce absorption features similar to those in observed spectra. At low redshifts, there are large regions of the spectrum with very low absorption. These regions are separated by prominent absorption features, most of which are just a single strong line. At higher redshifts, there is considerable absorption over most of the spectrum and many lines are blended. Many of the strong lines at $z=0.5$ can be traced back to higher redshifts. There is a clear decrease in the comoving number of lines with decreasing redshift. The evolution of the column density distribution with redshift is illustrated in figure~\ref{fig:fig1}. There is clear evolution in the simulated column density distribution with redshift. The rate of change depends on column density, with higher column density lines undergoing stronger evolution leading to steepening of the distribution. At $z=2$, the column density distribution is $\propto N_\H^{-1.6}$ whereas this has steepened to $\propto N_\H^{-2.1}$ at $z=0.5$ (see figure~\ref{fig:fig1}). The rate of evolution also depends on redshift, with considerably stronger evolution at higher redshifts. The number of weak \lya lines ($\le 10^{13.1}$ cm$^{-2}$) remains approximately constant. At redshifts 3 and 2, there is good agreement between the simulated column density distribution for our higher resolution simulation ($L=5.5$~Mpc) and the observed one. The drop in the number of lines with redshift can be quantified by counting the number of lines within a given column density range. The simulation results are compared to observations in figure~\ref{fig:fig2}. The simulations reproduce well the number of lines at a given redshift for all three column density cuts. Crucially, they also match very well the observed number of lines at low redshift. Consequently, the hierarchical picture of galaxy formation in a critical density universe, coupled with the observed evolution in the quasar luminosity function, can explain the observed evolution of the number of \lya lines over the entire observed redshift range. We have re-analysed the $z=0.5$ output time after increasing the imposed ionization flux from its $z=0.5$ value to the value appropriate to $z=2$. The number of lines with $10^{13.77}$ cm$^{-2}\le N_\H\le 10^{16}$ cm$^{-2}$ is shown by the open pentagon in figure~\ref{fig:fig2}. This point falls onto the extrapolation for the number density evolution for $z\ge 2$. Consequently, the dominant reason for the higher number of lines at low $z$ compared to what would be expected by extrapolation from high $z$, is the decrease in ionizing flux from $z=2$ to $z=0$, itself a consequence of the evolution of the quasar luminosity function. An estimate of the reliability of these simulations can be obtained by comparing the two simulations run at different resolutions. The higher resolution simulation produces more lines at all column densities, but the difference between the two simulations is well within the error bars of the observational results. This gives us confidence that we can reliable predict the number of lines from these simulations. In summary: our numerical simulations show that the properties of the \lya forest are in excellent agreement with what is expected in a cold dark matter universe with a photo-ionizing background dominated by quasar light. In particular, our simulations show that the observed decrease in the rate of evolution of \lya absorption lines at $z\le 2$ can be explained by the steep decline in the photo-ionizing background resulting from the rapid decline in quasar numbers at low redshift. | 98 | 3 | astro-ph9803245_arXiv.txt |
9803 | astro-ph9803303_arXiv.txt | We report on the March-April 1997 BeppoSAX observations of Aql X-1, the first to monitor the evolution of the spectral and time variability properties of a neutron star soft X--ray transient from the outburst decay to quiescence. We observed a fast X--ray flux decay, which brought the source luminosity from $\sim 10^{36}$ to $\sim 10^{33}\ergs$ in less than 10 days. The X--ray spectrum showed a power law high energy tail with photon index $\Gamma\sim 2$ which hardened to $\Gamma \sim 1-1.5$ as the source reached quiescence. These observations, together with the detection by RossiXTE of a periodicity of a few milliseconds during an X--ray burst, likely indicate that the rapid flux decay is caused by the onset of the propeller effect arising from the very fast rotation of the neutron star magnetosphere. The X--ray luminosity and hard spectrum that characterise the quiescent emission can be consistently interpreted as shock emission by a turned-on rotation-powered pulsar. | Soft X--Ray Transients (SXRTs), when in outburst, show properties similar to those of persistent Low Mass X--Ray Binaries containing a neutron star (LMXRBs; White et al. 1984; Tanaka \& Shibaza\-ki 1996; Campana et al. 1998). The large variations in the accretion rate that are characteristic of SXRTs allow the investigation of a variety of regimes for the neutron stars in these systems which are inaccessible to persistent LMXRBs. While it is clear that, when in outbursts, SXRTs are powered by accretion, the origin of the low luminosity X--ray emission that has been detected in the quiescent state of several SXRTs is still unclear. An interesting possibility is that a millisecond radio pulsar (MSP) turns on in the quiescent state of SXRTs (Stella et al. 1994). This would provide a ``missing link" between persistent LMXRBs and recycled MSPs. \begin{figure*}[!htb] \psfig{figure=aql_fig.ps,width=4.3cm} \caption{ Light curve of the Feb.-Mar. 1997 outburst of Aql X-1 (panel {\it a}. Data before and after MJD 50514 were collected with the RossiXTE ASM (2--10 keV) and the BeppoSAX MECS (1.5--10 keV), respectively. RossiXTE ASM count rates are converted to (unabsorbed) luminosities using a conversion factor of $4\times 10^{35}\ergs$ (before MJD 50512) and $2\times10^{35}\ergs$ (after MJD 50512) as derived from RossiXTE spectral fits (Zhang, Yu \& Zhang 1998). BeppoSAX luminosities are derived directly from the spectral data (see text). The evolution of the flux from MJD 50480 to MJD 50512 is well fit by a Gaussian centered on MJD 50483.2. This fit however does not provide an acceptable description for later times (see the dot-dashed line), not even if the accretion luminosity is calculated in the propeller regime (dashed line). The straight solid line represents the X--ray luminosity corresponding to the closure of the centrifugal barrier $L_{\rm min}$ (for a magnetic field of $10^8$ G and a spin period of 1.8 ms) and the straight dashed line the luminosity gap due to the action of the centrifugal barrier, $L_{\rm cor}$. The dotted line marks the minimum luminosity in the propeller regime ($L_{\rm lc}$). Panel {\it b} shows the BeppoSAX unfolded spectra of Aql X-1 during the early stages of the fast decline (1) and during the quiescent phase (3--6, summed). The best fit spectral model (black body plus power law) is superposed to the data.} \label{tot} \end{figure*} Aql X-1 is the most active SXRT known: more than 30 X--ray and/or optical outbursts have been detected so far. The companion star has been identified with the K1V variable star V1333 Aql and an orbital period of 19 hr has been measured (Chevalier and Ilovaisky 1991). The outbursts of Aql X-1 are generally characterised by a fast rise (5--10 d) followed by a slow decay, with an $e-$folding time of 30--70~d (see Tanaka \& Shibazaki 1996 and Campana et al. 1998 and references therein). Type I X--ray bursts were discovered during the declining phase of an outburst (Koyama et al. 1981), testifying to the presence of a neutron star. Peak X--ray luminosities are in the $\sim (1-4)\times 10^{37}\ergs$ range (for the $\sim 2.5$ kpc distance inferred from its optical counterpart; Thorstensen et al. 1978). Close to the outburst maximum the X--ray spectrum is soft with an equivalent bremsstrahlung temperature of $k\,T_{\rm br} \sim 4-5$~keV. Sporadic observations of Aql X-1 during the early part of the outburst decay (Czerny et al. 1987; Tanaka \& Shibazaki 1996; Verbunt et al. 1994) showed that when the source luminosity drops below $\sim 10^{36}\ergs$ the spectrum changes to a power law with a photon index of $\Gamma\sim 2$, extending up to energies of $\sim 100$ keV (Harmon et al. 1996). ROSAT PSPC observations revealed Aql X-1 in quiescence on three occasions at a level of $\sim 10^{33}\ergs$ (0.4--2.4 keV; Verbunt et al. 1994). In this lower energy band the spectrum is considerably softer and consistent with a black body temperature of $k\,T_{\rm bb} \sim 0.3$~keV. | \begin{table*} \begin{center} \caption{Summary of SAX NFIs observations.} \label{tab1} \begin{tabular}{cccccc} Obs./Date & LECS/MECS-PDS & LECS Count Rate & MECS Count Rate & PDS Count Rate\\ & Expos. (s) & (c s$^{-1}$) & (c s$^{-1}$) & (c s$^{-1}$)\\ \hline 1/March 8$^{th}$, 1997 &\, 5240/21342 & $0.84 \pm 0.02$ & $2.2\pm 0.01$ & $0.87\pm0.06$ \\ 2/March 12$^{th}$, 1997 &\, 3247/21225 & $(2.5\pm0.4) \times 10^{-2}$ & $(7.4\pm 0.2) \times 10^{-2}$ & $\lsim 0.19$ \\ 3/March 17$^{th}$, 1997 &\, 5755/17258 & $(5.6\pm1.7) \times 10^{-3}$ & $(1.3\pm 0.1) \times 10^{-2}$ & $\lsim 0.24$ \\ 4/March 20$^{th}$, 1997 &\, 4287/22589 & $(5.8\pm1.9) \times 10^{-3}$ & $(1.6\pm 0.1) \times 10^{-2}$ & $\lsim 0.17$\\ 5/April 2$^{nd}$, 1997 &\, 8440/23576 & $(6.2\pm1.3) \times 10^{-3}$ & $(1.2\pm 0.1) \times 10^{-2}$& $\lsim 0.17$ \\ 6$^{\rm o}$/May 6$^{th}$, 1997 & 11789/21703 & $(6.7\pm1.1) \times 10^{-3}$ & --- & $\lsim 0.19$ \\ \hline \end{tabular} \end{center} {\noindent $^{\rm o}$ No MECS data were obtained.} \end{table*} \begin{table*} \begin{center} \caption{Summary of spectral fits.} \label{tab2} \begin{tabular}{ccccccc} Obs.$^*$ & Black body & Black body & Power law & PL/BB & Mean Luminosity$^\dag$ & Reduced\\ & $k\,T_{\rm bb}$ (keV) & $R_{\rm bb}$ (km)& $\Gamma$ & flux ratio$^{\ddag}$&(erg s$^{-1}$)& $\chi^2$\\ \hline 1 & $0.42\pm0.02$&$2.6\pm0.3$&$1.9\pm0.1$& 3.7 & $9\times 10^{34}$ &1.0 \\ 2 & $0.3_{-0.2}^{+0.1}$&$0.7_{-0.3}^{+6.6}$&$1.8\pm0.7$& 1.6 & $2\times 10^{33}$ &0.9 \\ \ 3--6 & $0.3\pm0.1$&$0.8_{-0.1}^{+0.4}$&$1.0\pm0.3$& 0.7 & $6\times 10^{32}$ &1.3 \\ \hline \end{tabular} \end{center} { \noindent Errors are $1\,\sigma$. \noindent $^*$ Spectra from the LECS and MECS (and PDS for the first observation) detectors have been considered. The spectra corresponding to the quiescent state have been summed up, in order to increase the statistics and an upper limit from the summed PDS data was also used to better constrain the power law slope. \noindent $\dag$ Unabsorbed X--ray luminosity in the energy range 0.5--10 keV. In the case of the first observation the PDS data were included in the fit (the unabsorbed 0.5--100 keV luminosity amounts to $2\times 10^{35}\ergs$). \noindent $\ddag$ Power law to black body flux ratio in the 0.5--10 keV energy range.} \end{table*} The BeppoSAX observations enabled us to follow for the first time the evolution of a SXRT outburst down to quiescence. The sharp flux decay leading to the quiescent state of Aql X-1 is reminiscent of the final evolution of dwarf novae outbursts (e.g. Ponman et al. 1995; Osaki 1996), although there are obvious differences with respect to the X--ray luminosities and spectra involved in the two cases, likely resulting from the different efficiencies in the gravitational energy release between white dwarfs and neutron stars. Models of low mass X--ray transient outbursts hosting an old neutron star or a black hole are largely built in analogy with dwarf novae outbursts. In particular, van Paradijs (1996) showed that the different range of time-averaged mass accretion rates over which the dwarf nova and low mass X--ray transient outbursts were observed to take place is well explained by the higher level of disk irradiation caused by the higher accretion efficiency of neutron stars and black holes. However, the outburst evolution of low mass X--ray transients presents important differences. In particular, the steepening in the X--ray flux decrease of Aql X-1 has no clear parallel in low mass X--ray transients containing Black Hole Candidates (BHCs). The best sampled light curves of these sources show an exponential-like decay (sometimes with a superposed secondary outburst) with an $e-$folding time of $\sim 30$ d and extending up to four decades in flux, with no indication of a sudden steepening (Chen et al. 1997). In addition, BHC transients display a larger luminosity range between outburst peak and quiescence than neutron star SXRTs (Garcia et al. 1998 and references therein). Being the mass donor stars and the binary parameters quite similar in the two cases, it appears natural to attribute these differences to the different nature of the underlying object: neutron stars possess a surface and, likely, a magnetosphere, while BHCs do not. When in outburst accretion down to the neutron star surface takes place in SXRTs, as testified by the similarity of their properties with those of persistent LMXRBs, especially the occurrence of type I bursts and the X--ray spectra and luminosities. The mass inflow rate during the outburst decay decreases, causing the expansion of the magnetospheric radius, $r_{\rm m}$. Thus, accretion onto the neutron star surface can continue as long as the centrifugal drag exerted by the corotating magnetosphere on the accreting material is weaker than gravity (Illarionov \& Sunyaev 1975; Stella et al. 1986). This occurs above a limiting luminosity $L_{\rm min}=G\,M\,\mdot_{\rm min}/R \sim 4\times10^{36} \,B_8^2\,P_{-3}^{-7/3} ~{\rm erg\,s^{-1}}$, where $G$ is the gravitational constant; $M$, $R$, $B=B_8\,10^8$ G and $P=P_{-3}\,10^{-3}$ ms are the neutron star mass, radius, magnetic field and spin period, respectively (here and in the following we assume $M=1.4\msole$ and $R=10^6$ cm). As $r_{\rm m}$ reaches the corotation radius, $r_{\rm cor}$, accretion onto the surface is inhibited and a lower accretion luminosity ($<L_{\rm {min}}$) of $L_{\rm cor}=G\,M\,\mdot_{\rm min}/r_{\rm cor}\sim 2\times 10^{36}\,B_8^2\,P_{-3}^{-3}\ergs$ is released. After this luminosity gap the source enters the propeller regime. If the mass inflow rate decreases further, the expansion of $r_{\rm m}$ can continue up to the light cylinder radius, $r_{\rm lc}$, providing a lower limit to the accretion luminosity that can be emitted in the propeller regime $ L_{\rm lc}=G\,M\,\mdot_{\rm lc}/r_{\rm lc}\sim 2\times 10^{34}\,B_8^2\,P_{-3}^{-9/2}\ergs$. Below $L_{\rm {lc}}$ the radio pulsar mechanism may turn on and the pulsar relativistic wind interacts with the incoming matter pushing it outwards. Matter inflowing through the Roche lobe is stopped by the radio pulsar radiation pressure, giving rise to a shock front (Illarionov \& Sunyaev 1975; Shaham \& Tavani 1991). Clearly these regimes have no equivalent in the case of black hole accretion. \subsection{The onset of the propeller} During the February-March 1997 outburst of Aql X-1, RossiXTE observations led to the discovery of a nearly coherent modulation at $\sim 550$~Hz ($\sim 1.8$ ms) during a type I X--ray burst. A single QPO peak, with a centroid frequency ranging from $\nu_{QPO}\sim 750$ to 830 Hz, was also observed at two different flux levels, when the persistent luminosity was $\sim 1.2\times 10^{36}\ergs$ and $\sim 1.7\times 10^{36}\ergs$ (Zhang et al. 1998). In the presence of a single QPO peak, the magnetospheric and sonic point beat frequency models (Alpar \& Shaham 1985; Miller et al. 1997) interpretation is ambiguous in that the QPO peak could represent either the Keplerian frequency at the inner disk boundary or the beat frequency. Moreover, the burst periodicity at $\sim 550$~Hz may represent the neutron star spin frequency, $\nu_s$, or half its value (Zhang et al. 1998). In either cases, the possibility that accretion onto the neutron star surface takes place even in the quiescent state of Aql X-1 faces serious difficulties: for a quiescent luminosity of order $10^{33}\ergs$ a magnetic field of only $\lsim 5\times 10^6$ G would be required, in order to fulfill the condition $r_{\rm m}\lsim r_{\rm cor}$. For such a low magnetic field, Aql~X-1 and, by inference, LMXRBs with kHz QPOs can hardly be the progenitors of recycled MSPs. More crucially, the marked steepening in the outburst decay that takes place below $\sim 1\times 10^{36}\ergs$, is accompanied by a marked spectral hardening, resulting from a sudden decrease of the flux in the black body spectral component. This is clearly suggestive of a transition taking place deep in the gravitational well of the neutron star, where most of the X--rays are produced. The most appealing mechanism is a transition to the propeller regime, where most of the inflowing matter is stopped at the magnetospheric boundary (Zhang, Yu \& Zhang 1998). In Fig. 1a, the luminosity at MJD 50512 is identified with $L_{\rm {min}}$ and the lower horizontal lines indicate the luminosity interval during which Aql X-1 is likely in the propeller regime. Additional information on the neutron star magnetic field (and spin) can be inferred as follows. The observed ratio of the luminosity, $L_{QPO}$, when the QPO at $\sim 800$~Hz were detected and the luminosity $L_{\rm min}$ when the centrifugal barrier closes is $L_{QPO}/L_{\rm min} \sim 1.2-1.7$. At $L_{\rm min}$ the Keplerian frequency of matter at the magnetospheric boundary is, by definition, equal to the spin frequency, i.e. $\nu_s \sim 550$ or $\sim 275$~Hz for Aql X-1. Based on beat-frequency models, at $L_{QPO}$ the Keplerian frequency at the inner disk boundary can be either $\nu_{K,QPO}\sim 800$~Hz or $\nu_{K,QPO}\sim \nu_s + 800$~Hz, depending on whether the single kHz QPOs observed corresponds to the Keplerian or the beat frequency. In the magnetospheric beat-frequency models, simple theory predicts that the Keplerian frequency at the magnetospheric boundary is $\propto L^{3/7}$; in the radiation pressure-dominated regime relevant to the case at hand, the Ghosh and Lamb (1992) model predicts instead $\propto L^{3/13}$. Therefore we expect $\nu_{K,QPO}/\nu_s \sim (L_{QPO}/L_{\rm min})^{3/7} \sim 1.2$ and $\nu_{K,QPO}/\nu_s \sim (L_{QPO}/L_{\rm min})^{3/13} \sim 1.1$, in the two models, respectively. Such a low ratio clearly favors the interpretation in which $\nu_{K,QPO}\sim 800$~Hz and $\nu_s\sim 550$~Hz. In the sonic point beat-frequency model (Miller et al. 1997), the innermost disk radius is well within the magnetosphere, implying that the Keplerian frequency at the magnetospheric boundary is $< \nu_{K,QPO}$. In this case all possible combinations of $\nu_s$ and $\nu_{K,QPO}$ are allowed. By using the observed $L_{\rm min}$, a neutron star magnetic field of $B\sim 1-3 \times 10^8$~G (depending on the adopted model of the disk-magnetosphere interaction) is obtained in the case $\nu_s\sim 550$~Hz and $B\sim 2-6 \times 10^8$~G in the case $\nu_s\sim 275$~Hz. Once in the propeller regime, the accretion efficiency decreases further as the magnetosphere expands for decreasing mass inflow rates ($L_{\rm acc}\propto \mdot^{9/7}$). The $\sim 1$~d exponential-like luminosity decline observed with BeppoSAX is considerably faster than the propeller accretion luminosity extrapolated from the first part of the outburst (e.g. the Gaussian profile shown by the dashed line in Fig. 1a). We note here that the spectral transition accompanying the onset of the centrifugal barrier may also modify the irradiation properties of the accretion disk, contributing to X--ray luminosity turn off. Alternatively, an active contribution of the ``propeller'' mechanism or the neutron star spin-down energy dissipated into the inflowing matter cannot be excluded. \subsection{A turned-on rotation-powered pulsar?} It is unlikely that the quiescent luminosity of Aql~X-1 is powered by magnetospheric accretion in the propeller regime. As shown in Fig. 1a, the quiescent X--ray luminosity is probably lower than the minimum magnetospheric accretion luminosity $L_{\rm lc}$ allowed in the propeller phase (this remains true for $B\gsim 6\times 10^7$ G if $\nu_s\sim 550$ Hz, and for $B\gsim 3\times 10^8$ G if $\nu_s\sim 275$ Hz). Moreover the BeppoSAX X--ray spectrum shows a pronounced decrease in the power law to black body flux ratio together with a flattening of the power law component between the fast decay phase and quiescence, suggesting that a transition to shock emission from the interaction of a radio pulsar wind with the matter outflowing from the companion star has taken place. Note that an X--ray spectrum with a slope of $\Gamma \sim 1.5$ has been observed from the radio pulsar PSR~B1259--63 immersed in the wind of its Be star companion. Models of this interaction predict that a power law X--ray spectrum with a slope around $\Gamma \sim 1.5$ should be produced for a wide range of parameters (Tavani \& Arons 1997). The additional soft X--ray component observed during the outburst decay (see Table 2) might be emitted at the polar caps as a result of the residual neutron star accretion in the propeller phase. Note that the equivalent black body radius decreases for decreasing X--ray luminosities, just as it would be expected if the magnetospheric boundary expanded. Alternatively, the black body-like spectral component observed in quiescence could be due to the stre\-aming of energetic particles that hit the polar caps, in close analogy to the soft X--ray component observed, in MSPs, at the weaker level of $\sim 10^{30}-10^{31}\ergs$ (Becker \& Tr\"umper 1997). Assuming a magnetic field in the range derived in section 3.1 (i.e. $B\sim 1-3\times10^8$ G for $\nu_s=550$ Hz and $B\sim 2-6\times10^8$ G for $\nu_s=275$ Hz), we can consistently explain the $\sim 10^{33}\ergs$ quiescent X--ray luminosity as powered by a radio pulsar enshrouded by matter outflowing from the companion star, if the conversion efficiency of spin-down luminosity to X--ray is $\sim 0.1-10$\%. This is consistent with modeling and observations of enshrouded pulsars (Tavani 1991; Verbunt et al. 1996). There are chances of observing a MSP (a simple scaling from MSPs implies a signal at 400 MHz of $\sim 10$ mJy; see Kulkarni et al. 1990), even though the emission would probably be sporadic, like in the case of the pulsar PSR~1744--24A due to the large amount of circumstellar matter (see Lyne et al. 1991; Shaham \& Tavani 1991). In summary Aql~X-1 appears to provide the first example of an old fast rotating neutron stars undergoing a transition to the propeller regime at first, followed by a transition to the radio pulsar regime, as the transient X--ray emission approaches its quiescent level. Therefore, Aql X-1 (and possibly SXRTs in general) likely represents the long-sought ``missing link'' between LMXRBs and recycled MSPs. | 98 | 3 | astro-ph9803303_arXiv.txt |
9803 | hep-ph9803293_arXiv.txt | The M-theory, the strong-coupling heterotic string theory, presents various interesting new phenomenologies. The M-theory bulk axion is one of these. The decay constant in this context is estimated as $F_a\simeq 10^{16}$~GeV. Direct searches for the M-theory axion seem impossible because of the large decay constant. However, we point out that large isocurvature fluctuations of the M-theory axion are obtained in a hybrid inflation model, which will most likely be detectable in future satellite experiments on anisotropies of cosmic microwave background radiation. | It is widely believed that nonperturbative effects play a central role in describing the real world in superstring theories. Horava and Witten~\cite{Horava-Witten} have argued that the strong-coupling heterotic string theory is dual to an M-theory compactified on ${\bf S}^1/{\bf Z}_2$. The low-energy limit of this M-theory is well described by eleven-dimensional supergravity, where the fundamental energy scale is the eleven-dimensional Planck mass $M_{11}\simeq 6\times 10^{16}$~GeV, rather than the four-dimensional one $M_4\simeq 2\times 10^{18}$~GeV~\cite{Witten}. The four-dimensional Planck mass $M_4$ is only an effective parameter appearing at low energies. This M-theory description of strong-coupling heterotic string theory leads to various interesting new phenomenologies. Banks and Dine~\cite{Banks-Dine-1} have pointed out that the bulk moduli fields provide axion candidates in the M-theory compactified further on a Calabi-Yau manifold, and some of the axions survive at low energies, since world-sheet instanton effects are suppressed owing to the large compactification radius in the string tension unit. One of the string axions may acquire its mass dominantly from the QCD anomaly, and it plays the role of the Peccei-Quinn axion~\cite{Peccei-Quinn} in solving the strong $CP$ problem. The decay constant $F_a$ of the M-theory axion is estimated as~\cite{Banks-Dine-1,Choi} \begin{equation} F_a\simeq 10^{16} {\rm GeV}. \end{equation} This value greatly violates the constraint $10^{10} {\rm GeV}\lesssim F_a\lesssim 10^{12}$~GeV, derived in standard cosmology~\cite{Kolb-Turner}. However, this problem may be solved by late-time entropy production through decays of moduli fields~\cite{Kawasaki-Moroi-Yanagida-1,Banks-Dine-2} or a thermal inflaton field~\cite{Lyth-Stewart}. Since a very low reheating temperature, such as $T_R\simeq 1$ - $10$MeV, is required to solve the above problem, it is hard to imagine a consistent production of the lightest supersymmetric (SUSY) particles, and they cannot be the dark matter~\cite{Kawasaki-Moroi-Yanagida-2}. Thus, the M-theory axion seems the most plausible candidate for the dark matter in our universe. If the M-theory axion is indeed the dark matter, fluctuations of the axion density should consist of mixtures of adiabatic and isocurvature modes in general~\cite{Turner,Lyth,ksy}. In this paper we show that a hybrid inflation model proposed by Linde and Riotto~\cite{Linde-Riotto} naturally produces isocurvature fluctuations of the axion comparable to the adiabatic fluctuations that are in the accessible range of future satellite experiments on anisotropies of the cosmic microwave background radiation (CMB). The present analysis, therefore, confirms the previous proposal~\cite{Kawasaki-Yanagida}. | The mixture of isocurvature and adiabatic fluctuations is astrophysically interesting. Since isocurvature fluctuations yield anisotropies of the CMB that are six times larger than those caused by adiabatic fluctuations, mixed fluctuations reduce the amplitude of the power spectrum if the amplitude is normalized by COBE. It is well known that the standard cold dark matter scenario ($\Omega_0=1, h=0.5$) with COBE-normalized pure adiabatic fluctuations predicts density fluctuations that are too large on scales of galaxies and clusters. This problem is avoided if the isocurvature fluctuations are mixed with adiabatic ones, as is pointed in Ref.~\cite{ksy}. Furthermore, it can be shown that for the general flat universe ($\lambda_0 +\Omega_0=1$, with $\lambda_0$ the cosmological constant), the shape and amplitude of the power spectrum are in good agreement with observations if $\alpha \sim 0.05$~\cite{KKSY}. The CMB anisotropies induced by the isocurvature fluctuations can be distinguished from those produced by pure adiabatic fluctuations~\cite{ksy}, because the shapes of the angular power spectrum of CMB anisotropies are quite different from each other on small angular scales. The most significant effect of the mixture of the isocurvature fluctuation is that the acoustic peak in the angular power spectrum decreases. In Fig.~\ref{fig:mix_sample} we show the angular power spectrum for $\alpha=0.05$, $\Omega_0=0.4$ and $h=0.7$ as an example. It is seen that the height of the acoustic peak ($\ell \sim 200$) in the case of mixed fluctuations is greatly reduced compared with the pure adiabatic case. Since the axion decay constant $F_a$ is much higher than $10^{12}$~GeV, a direct search for the M-theory axion is implausible. Therefore, observations of CMB anisotropies by future satellite experiments (MAP~\cite{MAP}, PLANCK~\cite{PLANCK}) are very crucial to test the M-theory axion hypothesis. | 98 | 3 | hep-ph9803293_arXiv.txt |
9803 | astro-ph9803317_arXiv.txt | s{ We report the development of numerical tools for the topological analysis of sub--degree resolution, all--sky maps. Software to be released in the HEALFAST (V0.9) package defines neighbour relationships for the HEALPIX tessellation of the sphere. We apply this routine to a fast extrema search which scales strictly linearly in the number of pixels, $N_{p}$. We also present a highly efficient algorithm for simulating the gradient vector and curvature tensor fields ``on--the--fly'' with the temperature map, needing only of order $N_{p} \log_2 N_{p}$ more operations.} | \label{sec:intro} The Microwave Anisotropy Probe (MAP) and the Planck Surveyor satellite missions promise to produce high--resolution full sky maps of the Cosmic Microwave Background (CMB) within the next decade. These maps will contain a great deal of cosmologically relevant information and therefore present a tremendous opportunity for doing high--precision cosmology. Owing to the unprecedented wealth of data, this opportunity brings with it the challenge of developing tools which help extract this information in manageable processing time. Much attention has been given to the problem of estimating the angular power--spectrum of CMB anisotropies. To illustrate the difficulties one is facing, we note that even choosing the fastest presently available algorithm and allowing for parallelisation and technological advance, the estimation of the angular power--spectrum $C_l$ of CMB anisotropies will take $\gtrsim 6$ years \cite{borrill} for future million pixel maps. While the $C_l$ spectrum is the statistic of choice for a large class of cosmological models where the primordial fluctuations are Gaussian distributed, the possibility remains that another physical mechanism produced structure in the universe. In this case the CMB maps will almost certainly be non--Gaussian and contain more information than merely the power on different angular scales. Failing this, non--linear gravitational effects and foregrounds will produce detectable non--Gaussian signals. Even for Gaussian maps, it has been shown that extrema statistics capture information about cosmological parameters and their theory is well developed \cite{sazhin,BE,VJ,BSMCS}. The importance of studying these and other topological properties of CMB temperature and polarisation has been noted by several authors \cite{topos}. Extrema have the advantage that they dominate the noise and are therefore very robust. As a consequence they have been among the first scientific results reported from recent CMB observations (one of them even at this conference \cite{spot}!). The aim of this talk is to present tools which allow the implementation of a wide range of data analysis and processing techniques on the sphere. \newpage | In this talk we presented tools which are generally applicable for the analysis of topological properties of maps on the sphere such as the detection of extrema, saddle points and zero crossings of a function, even though they can be used more generally for any kind of local analysis of spherical data sets. The computational time required for all sky computations scales strictly as the number of pixels in the map, $N_{p}$. Further, we extend existing simulation tools for Gaussian fields to generate not only temperature maps but also the associated gradient vector and the curvature tensor fields, with the extra computational cost scaling as $N_{p} \log_2 N_{p}$. | 98 | 3 | astro-ph9803317_arXiv.txt |
9803 | astro-ph9803251_arXiv.txt | \rightskip = 0pt \noindent We suggest that the high--velocity clouds (HVCs) are large clouds, with typical diameters of 25 kpc and containing $5 \times 10^7$ solar masses of neutral gas and $3 \times 10^8$ solar masses of dark matter, falling onto the Local Group; altogether the HVCs contain 10$^{11}$ solar masses of neutral gas. Our reexamination of the Local--Group hypothesis for the HVCs connects their properties to the hierarchical structure formation scenario and to the gas seen in absorbtion towards quasars. We begin by showing that at least one HVC complex (besides the Magellanic Stream) must be extragalactic at a distance ${>} 40$ kpc from the Galactic center, with a diameter ${>} 20$ kpc and a mass ${>} 10^8$ solar masses. We then discuss a number of other clouds that are positionally associated with the Local Group galaxies. The kinematics of the entire ensemble of HVCs is inconsistent with a Galactic origin, but implies that the HVCs are falling towards the Local Group barycenter. The HVCs obey an angular--size/velocity relation consistent with the Local Group infall model. We simulate the dynamical evolution of the Local Group. The simulated properties of material falling into the Local Group reproduce the location of two of the three most significant groupings of clouds and the kinematics of the entire cloud ensemble (excluding the Magellanic Stream). We interpret the third grouping (the A, C, and M complexes) as tidally unstable nearby material falling onto the Galactic disk. We interpret the more distant HVCs as dark matter ``mini--halos'' moving along filaments towards the Local Group. Most poor galaxy groups should contain HI structures to large distances bound to the group. We suggest that the HVCs are local analogues of the Lyman--limit absorbing clouds observed against distant quasars. We argue that there is a Galactic fountain in the Milky Way, but that the fountain does not explain the origin of the HVCs. Our analysis of the HI data leads to the detection of a vertical infall of low--velocity gas towards the plane. We suggest that the fountain is a local phenomenon involving neutral gas with characteristic velocities of 6 \kmse rather than 100 \kms. This implies that the chemical evolution of the Galactic disk is governed by episodic infall of metal-poor HVC gas that only slowly mixes with the rest of the interstellar medium. The Local--Group infall hypothesis makes a number of testable predictions. The HVCs should have sub-solar metallicities. Their H$\alpha $ emission should be less than that seen from the Magellanic Stream. The clouds should not be seen in absorption to nearby stars. The clouds should be detectable in both emission and absorption around other groups. We show that current observations are consistent with these predictions and discuss future tests. | \label{sec:intro} Since their discovery in 1963 by \markcite{Muller63}Muller, Oort, \& Raimond, the nature of the high--velocity hydrogen clouds (HVCs) has remained a mystery. HVCs are clouds that deviate from Galactic circular rotation by as much as several hundred kilometers per second. Although astronomers have speculated about the origin of HVCs since their detection, no single explanation has encompassed the vast quantity of data that has been collected on the clouds (see \markcite{WvW97} Wakker \& van Woerden 1997, hereafter WvW97, and references therein). Particularly important is the lack of agreement on a characteristic distance for the clouds, because most of the relevant physical parameters depend on distance to one order or another. In the 1970s, a well--defined subset of the clouds was identified as a tidal stream associated with the Magellanic Clouds (\markcite{Mathewson74}Mathewson, Cleary, \& Murray 1974), but since then no consensus has arisen regarding the nature of the remaining clouds which constitute the majority of HVCs. In this paper, we suggest that the HVCs represent infall of the intergalactic medium onto the Local Group. Previous authors have explored the possibility that the HVCs are infalling primordial gas (\markcite{Oort66,Oort70}Oort 1966, 1970) and have associated the HVCs with the Local Group (\markcite{Verschuur69}Verschuur 1969; \markcite{Kerr69}Kerr \& Sullivan 1969; \markcite{Arp85}Arp 1985; \markcite{Bajaja87}Bajaja, Morras, \& P\"oppel 1987; \markcite{Arp91}Arp \& Sulentic 1991), but subsequent observations always produced fundamental difficulties for the particular models considered. Here, we assemble evidence based on new general--purpose surveys of atomic hydrogen gas by \markcite{Stark92}Stark et al. (1992) and by \markcite{Hartmann97}Hartmann \& Burton (1997), and on the HVC surveys by \markcite{Hulsbosch88}Hulsbosch \& Wakker (1988) and by \markcite{Bajaja85}Bajaja et al. (1985), and consider theoretical arguments in the context of modern cosmology. We argue that the clouds are matter accreting onto the Local Group of galaxies. Their velocities would thus largely reflect the motion of the clouds in the gravitational potential of the Local Group and the motion of the LSR about the Galactic center. We suggest that the clouds represent the building blocks from which the Local Group was assembled and that they continue to fuel star formation in the disk of the Milky Way. The evidence is presented as follows. In Section 2, we review some of the observed properties of the high--velocity clouds, and indicate those which are not consistent with a Galactic origin for the HVCs. In Section 3, we detail the observations entering our analysis. In Section 4, we discuss the stability of the HVCs against gravitational collapse and against Galactic tidal forces, and suggest that these considerations imply that the most of the clouds are extragalactic at distances typical of the Local Group. The stability criteria imply, however, that the largest clouds are nearby and possibly are interacting with the Milky Way. In Section 5, we discuss three individual clouds, one of which must be beyond the disk of the Milky Way, and two others that appear to be associated with M31 and M33, suggesting that at least some of the HVCs may be extragalactic. We identify a subset of the HVCs centered near the barycenter of the Local Group, and show that its kinematics as well as that of the entire HVC ensemble are well described as being at rest with respect to the Local--Group Standard of Rest (LGSR); the kinematics are thus inconsistent with a Galactic origin. The entire HVC ensemble is also shown to exhibit an angular--size/velocity relation consistent with membership in the Local Group. In Section 6, we simulate the accretion history of the Local Group and show that the simulation accounts for the observed distribution and kinematics of the HVC ensemble. The agreement between the simulation and the observations supports inferences about similarities between the Local Group HVCs and the Ly--$\alpha$ absorbing clouds observed toward quasars. We show that the velocity extrema observed for the HVCs are consistent with their membership in the Local Group. In Section 7, we discuss the distances and abundances of the HVCs in the context of the Local Group HVC hypothesis, and show that the hypothesis is consistent with all of the observations made to date. We review extragalactic HI searches for HVCs which have revealed clouds with properties similar to those we derive, in about the expected numbers. In Section 8, we discuss the implied mass accretion rate, and implications for the chemical evolution of the disk of the the Milky Way. We also present evidence for the Galactic fountain in low--velocity HI which suggests that the HI disk of the Milky Way is not in hydrostatic equilibrium. In Section 9, we conclude by discussing predictions and future tests of the model, and summarize the principal arguments made in this paper. | \label{sec:finale} Most cosmologists believe that galaxy formation is a hierarchical process: galaxies grow by accreting small clouds of gas and dark matter. This process is a continuing one and we expect that galaxies and groups are currently accreting new clouds. We simulated this process for the Local Group and found that properties of the accreted clouds are similar to certain properties of the high--velocity--cloud phenomenon (excluding the Magellanic Stream HVCs): $\bullet $ Most of the HVCs are located either near the general direction of M31, towards the barycenter of the Local Group, or in the antibarycenter direction, some 180\dege from the direction of M31 (see Figure~\ref{fig:lb}). $\bullet $ HVCs have chemical abundances similar to that of intra--group gas, and different from the abundances characteristic of the inner Galaxy. If HVCs were ejected from the inner Galaxy as part of a Galactic fountain, then their metal abundance would exceed the solar value, and this is not observed. $\bullet $ HVCs have an angular--size/velocity relation that is consistent with the clouds being nearly self--gravitating, and at a distance of $\sim 300$ kpc. If the Local--Group HVC hypothesis discussed in this paper is correct, then studies of HVCs can directly probe the process of galaxy formation. The validity of this hypothesis can be tested by a number of future observations: $\bullet $ Most observations of nearby galaxies would not have detected the gas clouds that are equivalent to the HVCs. Moreover, many HI maps of external galaxies extend just beyond the Holmberg radius. Our discussion would have the typical HVC located nearly a Mpc from the galactic center. It will be interesting to test our hypothesis with deep HI observations of isolated groups and filaments, searching for HI clouds associated with groups, rather than with individual galaxies. $\bullet $ Lyman--limit clouds, which are seen in absorption towards distant quasars, have column densities similar to those of the HVCs. Observations of nearby Lyman--limit and Lyman--$\alpha $ systems show that they are not all associated directly with individual galaxies, but rather with groups of galaxies (\markcite{Oort81}Oort 1981; \markcite{Stocke95}Stocke et al. 1995; \markcite{vanGorkom96}van Gorkom et al. 1996; \markcite{Rauch96}Rauch, Weymann, \& Morris 1996). In the scenario outlined in this paper, we expect that these clouds would have properties similar to those of the local HVCs. Thus, it would be interesting to use STIS to look for lower--column--density high--velocity HI clouds, which would correspond to the Lyman--$\alpha $ clouds. $\bullet $ The simulations predict large amounts of gas accreting onto M31 and the Local Group from the region of space beyond M31, under the gravitational attraction of both M31 and our own Galaxy. Because this gas is several Mpc away, the gas clouds are expected to have small angular sizes and relatively low column densities. Deep HI observations in the M31 direction should be able to detect this gas. The hypothesis central to this paper, namely that HVCs are at distances of around 1 Mpc, would be falsified by the detection of absorbtion in an HVC seen against stars in the Milky Way halo in the direction of M31 or in the anti--M31 direction. On the other hand, further measurements of low levels of H$\alpha$ emission towards these HVCs will strengthen the case for their extragalactic nature. | 98 | 3 | astro-ph9803251_arXiv.txt |
9803 | astro-ph9803067_arXiv.txt | A solitary millisecond pulsar, if near the mass limit, and undergoing a phase transition, either first or second order, provided the transition is to a substantially more compressible phase, will emit a blatantly obvious signal---spontaneous spin-up. Normally a pulsar spins down by angular momentum loss to radiation. The signal is trivial to detect and is estimated to be ``on'' for 1/50 of the spin-down era of millisecond pulsars. Presently about 25 solitary millisecond pulsars are known. The phenomenon is analogous to ``backbending'' observed in high spin nuclei in the 1970's. | The formation of a new phase of matter, a softer one than nuclear matter, may cause a rapidly rotating pulsar to produce a prolonged signal that is dramatic, easy to detect and easy to understand \cite{glen97:a}. The most plausible high density phase transition is deconfinement as predicted by QCD \cite{asymptotic}. The signal I will describe will occur for either a first or second order transition so long as it is accompanied by a sufficient softening of the \eosp. (Cf. Fig.\ \ref{eos}.) Strictly speaking we do not even know that quarks can be deconfined under extreme conditions or otherwise. It is an `expectation' based on the QCD property of asymptotic freedom \cite{asymptotic}. We would like to prove that this phase is a possible phase of matter. If so, it would have pervaded the very early universe, but quark confinement in hadrons occurred at an early time and the thermal equilibrium that existed then leaves no signal today. \begin{figure}[htb] \vspace{-.25in} \begin{minipage}[t]{80mm} \makebox[79mm] {\psfig{figure=ps.qm1,width=3in,height=3.6in} } \caption {The \eos (labeled `Hybrid') of neutron star matter with a first order deconfinement phase transition. The normal phase contains nucleons, hyperons and leptons in equilibrium. The mixed phase contains as well, the three light flavor quarks. Comparison is made with a case in which deconfinement is not taken account of (labeled `n+p+H'). (Nuclear properties include the observed binding, saturation density, symmetry energy and $K=300$ MeV, $m^\star_{{\rm sat}}/m=0.7$) \label{eos_k300_y_h} \label{eos} } \end{minipage} \hspace{\fill} \begin{minipage}[t]{75mm} \makebox[74mm] {\psfig{figure=ps.qm2,width=3in,height=3.6in}} \caption { Density profiles of two stars of the same mass $M=1.42 \msun$ but differing composition; (1) Hyperon star (neutron-proton-hyperon-lepton), (2) Hybrid (a pure quark matter core surrounded by mixed phase and outer pure hadronic confined phase). \Eoss as in Fig.\ \protect\ref{eos_k300_y_h}. Interior differences are dramatic but not directly measureable. (For a description of neutron star matter and relativistic stars see Ref.\ \protect\cite{book}. \label{prof_k240} } \end{minipage} \end{figure} From the balance of gravitational and centrifugal forces on a particle at the surface of rapidly rotating stars such as the millisecond pulsars, we know that the central density is a few times nuclear, the same range of energy densities as are expected to be produced in relativistic nuclear collisions. Let us assume that the critical deconfinement density occurs in the density range spanned by spherical stars and therefore in the population of slow pulsars to which the Crab belongs and of which there are about 800 presently known. In this case newly born neutron stars have a quark core essentially from birth. But we have no way to tell if this is indeed the case. Models of neutron stars having different composition generally differ in the range of masses and radii permitted, and their density profiles may be very different (cf. Fig.\ \ref{prof_k240}). However as far as measureable properties are concerned, ordinary neutron stars and hybrid stars (neutron stars with a quark core) are practically indistinguishable. Cooling rates are presumeably sensitive to the internal composition, but theoretical estimates are very uncertain. I will sketch the generally accepted evolutionary life of pulsars \cite{heuvel91:a}. There are two distinct populations of pulsars (see Figs.\ \ref{period} and \ref{pulsar}), the canonical pulsars (about 800 of them now known) with periods between 1/50 sec and 8 sec, and the millisecond pulsars (about 50), which are believed to be more evolved than the former. Stars are born into the first of these populations, and may, on a very long time-scale, evolve into the second. (See Fig.\ \ref{pulsar}.) \begin{figure}[htb] \vspace{-.25in} \begin{center} \begin{minipage}[t]{130mm}\hspace{.8in} \makebox[79mm] {\psfig{figure=ps.qm3,width=4.2in,height=3.5in} } \caption { Distribution of pulsar periods. The lower group consists of `recycled' millisecond pulsars, the higher to the canonical pulsars in the first stage of their evolution (see Fig.\ \protect\ref{pulsar}). \label{period} } \end{minipage} \end{center} \vspace{-.35in} \end{figure} As the stellar core of a luminous star collapses to form a neutron star, it is spun up by conservation of angular momentum and acquires an enormous magnetic field of $10^{12}{\rm~to~}10^{13}$ gauss because of flux conservation. The star is born as a rotating magnetic dipole. It has a tremendous store of rotational energy that will keep it spinning for 10 million years. The electromagnetic radiation beamed along the spinning dipole is what we see as pulsed radio emission once each rotation, if as observers, we lie on the cone swept out by the beam. \begin{figure}[htb] \vspace{-.25in} \begin{center} \begin{minipage}[t]{130mm}\hspace{.8in} \makebox[79mm] {\psfig{figure=ps.qm4,width=4.2in,height=3.5in} } \vspace{-.35in} \caption { Pulsars are born with field $B\sim 10^{12} {\rm~to~} 10^{13}$ gauss and evolve toward the right. Periods change with time as $\sim 1/P$ and so pulsars accumulate at large $P$. They become radio silent in about $10^7$ years and remain stagnant until they capture a companion star, or unless they had one all along. Accretion from the less dense companion spins them up along a line like `silent evolution'. A combination of the now weaker field but higher frequency turns them on again as `recycled' millisecond pulsars. They then again evolve toward shorter period, but now on a very long time-scale because of the weaker field. \label{pulsar} } \end{minipage} \end{center} \vspace{-.35in} \end{figure} In a plot of magnetic field $B$ vs period $P$ (Fig.\ \ref{pulsar}), stars move from top left to right because of loss of angular momentum to radiation. They disappear as active pulsars when a combination of angular velocity and field strength is insufficient to produce radiation. It takes about $10^7$ years to complete this first phase. However, either from birth, or afterward, the star may have or capture a lower density companion as is often evidenced by the presence of an orbiting white dwarf. During an accretion era, the compact star is spun up by infalling matter that it tears off from its less dense companion. It looses some magnetic field during accretion, perhaps by ohmic resistance during the long radio silent era. In this `silent' era the the star moves diagonally from top right to bottom left in the $B-P$ diagram. The neutron star becomes centrifugally flattened as it approaches millisecond periods. The central density falls. The core of quark matter shrinks as quarks recombine to form baryons. The quark core may disappear altogether. The silent neutron star, having completed a part of its life cycle, turns on again as a millisecond pulsar of low magnetic field when the lower field but higher angular velocity can once again produce radiation. Presently 50 such pulsars have been discovered, half of which still have a binary companion. The number of millisecond pulsars presently known is believed to be a fraction of the total population because of search selection effects. Millisecond pulsars, which have weaker fields ($10^8{\rm~to~}10^9$ gauss), spin down very slowly since the deceleration is proportional to $B^2$. Their characteristic age is $P/2\dot{P}\sim 10^9$ years. The central density is initially centrifugally diluted but as it spins down, the central density will rise again and the critical density will be reached, first at the center, and then in an expanding region. The growth of the central region of deconfined matter is paced by the slow spin-down, slow because of the coupling of rotation of the stellar magnetic dipole to electromagnetic processes. Stiff nuclear matter is being replaced in the core by highly compressible quark matter. The weight of the overlaying layers of nuclear matter weigh down on the core and compress it. Its density rises. The star shrinks---mass is redistributed with growing concentration at the center. The by-now more massive central region gravitationally compresses the outer nuclear matter even further, amplifying the effect. The density profile for a star at three angular velocities, (1) the limiting Kepler velocity which is stretched in the equatorial plane and centrally diluted, (2) an intermediate angular velocity, and (3) a non-rotating star, are shown in Fig.\ \ref{prof_k300b180}. We see that the central density rises with decreasing angular velocity by a factor of three and the equatorial radius decreases by 30 percent. In contrast, for a model for which the phase transition did not take place, the central density would change by only a few percent \cite{weber90:d}. The phase boundaries are shown in Fig.\ \ref{omega_r_k300B180} from the highest rotational frequency to zero rotation. \begin{figure}[htb] \vspace{-.5in} \begin{minipage}[t]{80mm} \makebox[79mm] {\psfig{figure=ps.qm5,width=3in,height=3.6in} } \caption { Mass profiles as a function of equatorial radius of a star rotating at three different frequencies, as marked. At low frequency the star is very dense in its core, having a 4 km central region of highly compressible pure quark matter. At intermediate frequency, the pure quark matter phase is absent and the central 8 km is occupied by the mixed phase. At higher frequency (nearer $\Omega_K$) the star is relatively dilute in the center and centrifugally stretched. Inflections at $\epsilon=220$ and $950$ are the boundaries of the mixed phase. \label{prof_k300b180} } \end{minipage} \hspace{\fill} \begin{minipage}[t]{75mm} \makebox[74mm] {\psfig{figure=ps.qm6,width=3in,height=3.6in} } \caption { Radial boundaries at various rotational frequencies separating (1) pure quark matter, (2) mixed phase, (3) pure hadronic phase, (4) ionic crust of neutron rich nuclei and surface of star. The pure quark phase appears only when the frequency is below $\Omega \sim 1370$ rad/s. Note the decreasing radius as the frequency falls. The frequencies of two pulsars, the Crab and PSR 1937+21 are marked for reference. \label{omega_r_k300B180} } \end{minipage} \end{figure} The redistribution of mass and shrinkage of the star change its moment of inertia and hence the characteristics of its spin behavior. The star must spin up to conserve angular momentum which is being carried off only slowly by the weak electromagnetic dipole radiation. The star behaves like an ice skater who goes into a spin with arms outstretched, is slowly spun down by friction, temporarily spins up by pulling the arms inward, after which friction takes over again. It is that simple to describe, and that is the blatant signal I mentioned---the spontaneous spin-up of an isolated millisecond pulsar that is radiating angular momentum and ought otherwise to be slowing down. | An isolated millisecond pulsar will spin up over an epoch of $2 \times10^7$ years out of a spin-down life of $10^9$ years if it undergoes a phase transition obeying the two conditions (i) the transition causes a substantial softening of the \eosp, and (ii) the critical density is attained in stars very near the mass limit. The spin-up epoch, compared to the spin-down life of the pulsar, corresponds to an `event rate' of 1/50. The determination of whether a pulsar is spinning up or down is trivial. Of the presently known millisecond pulsars, about 25 are isolated. We are approaching the moment of truth for this observable signal of a phase transition. We have emphasized that the transition need not be of first order so long as it is accompanied by a sufficient softening of the \eosp. We do not have a measure of what we mean by this. Of course our model does possess the requisite softening, or our results would not have exhibited backbending. If no pulsar is observed to produce the signal, little is learned. Just as in the search for deconfinement in high energy nuclear collisions, failure to observe a signal does not inform us that the deconfined phase does not exist. | 98 | 3 | astro-ph9803067_arXiv.txt |
9803 | astro-ph9803192_arXiv.txt | We construct evolutionary synthesis models for simple stellar populations using the evolutionary tracks from the Padova group (1993, 1994), theoretical colour calibrations from Lejeune et al. (\cite{lejeune}, \cite{lejeune1}) and fit functions for stellar atmospheric indices from Worthey et al. (\cite{worthey}). A Monte-Carlo technique allows us to obtain a smooth time evolution of both broad band colours in UBVRIK and a series of stellar absorption features for Single Burst Stellar Populations ({\bf SSPs}). We present colours and indices for SSPs with ages from $1 \cdot 10^9$ yrs to $1.6 \cdot 10^{10}$ yrs and metallicities $[M/H]$ = -2.3, -1.7, -0.7, -0.4, 0.0 and 0.4. Model colours and indices at an age of about a Hubble time are in good agreement with observed colours and indices of the Galactic and M 31 GCs. | Colour distributions of Globular Cluster ({\bf GC}) systems are observed for a large number of early-type galaxies (E, S0, dE, cD) using ground-based Washington or HST broad band photometry. In most cases double-peak or broad/mul\-ti-peak colour distributions are seen (e.g. Zepf \& Ashman \cite{zepf}, Elson \& Santiago \cite{elson}, Kissler-Patig \etal \cite{kissler}). If the different colour subpopulations of GCs are formed in different events then they may contain clues to the formation history of their parent galaxies. For example a two-peak colour distribution may result, if in addition to a primary initial collapse population of GCs, a secondary population of GCs were formed either in a merger-induced starburst (Schweizer \cite{schweizer}, Ashman \& Zepf \cite{ashman}, Fritze -- v. Alvensleben \& Gerhard \cite{fritze1}, Fritze -- v. Alvensleben \& Burkert \cite{fritze2}) or else in some distinct secondary phase of cluster formation within the original galaxy (Forbes \etal \cite{forbes}). Likewise the broad or multi-peaked colour distribution often observed in GC systems around cD galaxies may point to a series of GC formation events during the hierarchical assembly of the parent galaxy or to some protracted GC formation or accretion mechanism. A well-known difficulty with the interpretation of colour distributions is the degeneracy of colours with respect to age and metallicity. While for Washington photometry there are well established and reliable calibrations of colours in terms of metallicity, the situation with HST broad band observations of GC systems is less clear. A better understanding of the formation of composite GC systems would be possible if separate age and metallicity distributions could be disentangled from an observed colour distribution. A second issue concerns the interpretation of colours for young star cluster systems detected with HST in many interacting and starburst galaxies. The question is, if these YSC systems -- at least some fraction of them -- are the progenitors of GC systems. In an attempt to answer this question star clusters are being imaged with HST in an age sequence of interacting galaxies -- from early stages of interaction through merger remnants up to E/S0s (eg. Schweizer \etal \cite{schweizer1}, Whitmore \etal \cite{whitmore}, Miller \etal \cite{miller}). With 10 m class telescopes, spectroscopy of the brighter members of young star cluster populations is becoming possible (Kissler-Patig \etal \cite{kissler1}, Brodie \etal \cite{brodie}, but see also 4 -- 5 m class spectra by Schweizer \& Seitzer \cite{schweizer} or Zepf \etal \cite{zepf1}). Spectroscopy will only be possible for a subsample of YSCs. Thus the determination of ages and metallicities from broad band colors will still be necessary. It is thus desirable to study the evolution of broad band colours and absorption indices for single burst stellar populations of various metallicities using the most recent and complete stellar evolutionary tracks as well as careful colour and index calibrations.This allows one to obtain theoretical calibrations of broad band colours and indices in terms of metallicity over the full range of ages under investigation, i.e. from $10^7$ yr to a Hubble time. Since theoretical calculations for the evolution of stars are only available for a discrete grid of masses, some means for obtaining a smooth evolution of the composite population is needed. Applying the tracks as they are would create discontinuities because all stars of a given mass would reach the giant branch at the same time, dominating the integrated light until they die. This effect is large for populations with stars that have about the same age. The effect also increases with age of the whole population, since the differences in both the lifetimes and luminosities between the main sequence and the later stages increase with decreasing mass. For this work, we use the \emph{Monte Carlo} method to bypass this problem while still avoiding the interpolation of tracks with its accompaining danger of creating artificial states. This is described in detail in section \ref{sec_nummethod}. The star formation history of any stellar system can be described by a superposition of SSP models of different ages and metallicities. An example of this is given by Cellone \& Forte et al (\cite{cellone}) in their study of Low Surface Brightness galaxies or Contardo et al. (\cite{contardo}) who investigate the formation and evolution of galaxies in a cosmological scenario. | In this work we present Monte-Carlo evolutionary synthesis models for SSPs which cover a wide range in metallicity, from $Z=0.0001$ to $Z=0.05$, using most recent and complete sets of input physics: stellar evolutionary tracks for stellar masses from $0.08 M_\odot$ to $120 M_\odot$, including post - helium flash evolution and mass loss, model atmosphere libraries also covering late stellar types and giving colours from $U$ to $K$ in agreement with observations and empirical calibrations for a series of absorption indices. We obtain theoretical calibrations of colours and indices in terms of metallicity which for model ages of 10-15 Gys agree closely with observations of GCs. The theoretical calibrations extend beyond the range of observed GCs, i.e. to a metallicity up to $[M/H] \leq 0.4$. Moreover, our models provide these theoretical calibrations for all ages from cluster formation to 15 Gyrs and thus can also be applied in the interpretation of young star cluster systems observed in many interacting and starburst galaxies. The complete model files are available via WWW on {\tt http://www.uni-sw.gwdg.de/\~\ okurth/ssp.html}. There are also models with different parameters for the IMF and for mass loss available. | 98 | 3 | astro-ph9803192_arXiv.txt |
9803 | astro-ph9803100_arXiv.txt | I present several simple figures to illustrate cosmology and structure formation in a nutshell. Then I discuss the following argument: if we assume that $\Omega_{\Lambda} = 0$ then the CMB results favor high $\Omega_{m}$ while the supernova results favor low $\Omega_{m}$. This large inconsistency is strong evidence for the incorrectness of the $\Omega_{\Lambda}=0$ assumption. Finally I discuss recent CMB results on the slope and normalization of the primordial power spectrum. | The Big Bang model became the standard cosmological model soon after the discovery of the cosmic microwave background (CMB). The Big Bang model has a hot, dense early epoch (see Figure 1) when nucleosynthesis occurred. It also has an opaque surface that can naturally produce the Planckian spectrum of the CMB. The Steady State universe was not hotter in the past, has no epoch of Steady State Nucleosynthesis and has no opaque surface to produce the CMB. Gravitational collapse is the leading model of structure formation (Figure 2). Slight over-densities are gravitationally unstable and collapse under their own self-gravity. In an alternative family of models, structure forms from topological defects. In gravitational collapse models CMB anisotropies larger than $\sim 1$ degree are acausal and rely on inflation to explain their existence. In defect models these large anisotropies are close by, causal and sub-horizon sized. \thefirstfig \clearpage \thesecondfig \setcounter{figure}{2} \begin{figure}[hbt] \centerline{\psfig{file=wearehere.eps,height=70mm,width=110mm,angle=-90}} \vspace{0cm} \caption{Galaxies are CMB anisotropies are Quantum fluctuations. According to the inflationary scenario, quantum fluctuations of a scalar field are the origin of all structures. These quantum fluctuations are not caused by any preceeding event in the same sense as radioactive decay or quantum tunneling are not caused. They are non-deterministic prime movers. Inflation of the universe by a factor of more than $10^{26}$ transforms these quantum fluctuations into super-horizon classical density fluctuations. On their way to becoming galaxies we can monitor their progress by looking at CMB maps.} \end{figure} \clearpage One of the most important questions in cosmology is: What is the origin of all the galaxies, clusters, great walls, filaments and voids we see around us? The inflationary scenario provides the most popular explanation for the origin of these structures: they used to be quantum fluctuations. Figure 3 illustrates the metamorphosis of quantum fluctuations to CMB anisotropies to galaxies. Primordial quantum fluctuations of a scalar field get amplified and evolve to become classical seed perturbations and eventually large scale structure. This process can be monitored by CMB observations since matter fluctuations produce temperature fluctuations in the CMB: $\frac{\delta \rho}{\rho} \propto \frac{\Delta T}{T}$. How does a particular fluctuation know whether it will become a spiral or an elliptical galaxy? Does the density and irregularity of its environment determine its morphology by controlling its angular momentum and the amount of merging? With a full understanding of galaxy formation we may be able to look at CMB cold spots and their neighborhoods and predict where they will end up in the Hubble tuning fork diagram of galaxy types. The distribution of morphological types at high redshift discussed by Driver in these proceedings would then be a derivable function of the characteristics of the CMB anisotropies. \thefourthfig \subsection{There is no scale beyond which the universe is homogeneous} It has been claimed that some recent, deep, galaxy redshift surveys have reached the scale at which the Universe becomes homogeneous. Strictly speaking however there is no scale beyond which the universe is homogeneous. The amplitude of the density contrast ($\delta \rho / \rho \propto k^{3}P(k)$ decreases for larger scales but is never zero. A more meaningful question is: Where is the turnover in the power spectrum? This turnover is due to a suppression of growth of a given k mode by $k^{4}$ relative to modes which enter the horizon during matter domination (assuming $\oo = 1$). Thus, the horizon scale at matter-radiation equality is an important diagnostic of this fundamental scale. See Figure 4. Lineweaver \& Barbosa (1998) have used current CMB anisotropy measurements to determine the position of the adiabatic peak in the CMB spectrum under the assumption of open or critical density CDM dominated universes: $\ell_{peak} = 260^{+30}_{-20}$. Figure 5 illustrates how harmonic sound bumps appear in the CMB power spectrum driven by the wells and valleys of the CDM potentials. The epoch when matter and radiation densities are equal has a redshift of $z_{eq}$ while decoupling occurs at $z_{dec}$. The number of oscillations between $z_{eq}$ and $z_{dec}$ and thus the phase of the oscillations at $z_{dec}$ is determined by i) the physical size of the potential well, ii) the speed of sound and iii) the time interval between $z_{eq}$ and $z_{dec}$. \vspace{0.7cm} \thefifthfig \clearpage | 98 | 3 | astro-ph9803100_arXiv.txt |
|
9803 | astro-ph9803336_arXiv.txt | This paper presents the preliminary results of the ESO Imaging Survey (EIS), a public survey being carried out by ESO and member states, in preparation for the VLT first-light. The survey goals, organization, strategy and observations are discussed and an overview is given of the survey pipeline developed to handle EIS data and produce object catalogs. A report is presented on moderately deep I-band observations obtained in the first of four patches surveyed, covering a region of 3.2 square degrees centered at $\alpha \sim 22^h 40^m$ and $\delta =-40^\circ$. The products available to the community, including pixel maps (with astrometric and photometric calibrations) and the corresponding object catalogs, are also described. In order to evaluate the quality of the data, preliminary estimates are presented for the star and galaxy number counts, and for the angular two-point correlation function obtained from the available data. The present work is meant as a preview of the final release of the EIS data that will become available later this year. | With the advent of very large telescopes, such as the VLT, a largely unexplored domain of the universe becomes accessible to observations which may dramatically enhance on our understanding of different physical phenomena, in particular the origin and evolution of galaxies and large scale structures. In the next few years a wide array of 8-m telescopes will become available world-wide. Among these, the European VLT project is particularly striking because of its four 8-m telescopes and an impressive array of complementary instrumentation. Viewed as a unit, the VLT provides great flexibility by combining complementarity for certain programs with multiplexing capabilities for others. First-light for the VLT is scheduled for May 1998, with regular science operation starting in April 1999. In order to take full-advantage of the VLT from the start of its operation, ESO and its Observing Programmes Committee (OPC) decided to coordinate an imaging survey to provide candidate targets well-suited to the first set of VLT instruments. The ESO Imaging Survey (EIS) has been conceived as a collaborative effort between ESO and astronomers in its member states. Following the recommendation of the OPC, the survey has been overseen by a Working Group (WG). The EIS WG is composed of leading experts in different fields and has the responsibility of defining the survey science goals and strategy, and monitor its progress. In order to carry out the survey a dedicated team was assembled, starting March 1997. To stimulate cooperation between ESO and the astronomical community of the member states, EIS has sponsored the participation of experts as well as students and post-docs from the community in the development of software, observations and data reduction. As described by Renzini \& da Costa (1997) (see also ``http://www.eso.org/eis''), EIS consists of two parts: EIS-wide to search for rare objects (\eg distant clusters and quasars) and EIS-deep to define samples of high-redshift galaxies. These science goals were chosen to match as well as possible the capabilities of the first VLT instruments, FORS, ISAAC and UVES. EIS is also an essential first step in the long-term effort, currently underway at ESO, to provide adequate imaging capabilities in support of VLT science (Renzini 1998). The investment made in EIS will be carried over to a Pilot Survey utilizing the ESO/MPIA 2.2m telescope at La Silla, with its new wide-field camera. This Pilot Survey, which will follow the model of EIS, has been recommended by the EIS WG and is being submitted to the OPC. The goal of this paper is to describe the characteristics of I-band observations carried out in the fall of 1997 over a region of 3.2 square degrees (EIS patch A, da Costa \etal 1998a) and of the corresponding data products, in the form of calibrated images and single frame catalogs. These products have been made publicly available through the ESO Science Archive, as a first step towards the full distribution of the EIS data. The purpose of the present release is also to provide potential users with a preview of the data, which may help them in the preparation of VLT proposals, and to encourage the community to provide constructive comments for the final release. It is important to emphasize that due to time limitations the results presented here should be viewed as preliminary and improvements are expected to be made before the final release of the EIS data later this year. In section 2, a brief description is presented of the criteria adopted in the field selection, the strategy of observations and the characteristics of the data in patch A already completed. It also describes the filters used, the definition of the EIS magnitude system and its relation to other systems, and the data used for the photometric calibration of the survey. In section 3, a brief description of the data reduction pipeline is presented, followed in section 4 by a description of the data products made publicly available in this preliminary data release. In section 5, the algorithms used to detect and classify objects, and the information available in the catalogs being distributed are described. Preliminary results from a scientific evaluation of the data is presented in section 6. In section 7, future plans are presented, followed in section 8 by a brief summary. | The ESO Imaging Survey is being carried out to help the selection of targets for the first year of operation of VLT. This paper describes the motivation, field and filter selection, and data reduction pipeline. Data for the first completed patch, in the form of astrometric and photometric calibrated pixels maps, single-frame catalogs, on-line coadded section images and further information on the project are available on the World Wide Web at ``http://\-www.eso.org/eis''. Preliminary evaluation of the data shows that the overall quality of is good and the completeness limit of the extracted catalogs is sufficiently deep to meet the science requirements of EIS. Furthermore, the results for the other patches should improve as the observing conditions were considerably better than those in the period patch A was observed. The final and complete release of the data products of EIS is scheduled as follows: 1) EIS-wide, except the U-band : July 31, 1998, before the first call for proposals for the VLT; 2) EIS-deep and EIS-wide U-band on December 31, 1998. | 98 | 3 | astro-ph9803336_arXiv.txt |
9803 | astro-ph9803046_arXiv.txt | Thermal radiation from surfaces of several radio pulsars has been detected \linebreak recently in soft X-rays by {\it ROSAT\/} and {\it ASCA\/} (see Becker and Tr\"umper,~1998). A number of the point-like X-ray sources has also been discovered and identified as radio-silent isolated cooling neutron stars (NSs) (see Caraveo et al.,~1996; Walter et al.,~1996; Vasisht et al.,~1997). In several cases, the observations seem to be better explained if the NSs possess an envelope of matter of low atomic weight, presumably hydrogen (Page,~1997), and if NS atmosphere models are applied to fit the data (Pavlov and Zavlin,~1998). There are indications that this may be the case for NSs with different magnetic fields $B$: from weak, $B\ll10^{10}$~G (Rajagopal and Romani,~1996; Zavlin and Pavlov,~1998; Zavlin et al.,~1998), to strong, $B\sim10^{11-13}$~G (Page et al.,~1996; Zavlin et al.,~1998), and superstrong, $B>10^{14}$~G (Heyl and Hernquist,~1997). To justfy this, one needs in further observations and advanced atmosphere models of NSs with high $B$. Thermal motion of atoms accross the magnetic field induces an electric field in the frame of the atom. This affects atomic structure, atmospheric thermodynamics and opacities (Ventura et al.,~1992; Pavlov and M\'esz\'aros,~1993), being a critical point in the construction of models and data interpretation. Here we study this problem for hydrogen plasma at temperatures $T\sim 10^{5.5}-10^{6}$~K, densities $\rho\sim10^{-3}-10^1{\rm~g~cm}^{-3}$, and magnetic fields $B\sim 10^{12}-10^{13}$~G, typical for atmospheres of middle-aged cooling NSs. We construct an analytic model of the plasma free energy and derive a generalized Saha equation which is used to obtain the opacities. | 98 | 3 | astro-ph9803046_arXiv.txt |
||
9803 | astro-ph9803270_arXiv.txt | Warped \ion{H}{1} gas layers in the outer regions of spiral galaxies usually display a noticeably twisted structure. This structure almost certainly arises primarily as a result of differential precession in the \ion{H}{1} disk as it settles toward a preferred orientation in an underlying dark halo potential well that is not spherically symmetric. In an attempt to better understand the structure and evolution of these twisted, warped disk structures, we have adopted the ``twist--equation'' formalism originally developed by Petterson (1977) to study accretion onto compact objects. Utilizing more recent treatments of this formalism, we have generalized the twist--equation to allow for the treatment of non--Keplerian disks and from it have derived a steady--state structure of twisted disks that develops from free precession in a nonspherical, logarithmic halo potential. We have used this steady--state solution to produce \ion{H}{1} maps of five galaxies (M83, NGC 300, NGC 2841, NGC 5033, NGC5055), which match the general features of the observed maps of these galaxies quite well. In addition, the model provides an avenue through which the kinematical viscosity of the \ion{H}{1} disk and the quadrupole distortion of the dark halo in each galaxy can be quantified. This generalized equation can also be used to examine the time-evolutionary behavior of warped galaxy disks. | In simplest terms, spiral galaxy disks can be described as geometrically thin, flat, and circular. We understand that spiral disks are geometrically thin because the gas of which they are composed is cold (the sound speed of the gas is much smaller than its circular orbital velocity); and they are both circular and flat because, being dissipative, the gas is fairly efficient at both minimizing out-of-the-plane motions and radial excursions that would lead to departures from circular orbits. Describing spiral disks as {\it perfectly} circular and flat is clearly an oversimplification, however. In addition to the nonaxisymmetric structures that are obvious in optical photographs of many spiral disks, 21-cm maps of the projected velocity fields of spiral disks often reveal isovelocity contours that are significantly twisted (\cite{RLW}; \cite{RN78}; \cite{RCC79}; \cite{N80a}, 1980b; \cite{B81}; \cite{S85}). Kinematical tilted-ring models have been constructed in an effort to explain the presence of such twists in the velocity maps of \ion{H}{1} disks. The models indicate that the outer regions of many normal spiral disks are significantly warped out of the principal plane that is defined by the optically visible, central portion of each galaxy. The line of nodes that defines the intersection of adjacent rings of gas in these kinematical models also usually must be twisted significantly as a function of radius in order to explain the observed contour maps (for recent reviews, see \cite{Briggs90}; \cite{Bosma91}; \cite{CTSC93}, hereafter CTSC). It is not particularly surprising that many galaxies are observed to possess extended, rotationally flattened disks because such structures appear to be fairly ubiquitous in gravitationally bound, astrophysical systems. (There is strong evidence, for example, that rotationally flattened disks either exist now or have existed in the past around our sun, individual planets within our solar system, numerous protostars, the primary star of many mass--exchanging binary star systems, and active galactic nuclei.) What is peculiar about the \ion{H}{1} disks of many galaxies is that the disks are significantly warped. It is not clear why natural dissipative processes similar to those which work effectively to minimize out-of-the-plane motions in stellar or protostellar ``accretion'' disks are unable to suppress warps in the gaseous disks of galaxies. As Binney (1992) has reviewed, there still is no generally accepted dynamical model of spiral galaxies that satisfactorily explains either the origin or the current structure of warped \ion{H}{1} disks. Almost thirty years ago, Hunter \& Toomre (1969) examined whether normal, infinitesimal bending oscillations might be exhibited by thin, rotating disks of self-gravitating material, permitting them to sustain coherent warps for more than a Hubble time. They concluded that ``for any disk whose density tapers sufficiently gradually to zero near its edge,'' the frequency spectrum of such oscillations is at least partly continuous and, as a result, coherent warps cannot be sustained. Over the subsequent decade, a number of other ideas surfaced to explain the persistence of warps in galaxies, each one taking advantage of the demonstrated existence of dark matter halos around galaxies. As Toomre (1983) has reviewed, however, models proposing to use the halo as an active agent to excite warps in otherwise flat disks -- for example, via the Mathieu instability (Binney 1981) or via a flapping instability (Bertin \& Mark 1980) -- each present significant difficulties. Toomre (1983) and Dekel \& Shlosman (1983) proposed, instead, that untwisted steady-state warps may be sustained as a result of steady forcing by a nonspherical, tilted halo. Building on the early work of Hunter \& Toomre (1969) and the idea that forcing by a nonspherical dark halo can influence the dynamics of the visible disks of galaxies, Sparke (1986) and Sparke \& Casertano (1988) have shown that a discrete warping mode can persist if the disk is sufficiently self-gravitating and if it is embedded in a nonspherical halo whose equatorial plane is tilted with respect to the centralmost regions of the disk. Adopting this model of galaxy warps, the following evolutionary picture emerges. During the galaxy formation process, gas which falls into a spheroidal dark matter halo generally will find that its angular momentum vector is tipped at some nonzero angle, $i$, away from the symmetry axis of the halo. If the gas is cold, it will settle into a rotationally flattened disk that is tilted at the same angle $i$ with respect to the equatorial plane of the halo. In the centralmost regions of the galaxy, where the self-gravity of the gas (or, ultimately, the combined gas/star system) dominates over the gravitational influence of the halo, the gas will be content to remain in orbits that preserve this original tilt. In the outermost regions of the galaxy where the gravitational field of the halo dominates, however, the gas should settle into the halo's equatorial plane. As Toomre (1983) and Dekel \& Shlosman (1983) both sketched in their original concept papers, there also should be an intermediate region where the gas will be influenced significantly by the nonspherical gravitational fields of both the halo and the central gas (or gas/star) disk. Through their modeling efforts, Sparke (1986) and Sparke \& Casertano (1988) confirmed this earlier suspicion that in the intermediate region, the gas can reside in a steady-state, ``warped disk'' structure that provides a smooth radial transition between the separate ``flat disk'' orientations of the inner and outer regions of the galaxy. Concentrating on the dynamics of the centralmost and intermediate regions -- that is, by building models in which there was effectively no gas in the outermost regions -- Sparke \& Casertano (1988) showed that, in steady-state, the warped disk exhibits a straight line of nodes which precesses slowly and coherently in a direction retrograde to the orbital motion of the gas. In a time-dependent simulation, furthermore, Hofner \& Sparke (1994) showed that settling to this steady-state warped disk structure occurs from the inside, out, and is driven not by dissipative processes that are similar to those which are thought to drive settling in most stellar or protostellar accretion disks but, rather, because ``bending waves carry energy associated with transient disturbances out toward the disk edge.'' They also showed that, during an evolution as the bending waves propagate outward through the intermediate region of the disk, a twisted structure can develop and persist until the gas has had sufficient time to settle into the steady-state (constant line of nodes) configuration. Sparke \& Casertano (1988) and Hofner \& Sparke (1994) have demonstrated that, with an appropriate choice of parameters, this model of disk warping matches well the observed properties of several galaxies with warped disks. (See also Kuijken 1991 and Dubinski \& Kuijken 1995.) As Hofner \& Sparke (1994) have pointed out, in galaxies with extended \ion{H}{1} disks ``the outermost gas cannot be expected to form part of a coherent warping mode.'' They did not include normal dissipative forces in their simulations and therefore were unable to comment on how such forces might influence the settling process. In this paper, we examine the disk-settling process from the other extreme, ignoring the self-gravity of the gas but introducing an effective kinematical viscosity into the dynamical equations in order to simulate the effects of dissipative forces. Hence, our effort is complementary to the work of Hofner \& Sparke (1994) and is most relevant to galaxies with extended \ion{H}{1} disks -- although there is one galaxy used for model comparisons (NGC 2841) that is shared by both works. We adopt the view that warps in extended \ion{H}{1} disks which exhibit substantial twists are transient features. Independent of precisely what physical process was responsible for initially placing the gas in an orbit that is inclined to the halo's equatorial plane ({\it e.g.}, gas infall at the time of formation or a recent tidal encounter with another galaxy), the twisted structure can be understood as the result of differential precession in the gaseous disk as it dissipatively settles toward that ``preferred plane.'' In the past, there has been considerable concern (first enunciated by Kahn \& Woltjer [1959], but reiterated in the reviews by both Toomre [1983] and Binney [1991]) that differential precession will destroy any warped disk structure on a time scale that is short compared to a Hubble time and, therefore, that the mechanism we are examining cannot reasonably be used to explain the persistence of such structures. In the outermost regions of \ion{H}{1} disks, however, precession times are relatively long and, as was first pointed out by Tubbs \& Sanders (1979), a warped gas layer can persist for a Hubble time if the dark halo in which the disk is embedded deviates only slightly from spherical symmetry. By modeling carefully the process of disk settling that is driven by normal dissipative forces and comparing the models to the observed kinematical properties of galaxies with extended, warped \ion{H}{1} layers, we hope to be able to more carefully examine the viability of such models. Steiman-Cameron \& Durisen (1988, hereafter SCD88) have developed this idea rather extensively. They have adopted a numerical, cloud-fluid model to simulate the time-dependent evolution of a galaxy disk that intially is tilted out of the equatorial plane of an underlying, spheroidal dark halo. The disk is assumed to be composed of a set of annular mass elements, or ``clouds,'' which act like atoms in a viscous fluid. The SCD88 model has offered some valuable physical insight into the time-dependent settling process that is driven by normal dissipative forces and their dynamically generated model of a twisted galaxy disk has been surprisingly successful at matching the peculiar optical image of one particular galaxy, NGC 4753 (\cite{SCKD92}). Our model is analogous to the one developed by SCD88 but it derives from an analytical prescription of the viscous settling process. More specifically, we employ the ``twisted-disk'' equation formalism first developed by Bardeen \& Petterson (1975) and Petterson (1977, 1978) to describe the time-dependent settling of a thin, viscous disk in a nonspherical dark halo potential. This is a rather natural formalism to adopt because, as numerous kinematical ``tilted-ring'' models have demonstrated, warped \ion{H}{1} galaxy disks display a structure that resembles, at least qualitatively, the twisted geometry that had once been thought to be important in accretion disks which surround certain compact stellar objects (Bardeen \& Petterson 1975; see a recent rejuvenation of this idea put forward by Maloney \& Begelman 1997). In adapting the model to galaxy disks, we have replaced the approximate Keplerian gravitational potential used in earlier accretion disk work with a logarithmic potential appropriate to galaxy halos. (Pringle [1992] also recently described how the twisted--disk formalism may be adapted to galaxies.) In the limit of stress-free precession, our model reproduces the analytical prescription of disk settling first presented by SCD88, but our model is not constrained to this limit. A more general solution to the governing equations predicts an exponential settling rate that depends on time to the first power, rather than on time to the third power as has been derived in the limit of stress-free precession. Furthermore, an analytical, steady-state solution to the governing equations produces a twisted-disk structure that is very similar to previously constructed, kinematical models. We demonstrate that projected surface density maps and radial velocity maps derived from our analytical model match published \ion{H}{1} maps of five well-studied warped disk galaxies (M83, NGC 300, NGC 2841, NGC 5033, and NGC 5055) very well. | Utilizing the formalism originally introduced by Petterson to describe warped and twisted accretion disk structures in Keplerian potentials, we have derived a single, complex ODE to describe time dependent settling of an \ion{H}{1} disk in the logarithmic potential that appears to be typical of normal spiral galaxies. Over the interval $1 \lesssim x \lesssim 20$, the analytical function $w_{ss}\bigl(x\bigr)$ -- derived from an analysis of the steady-state limit of the general twisted-disk equation -- appears to describe quite accurately the general warped and twisted structure that is exhibited by a number of galaxy disks. It should be noted again that our analysis is based on the assumption that the warping angle $\beta$ is $\ll 1$. For galaxies with larger warps (including some that we have modeled) this is a simplifying assumption and should be disregarded in more proper treatments (see Pringle [1992]). From our model fits we conclude that, quite generally, the effective kinematical viscosity in these neutral hydrogen disks is $\nu\sim0.6$ km s$^{-1}$ kpc. That is, the effective Reynolds number in these systems is \begin{displaymath} R_e\sim\frac{r_{max} v_{\psi}}{\nu} \sim \frac{x_{max}^2 v_\psi}{r_{max}} \sim6000\;. \end{displaymath} According to equation (\ref{radvsub}), this also implies that the ratio of the radial inflow velocity of the gas to its orbital velocity is \begin{displaymath} \frac{v_r}{v_{\psi}}\approx\frac{1}{R_e}\sim\times{10}^{-4}\;. \end{displaymath} This model also provides a mechanism by which the parameter $\eta$ -- the quadrupole moment of the underlying dark halo potential well -- can be measured in spiral galaxies. Our fits to five normal spirals with well-studied warped \ion{H}{1} disks specifically indicate that $\eta\sim{10}^{-3}$. (This value can be increased to $\eta\sim{10}^{-2}$, with a corresponding factor of ten increase in viscosity, only if the age of these disks is assumed to be 0.1$\,{H_0}^{-1}$ -- which seems unlikely -- or $\sigma$ in expression 38c is found to be $\sim 0.1$.) Hence, we conclude that the dark halos in which these warped disks sit are, to quite high accuracy, spherically symmetric. This particular conclusion should not come as a surprise because some time ago Tubbs and Sanders (1979) pointed out that if warped disks are identified as transient structures, the warp can only be sustained for a Hubble time if the underlying halo potential is very nearly spherical. Although we have examined in detail here only the steady-state solution, the derived time-dependent twisted-disk equation provides a tool that can be utilized to model the time-evolution of warped \ion{H}{1} disks without resorting to elaborate numerical techniques. The twisted-disk formalism in general -- and the analytical function $w_{ss}\bigl(x\bigr)$ in particular -- provides an avenue through which models of warped \ion{H}{1} disks can advance from purely kinematical fits (e.g. tilted-ring models) to dynamical models based on a reasonable physical model. In the future, we also expect to use the time-dependent form of the twisted-disk equations as a point of comparison for fully three-dimensional gas dynamic simulations of \ion{H}{1} disks settling into distorted halo potentials. | 98 | 3 | astro-ph9803270_arXiv.txt |
9803 | astro-ph9803028_arXiv.txt | We present a total of 48~minutes of observations of the nearby, bright millisecond pulsar \psr~taken at the Parkes radio observatory in Australia. The data were obtained at a central radio frequency of 1380~MHz using a high-speed tape recorder that permitted coherent Nyquist sampling of 50 MHz of bandwidth in each of two polarizations. Using the high time resolution available from this voltage recording technique, we have studied a variety of single-pulse properties, most for the first time in a millisecond pulsar. We show that individual pulses are broadband, have pulse widths ranging from $\sim$10~$\mu$s ($\sim 0.6^{\circ}$ in pulse longitude) to $\sim$300~$\mu$s ($\sim 20^{\circ}$) with a mean pulse width of $\sim$~65$\mu$s ($\sim 4^{\circ}$), exhibit a wide variety of morphologies, and can be highly linearly polarized. Single pulse peaks can be as high as 205~Jy (over $\sim$40 times the average pulse peak), and have a probability distribution similar to those of slow-rotating pulsars. We observed no single pulse energy exceeding $\sim$4.4 times the average pulse energy, ruling out ``giant pulses'' as have been seen for the Crab and \ntts\ pulsars. \psr\ does not exhibit classical microstructure or show any signs of a preferred time scale that could be associated with primary emitters; single pulse modulation has been observed to be consistent with amplitude-modulated noise down to time scales of 80 ns. We observe a significant inverse correlation between pulse peak and width. Thus, the average pulse profile produced by selecting for large pulse peaks is narrower than the standard average profile. We find no evidence for ``diffractive'' quantization effects in the individual pulse arrival times or amplitudes as have been reported for this pulsar at lower radio frequency using coarser time resolution (\cite{amdv97}). Overall, we find that the single pulse properties of \psr\ are similar to those of the common slow-rotating pulsars, even though this pulsar's magnetosphere and surface magnetic field are several orders of magnitude smaller than those of the general population. The pulsar radio emission mechanism must therefore be insensitive to these fundamental neutron star properties. | \label{sec:intro} Single pulse studies of millisecond pulsars are considerably more difficult to perform than are those of slow pulsars. Faster data rates are needed to study millisecond pulsars with comparable pulse phase resolution, and finer radio frequency resolution is required to minimize the effect of interstellar dispersion. Also, interstellar scattering time scales comparable to the pulse duration render studying individual pulse morphologies impossible, while steep millisecond pulsar spectra preclude observations of sufficient sensitivity at higher radio frequencies where scattering is less important. Yet single pulse studies of millisecond pulsars are highly desirable for two reasons. First, the origin of the radio emission that has made isolated neutron stars famous is, even 30 years after their discovery, still a mystery. The high brightness temperatures ($\sim 10^{30}$K) associated with the radio emission point to coherent processes which are poorly understood even under less exotic conditions (\cite{mel96}). Previous observations of slow pulsars have not constrained the emission mechanism sufficiently; the study of radio emission properties of millisecond pulsars may provide important new clues. Millisecond pulsars, because of their fast spin periods, have much smaller light-cylinder radii, and hence magnetospheres, than slow pulsars. They also have lower surface magnetic field strengths than the general pulsar population (most likely resulting from their having been ``recycled'' by a binary companion through the accretion of mass and angular momentum). Were the radio emission mechanism at all dependent on such properties, millisecond pulsars should have different radio properties than the slower-spinning general population. The second reason single pulse studies of millisecond pulsars are important is that millisecond pulsar timing is well-known to be an unparalleled source of precision astrometric and astrophysical information. Among factors possibly limiting timing precision is the stability of the average profile, which depends on the properties of single pulses. Only recently has a systematic study of single pulses from millisecond pulsars become possible, largely due to improving computational and data recording technologies. To date, the only such investigation has been for the 1.5~ms pulsar \ntts\ (\cite{wcs84,sb95,bac95,cstt96}). Interestingly, \ntts\ exhibits giant radio pulses like those seen elsewhere only in the Crab pulsar. With the single pulse properties of only one millisecond pulsar having been studied in any detail, the question of whether all millisecond pulsars show similar properties naturally arises. Unfortunately, \ntts\ suffers interstellar scattering at time scales comparable to the duration of a single pulse at the radio frequencies at which it has been observed, rendering detailed study of individual pulse morphologies difficult. Here we report on high-time-resolution single pulse studies of a second millisecond pulsar, \psr. This pulsar's large flux density and low dispersion measure (DM), and the corresponding scarcity of line-of-sight scattering material, render it an obvious target for single pulse work. Some single pulse investigations of \psr\ have been reported (\cite{jlh+93,amdv97}) but none have had sufficient time resolution to resolve most individual pulses. Using a fast recording device and powerful supercomputers, we have been able to resolve all pulses in our data, the narrowest being $\sim$10~$\mu$s. | \label{sec:concl} We have presented the first detailed single pulse study of a millisecond pulsar in which sufficient time resolution was available to resolve single pulses as they would be seen in the pulsar vicinity. The similarity of the single pulse properties to those of normal slow pulsars is remarkable given the dramatically reduced magnetospheric volume and magnetic field strength of \psr. Indeed, without being told the absolute sample rate, it is unlikely that one could distinguish between this being a millisecond or slow pulsar. To summarize, our observations of \psr\ have: resolved individual pulses, and shown that they have a wide variety of morphologies, with multiple sub-pulse components not uncommon; shown that individual pulses are in general broadband; shown that individual pulses can have high linear polarization; provided no evidence for giant pulses as observed in the Crab pulsar and PSR B1937+21, nor for pulse nulling or drifting subpulse phenomena; found structure in the intensity fluctuation spectrum; revealed a correlation of pulse peak with pulse width so that the average profile formed from only the highest amplitude pulses is much narrower than the conventional average profile; not shown any evidence for micro-structure or preferred time scales $\geq$80~ns; shown that the emission is consistent with an amplitude modulated noise model; provided no evidence to support the claim made by Ables \etal\ (1997) of the detection of coherent radiation patterns. Because there exists no self-consistent radio emission mechanism context (see e.g. \cite{mel96}) in which to discuss these results, it is difficult to say how such models are constrained. Indeed previous slow pulsar single pulse studies suffered from the same difficulty. However, it is often the case that fundamental insights become apparent when known phenomenon are taken to extremes; this, and improved tape recording and computer technologies that permit single pulse studies of millisecond pulsars, provided our motivation in undertaking this analysis. Similar studies of other millisecond pulsars may eventually lead to an understanding of the radio emission mechanism. | 98 | 3 | astro-ph9803028_arXiv.txt |
9803 | astro-ph9803264_arXiv.txt | We present multi-frequency radio continuum VLBI observations of the gravitational lens system B0218+357 carried out using a global VLBI network and the VLBA. The source has been observed with resolutions from 0.2~mas to 5~mas and displays interesting structure. The spectral properties of various components show that the lensed object is a standard flat spectrum radio source which has many self-absorbed components. Based on the flux ratio of the lensed images as a function of frequency we propose a simple model for the background radio source. | The radio source B0218+357 has been identified as a gravitationally lensed system (Patnaik et al. 1993). The source, which has a core-halo structure at low frequency (1.4~GHz) and low resolution (5~arcsec), consists of two compact flat-spectrum components, A and B, separated by 335 milliarcsec. The weaker of the two, B, is surrounded by a faint Einstein ring of similar diameter. The spectral and polarisation characteristics of the components show them to be the lensed images of a single flat spectrum `core'. The lensing galaxy, suggested to be a spiral galaxy, has been observed by the HST (Jackson et al. 1997). It has a redshift of 0.6847 (Browne et al. 1993). Atomic hydrogen and many molecular species have been detected in absorption against the background source (Carilli et al. 1993, Wiklind \& Combes 1995, Menten \& Reid 1996). The redshift of the background object is suggested to be 0.96 (Lawrence 1996). In this contribution we explore the properties of the radio source from our multi-frequency VLBI observations. We give a summary of our observations and discuss the results in the context of a flat spectrum radio source. We note that the overall radio spectrum of B0218+357 is flat between 365~MHz and 43~GHz i.e. its flux density is constant within a factor of two between these frequencies. | We derive two basic results from these observations, namely the spectral decomposition of various components and the ratio of their flux densities as a function of frequency. Since we resolve the `core-jet' structure in A and B at frequencies higher than 8.4~GHz, we have 4 measurements of flux density for these components. The spectra are plotted in Fig. 2. The spectrum of the core (A1 and B1) is self-absorbed with a peak around 15~GHz. The jet (A2 and B2) has steep spectra above 8.4~GHz. However, the total flux density of A and B at 1.7 and 5~GHz suggest that both A2 and B2 must also be self-absorbed between these two frequencies. The Einstein ring emission has a steeper spectrum between 5 and 22~GHz and it must also be self-absorbed at lower frequencies since the total flux density provides a limit (Patnaik et al. 1993). In summary, the radio structure of B0218+357 consists of at least 3 distinct components, core, jet and the Einstein ring, which are all self-absorbed at different frequencies. The components are self-absorbed at progressively lower frequencies as one moves from the core along the jet. This behaviour of the radio source is consistent with its flat spectrum (Cotton et al. 1980). The core, being self-absorbed around 15~GHz, contributes very little to the emission at 1.7 and 5~GHz, where the emission is dominated by the jet as evident from the spectra. Perhaps it is not surprising therefore that the radio structures of both A and B at 1.7~GHz appear diffuse. However, this poses a difficulty in that a self-absorbed component, which must be compact, has a large observed size. In this case one must consider the intrinsic size of the source rather than the observed size which has been magnified by the lens. Another result from our observations is that the image flux ratios change with frequency. This is an apparent contradiction since gravitational lensing is achromatic. However, since the source is extended and spans areas of different image magnifications the observed flux ratio can indeed vary with frequency. The flux ratio of the core (i.e. A1/B1) is around 3.7 for frequencies higher than 8.4~GHz. This is similar to the ratio of A/B obtained from VLA observations at these frequencies (Patnaik et al. 1993). At lower frequencies the flux ratio A/B is around 2.6. Of course, we point out that the source is variable at high frequencies and thus the measured flux ratio can be different due to differential time delay. The above results lead to a rather simple model for B0218+357. The radio source consists of a number of different components which are self-absorbed at different frequencies thus conspiring to produce the observed flat spectrum. The core (imaged into A1 and B1) is located at the western-most edge, the jet (imaged into A2 and B2) lies between the core and the component giving rise to the Einstein ring. The core dominates the radio structure at frequencies higher than 8.4~GHz, and the jet and the Einstein ring dominate at lower frequencies. At lower frequencies one does not detect any compact feature in the source since the core contributes very little to the flux density. | 98 | 3 | astro-ph9803264_arXiv.txt |
9803 | astro-ph9803322_arXiv.txt | We have determined detailed radio spectra for 26 compact sources in the starburst nucleus of M82, between $\lambda\lambda$ 74 and 1.3 cm. Seventeen show low-frequency turnovers. One other has a thermal emission spectrum, and we identify it as an \HII region. The low frequency turnovers are due to absorption by interstellar thermal gas in M82. New information on the AGN candidate, 44.01+595, shows it to have a non-thermal falling powerlaw spectrum at the highest frequencies, and that it is strongly absorbed below 2 GHz. We derive large magnetic fields in the supernova remnants, of order 1-2 $(1+k)^{2/7} \phi^{-2/7}$ milliGauss, hence large pressures in the sources suggest that the brightest ones are either expanding or are strongly confined by a dense interstellar medium. From the largest source in our sample, we derive a supernova rate of 0.016 yr$^{-1}$. | Supernovae (SNe) and supernova remnants (SNR's) are thought to be the main ``drivers'' of the starburst phenomenon (\cite{kw75}). Detailed radio observations for supernovae and supernova remnants are few, due to the fact that only a small subset of the extragalactic SNe discovered each year produce detectable radio emission. Those events for which monitoring data is available (e.g. \cite{weiler88}) have revealed considerable information regarding the evolution of the properties of the expanding shock wave and its emission processes. In intense starbursts like M82, where a significant population of massive stars has had sufficient time to evolve to the SN stage, we have the unique opportunity to study a collection of SNR's of similar age and origin, and their interaction with the surrounding environment. The observations are briefly outlined in Section 2. We discuss the technique used to fit spectral models in Section 3. Then in Section 4 we discuss the physical implications of our modelling. Throughout the paper, we assume that the distance to M82 is 3.63 Mpc (\cite{freedman}, \cite{tammann}), so that 1$\farcs$0 corresponds to 17.6 pc. | 1. Of the 26 sources, one (42.21+590) is an \HII region, 22 are most likely young radio supernovae or supernova remnants, and three further sources remain of uncertain nature. 2. High magnetic field strengths and pressures are derived for the sources we identify as SNR's. The brighter ones do not appear to be in pressure equilibrium with the surrounding medium, and are probably expanding, or are strongly confined by the high pressure ISM. The source of the high nuclear pressure and halo winds are most likely the SNR's. 3. Sources that show a low-frequency turnover, do so as a result of absorption by ionized gas. This ionized gas may be intimately associated with the source, or may be in clouds located along the line of sight to the sources. The gas is most likely in the form of clumpy, high density clouds. 4. The resolved ring-like source in our survey is the oldest. We derive a supernova rate in M82 of $\sim$ 0.016 $\left( \frac{V}{5000 \;\;km/s} \right)$yr$^{-1}$. This value is a lower limit. | 98 | 3 | astro-ph9803322_arXiv.txt |
9803 | astro-ph9803114_arXiv.txt | As the Cosmic Microwave Background (CMB) radiation is observed to higher and higher angular resolution the size of the resulting datasets becomes a serious constraint on their analysis. In particular current algorithms to determine the location of, and curvature at, the peak of the power spectrum likelihood function from a general $N_{p}$-pixel CMB sky map scale as $O(N_{p}^{3})$. Moreover the current best algorithm --- the quadratic estimator --- is a Newton-Raphson iterative scheme and so requires a `sufficiently good' starting point to guarantee convergence to the true maximum. Here we present an algorithm to calculate bounds on the likelihood function at any point in parameter space using Gaussian quadrature and show that, judiciously applied, it scales as only $O(N_{p}^{7/3})$. Although it provides no direct curvature information we show how this approach is well-suited both to estimating cosmological parameters directly and to providing a coarse map of the power spectrum likelihood function from which to select the starting point for more refined techniques. | Planned observations of the Cosmic Microwave Background (CMB) will have sufficient angular resolution to probe the CMB power spectrum up to multipoles $l \sim 1000$ or more (for a general review of forthcoming observations see \cite{S}). If we are able to extract the multipole amplitudes $C_{l}$ from the data sufficiently accurately we will be able to obtain the values of the fundamental cosmological parameters to unprecedented accuracy. The CMB will then have lived up to its promise of being the most powerful discriminant between cosmological models \cite{Kn,HSS,KKJS}. Extracting the power spectrum is conceptually simple --- the raw data is cleaned and converted into a time-ordered dataset. This is then converted to a sky temperature map, which in turn is analysed to find the location of, and curvature at, the maximum of the likelihood function of the power spectrum. In practice as the size of the dataset increases the problem rapidly becomes intractable by conventional methods. This is particularly true of the final step --- the likelihood analysis of any reasonably general sky temperature map. An observation of the CMB contains both signal and noise \begin{equation} \Delta_{i} = s_{i} + n_{i} \end{equation} at each of $N_{p}$ pixels. For independent, zero-mean, signal and noise the covariance matrix of the data \begin{equation} M \equiv \left< {\bf \Delta} \, {\bf \Delta}^{T} \right> = \left< {\bf s} \, {\bf s}^{T} \right> + \left< {\bf n} \, {\bf n}^{T} \right> \equiv S + N \end{equation} is symmetric, positive definite and dense. For any theoretical power spectrum $C_{l}$ we can construct the corresponding signal covariance matrix $S(C_{l})$; knowing the noise covariance matrix $N$ for the experiment we now know the observation covariance matrix for that power spectrum $M(C_{l})$. The probability of the observation given the assumed power spectrum is then \begin{equation} \label{eq.lf} P({\bf \Delta} \, | \, C_{l}) = \frac{e^{-{\small\frac{1}{2}} \, {\bf \Delta}^{T} M^{-1} {\bf \Delta}}} {(2 \pi)^{N_{p}/2} \left| M \right|^{1/2}} \end{equation} Assuming a uniform prior, so that \begin{equation} P(C_{l} \, | \, {\bf \Delta}) \propto P({\bf \Delta} \, | \, C_{l}) \end{equation} we can restrict our attention to evaluating the right hand side of equation (\ref{eq.lf}). Unfortunately unless the noise covariance matrix is unrealistically simple (eg. diagonal) both direct evaluation \cite{G1,G2} and quadratic estimation \cite{T,BJK} of the likelihood function scale as at best $O(N_{p}^{3})$ \cite{B}, making them impractical for the forthcoming $10^4$ -- $10^6$ pixel datasets. | We have presented an algorithm to calculate probabilistic bounds on the power spectrum likelihood function from an $N_{p}$-pixel CMB map using Gaussian quadrature which scales as between $O(N_{p}^{7/3})$ and $O(N_{p}^{5/2})$ --- a very significant advance on existing algorithms for the exact calculation which scale as $O(N_{p}^{3})$. Since lowering the convergence constraint below the 1\% level gains us only marginally tighter final bounds at the expense of increasing the scaling power, it is not recommended for the forthcoming $10^{4}$ -- $10^{6}$ pixel CMB maps. Our final algorithm of choice therefore gives better than 3\% bounds on the logarithm of the likelihood function with $O(N_{p}^{7/3})$ operations with 99\% confidence. Since this algorithm gives no information about the local curvature of the likelihood function it is not as well suited as quadratic estimator techniques for searching a large multi-dimensional parameter space for its likelihood maximum. However for direct estimation of a small set of cosmological parameters this technique is certainly viable and fast. Moreover, even when the parameters are taken to be the multipole moments (individually or in bins), quadratic estimation, being a Newton-Raphson iteration, requires a starting point `sufficiently close' to the maximum to guarantee convergence; the algorithm presented here is then well suited to provide a coarse overall mapping of the likelihood function from which to select a starting point for more refined techniques. | 98 | 3 | astro-ph9803114_arXiv.txt |
9803 | astro-ph9803232_arXiv.txt | We study the influence of a possible H dibaryon condensate on the equation of state and the overall properties of neutron stars whose population otherwise contains nucleons and hyperons. In particular, we are interested in the question of whether neutron stars and their masses can be used to say anything about the existence and properties of the H dibaryon. We find that the equation of state is softened by the appearance of a dibaryon condensate and can result in a mass plateau for neutron stars. If the limiting neutron star mass is about that of the Hulse-Taylor pulsar a condensate of H dibaryons of vacuum mass $\sim 2.2$ GeV and a moderately attractive potential in the medium could not be ruled out. On the other hand, if the medium potential were even moderately repulsive, the H, would not likely exist in neutron stars. If neutron stars of mass $\sim 1.6 M_\odot$ were known to exist, attractive medium effects for the H could be ruled out. | Since Jaffe proposed that there may exist a stable dihyperon (a quark composite with baryon number two) \cite{Jaffe77}, an ongoing quest for this particle began \cite{Carroll78}. Recent searches using kaon beams \cite{Aoki90} or heavy ion beams \cite{Belz96,Belz97,Stotz97} found no candidates or are still in progress \cite{Craw98}. There exist some claims for evidence for the H dibaryon produced in proton-nucleus \cite{Shahba93} and in heavy-ion collisions \cite{Long95}. Nevertheless, these candidates might be misidentified $K^0_L$ as seen in \cite{Belz97}. For a most recent overview on the search for the H dibaryon we refer to \cite{HYP97}. There are numerous mass estimates for the H dibaryon and they are reviewed in \cite{Dover89}. The existence or nonexistence of the H dibaryon is strongly connected with the observation of double $\Lambda$ hypernuclei which has been discussed in \cite{Dalitz89}. Three double $\Lambda$ hypernuclei have been reported in literature: $^{~~6}_{\Lambda\Lambda}$He \cite{Danysz63}, $^{~10}_{\Lambda\Lambda}$Be \cite{Prowse66}, and $^{~13}_{\Lambda\Lambda}$B \cite{Aoki91,Dover91}. The two $\Lambda$'s can decay by strong interactions to the H dibaryon. As this has not been seen in the above events, the H must either be heavier than $m_H> 2 m_\Lambda+B_{\Lambda\Lambda}\approx 2.22$ GeV \cite{Kerb84} or the events are misidentified as an H hypernucleus with a shallow attractive nuclear potential \cite{Dover89}. A more stringent condition is the observation of the weak mesonic decay of the double $\Lambda$ hypernuclei giving $m_H> m_\Lambda + m_p + m_{\pi^-} + B_\Lambda \approx 2190$ MeV \cite{Jaffe91} where $B_\Lambda$ depends on the mass of the decay fragment and is $B_\Lambda=-3.1$ MeV for $^5_\Lambda$He. In all cases, a deeply bound H dibaryon seems to be ruled out by these events. If the H dibaryon exists, it will have a certain impact also on the properties of dense matter. It is quite established nowadays, that neutron stars have a large hyperon fraction in the core and might be described as giant hypernuclei, though bound by gravity \cite{Glen85}. Here again, the presence of hyperons might restrict certain properties of the H dibaryon. Recently, studies for neutron stars have been done for nuclear matter without hyperons but including H dibaryon condensation \cite{Fae97a} and limits have been set for the coupling constants of the H dibaryon \cite{Fae97b}. There might exist heavier partners of the H dibaryon, lumps of strange quark matter dubbed strangelets. There are several heavy-ion experiments dedicated to search for this novel form of matter \cite{Arm97,Beavies95,Apple96}. In the MIT bag model, strangelets with $A\leq 6$ are found to be unbound \cite{Aerts78}. Nevertheless, light strangelet candidates in the range of $6<A<40$ might be stable against weak hadronic decay \cite{Gilson93,Scha97} (an overview of the properties of strange matter can be found in \cite{GS96}). The H dibaryon as well as these light strangelets can occur in dense matter as a precursor of the phase transition to a quark plasma. In this paper, we study the influence of H dibaryons and other strangelet candidates on the composition and structure of neutron stars including the hyperon degree of freedom. We are particularly interested in the question of whether neutron stars and their masses can be used to say anything about the existence and properties of the H dibaryon. In section \ref{sec:comp}, we discuss the condition for the occurrence of dibaryons and strangelets in neutron star matter. The relativistic mean field model with hyperons and the H dibaryon is presented in section \ref{sec:model}. Implications for a H dibaryon condensate are discussed in section \ref{sec:results} and summarized in the last section. | \label{sec:summary} We are particularly interested in the question of whether neutron stars and their masses can be used to say anything about the existence and properties of the H dibaryon. We have studied the influence of the possible occurrence of an H dibaryon condensate and strangelets in neutron stars including hyperons. Without in-medium modifications, it is quite likely that especially negatively charged strangelets, if they exists, will be present in the dense interior of neutron stars. The appearance of H dibaryons in the stellar core depends crucially on their mass and on the chosen potential of the H in nuclear matter. Hyperons tend to shift the onset of the H to higher density or to prevent H dibaryon condensation. If the condensation happens and if the potential of the H is attractive enough to provide a substantial number density in the neutron star, the maximum mass of the neutron star is reduced compared to the case without the H dibaryon. The decrease of the maximum mass is moderate and allows for the presence of H dibaryons in the interior of neutron stars in accord with present neutron star mass data. If the H dibaryon feels an attractive potential in matter, it can lead to a plateau in the mass of neutron stars, as there exist a region of very slowly rising pressure with energy density. If the limiting neutron star mass is about that of the Hulse-Taylor pulsar a condensate of H dibaryons of vacuum mass $\sim 2.2$ GeV and a moderately attractive potential in the medium could not be ruled out. On the other hand, if the medium potential were even moderately repulsive, the H, would not likely exist in neutron stars. If neutron stars of mass $\sim 1.6 M_\odot$ were known to exist, attractive medium effects could be ruled out. For a mass limit of $1.44 M_\odot$, attractive potentials for an H mass below the $\Sigma$N threshold (1.3 GeV) are ruled out. H dibaryon or strangelet condensation might happen as a precursor to the phase transition to a quark plasma. In this respect, we note that this phase transition is of first order \cite{Glen92}. Hence, small bubbles of strange quark matter will appear in the mixed phase which are most likely negatively charged due to the isospin potential of the nuclear matter. This is in line with the results presented here. As the most stable strangelets have spin zero, the onset to a quark plasma will be initiated by a Bose condensation of strangelets (possibly including the H dibaryon). As the phase transition proceeds, the bubbles will overlap and will finally replace nuclear matter by essentially filling up the whole volume. ~~\\[2ex] Acknowledgments: J.S.B. acknowledges support by the Alexander-von-Humboldt Stiftung with a Feodor-Lynen fellowship. This work is supported by the Director, Office of Energy Research, Office of High Energy and Nuclear Physics, Nuclear Physics Division of the U.S. Department of Energy under Contract No.\ DE-AC03-76SF00098. | 98 | 3 | astro-ph9803232_arXiv.txt |
9803 | astro-ph9803142_arXiv.txt | Two fundamental empirical laws have been established in the analysis of galaxy space distribution. First, recent analyses have revealed that the three dimensional distribution of galaxies and clusters is characterized by large scale structures and huge voids: such a distribution shows fractal correlations up to the limits of the available samples. This has confirmed the earlier de Vaucouleurs power-law density - distance relation, now corresponding to a fractal structure with dimension $D \approx 2$, at least, in the range of scales $ \sim 1 \div 200 \; Mpc$ ($H_0 = 55 km/sec/Mpc$). An eventual cut-off towards homogenization has not been yet identified. Second, since Hubble's discovery, the linear redshift-distance law has been well established within $200 Mpc$ and also much deeper. The co-existence of these laws within the same scales is a challenge for the standard cosmology where the linear Hubble law is a strict consequence of homogeneity of the expanding universe. This puzzle is now sufficiently strong to raise doubts for the standard cosmology. | 98 | 3 | astro-ph9803142_arXiv.txt |
||
9803 | astro-ph9803138_arXiv.txt | We present new {$v\sin i$}~ measurements for 235 low-mass stars in the Pleiades. The differential rotational broadening has been resolved for all the stars in our sample. These results, combined with previously published measurements, provide a complete and unbiased rotation data set for stars in the mass range from 0.6 to 1.2{$M_{\odot}$}. Applying a numerical inversion technique on the {$v\sin i$}~ distributions, we derive the distributions of {\it equatorial\/} velocities for low-mass Pleiades members. We find that half of the Pleiades dwarfs with a mass between 0.6 to 1\,{$M_{\odot}$}~ have rotation rates lower than 10{\,km\,s$^{-1}$}. \\ Comparison of the rotational distributions of low-mass members between IC 2602/2391 ($\approx 35$\,Myr) and the Pleiades ($\approx 100$\,Myr) suggests that G dwarfs behave like solid-bodies and follow Skumanich's law during this time span. However, comparison between Pleiades and older clusters --M34 ($\approx 200$\,Myr) and Hyades ($\approx 600$\,Myr)-- indicates that the braking of slow rotators on the early main sequence is weaker than predicted by an asymptotical Skumanich's law. This strongly supports the view that angular momentum tapped in the radiative core of slow rotators on the zero age main sequence (ZAMS) resurfaces into the convective envelope between Pleiades and Hyades age. For the G-dwarfs, we derive a characteristic coupling time scale between the core and the envelope of about 100--200\,Myr, which accounts for the observed evolution of surface rotation from the ZAMS to the Hyades.\\ The relationship between rotation and coronal activity in the Pleiades is in agreement with previous observations in other clusters and field stars. We show that the Rossby diagram provides an excellent description of the X--ray activity for all stars in the mass domain studied. The Pleiades data for slow and moderate rotators fills the gap between the X-ray--rotation correlation found for slow rotators and the X-ray ``saturation plateau'' observed for young fast rotators. The transition between increasing X-ray flux with rotation and X-ray saturation is observed at $\log(P/\tau)=0.8\pm0.1$. These results strengthen the hypothesis that the ``saturation'' of the angular momentum loss process depends on the stellar mass. | In the past 10 years, numerous rotational velocity measurements have been obtained for low-mass pre-main sequence stars and members of nearby young open clusters. This set of observations has shown that the angular momentum of each star follows an evolutionary scenario from the star's birth to the age of the Sun. Though several models have been developed to describe the rotational history of low-mass stars, the data available to confront models with observations were still insufficient to provide an unambiguous validation of these models. The measurements of stellar rotation for various ages and the understanding of its evolution shall provide a way to get a better knowledge of the physical processes experienced by the star through its history. In particular, this data might offer additional insight into the physics of young stars and their interactions with their proto-stellar and maybe their proto-planetary disks. On the ZAMS, G type stars ($0.8-1.0${$M_{\odot}$}) exhibit a large spread of rotational velocities. A difference of the order of 150{\,km\,s$^{-1}$}~ is observed for G dwarfs in Alpha\,Per ($\approx 50$\,Myr) between the fastest and slowest rotators (Prosser 1992). At Hyades age (about 600\,Myr), the spread has disappeared and the stars with masses in the 0.6 to 1 {$M_{\odot}$}~ range show a monotically decreasing rotational velocity with mass (Radick et al. 1987; Stauffer et al. 1997a). Over this time span, the fast rotators have been strongly braked but the slow rotators have only suffered a little braking. Moreover, the efficiency of the braking process also depends on the stellar mass. At the Pleiades age ($\approx100$\,Myr), almost all G type stars have converged down to slow rotation while K dwarfs stars still exhibit a large roational spread (Mayor \& Mermilliod 1991; Soderblom et al. 1993). At the Hyades age, a significant spread in rotational velocity spread is only observed for stars less massive than 0.6{$M_{\odot}$} (Stauffer et al. 1997a). Models have difficulties to simultaneously reproduce the large diversity of rotators at early ages and their strong convergence in a few 100\,Myr years. The recent discovery of a bi-modal velocity distribution for pre-main sequence stars in Taurus and Orion (see Bouvier 1994; Choi \& Herbst 1996) suggests that the magnetic coupling between the disk and the star (Camenzind 1990; K\"onigl 1991) could prevent the young star from spinning up during its PMS contraction, yielding a large spread in the initial angular momentum distribution. On the main sequence, the star's rotation is mostly ruled by the angular momentum losses at the surface through a magnetized wind (Schatzmann 1962; Weber \& Davis 1967) and by the angular momentum transfer in the inner parts of the star (Endal \& Sofia 1978; Mac\,Gregor \& Brenner 1991). Depending on the efficiency of the transfer and the disk lifetime, various scenarios can be predicted. The latest models of angular momentum evolution are from Krishnamurthi et al. (1997a), Bouvier et al. (1997b) and Allain (1998). Numerous data on stellar rotation were available to these models, though not enough yet to derive the complete distributions of rotational velocity in young open clusters. The comparison between such distributions and the initial velocity distribution gathered from T\,Tauri observations would provide an important additional constraint on the models. Unbiased rotational velocity distributions are hard to build, numerous observations being needed to get a statistically significant distribution. In this case, measurements of the projected velocity ({$v\sin i$}~) are a good alternative to the direct measurement of rotation periods which requires a huge number of photometric measurements, difficult to achieve for a large sample of stars. Indeed, since the pioneering work of van Leeuwen \& Alphenaar (1982), there have been several attempts to measure photometric periods in the Pleiades (see O'Dell et al. 1995 for a census of the observations; Krishnamurthi et al. 1997b). Up to now, 42 periods have been measured in this cluster, for stars with {(B$-$V)$_{0}$} in the range 0.5 to 1.4. The detection of rotational periods for the whole sample of Pleiades stars in this mass range (more than 200 catalogued members) is difficult and will still take a few years. Measurements of {$v\sin i$}~ are more straighforward but are obviously affected by projection effects. Yet, the knowledge of the {$v\sin i$}~ for a large stellar sample does provide rich {\sl statistical} information. It is then possible to extract an accurate estimate of the distribution of {\it equatorial} velocities in a cluster from a large number of observed {\sl projected} rotational velocities if the stellar sample is statistically representative of the stellar population of the cluster and if it does not suffer from incompleteness (i.e., all the {$v\sin i$}~ must be resolved). We consider that the inclination angles are randomly distributed in clusters. In the next section we present new {$v\sin i$}~ measurements for 235 Pleiades low-mass members. We show that by taking into account the spectral type of stars, we can calibrate the intrinsic width of the spectral lines and measure the {$v\sin i$}~ of very slow rotators down to 1.5{\,km\,s$^{-1}$}. The calibration process is described in Sect.~2. We detect rotational broadening for all observed stars. More than 98\% Pleiades members of the original Hertzprung sample now have a resolved rotational velocity measurement up to {(B$-$V)$_{0}$}=1.35. The distributions of projected rotational velocities are then converted into distributions of equatorial velocities using a numerical algorithm described in Sect.~3. The velocity distributions for various mass ranges are presented in Sect.~4. In Sect.~5 we compare our results with those obtained for other open clusters and discuss the implications for the models of angular momentum evolution. We also examine the relationship between X-ray flux and stellar rotation for Pleiades stars. | From a complete and unbiased set of {$v\sin i$}~ measurements, we have computed the distribution of equatorial velocities in the Pleiades for various mass ranges between 0.5 and 1.5{$M_{\odot}$}. Comparison with the distribution of rotational velocities in the Hyades, M\,34, IC\,2391 and IC\,2602 yields a coherent picture for the angular momentum evolution of the convective envelope. The comparison with the younger clusters IC\,2391 and IC\,2602 suggests that most Pleiades G--dwarfs are in an unsaturated braking law regime. The rotational evolution of moderate rotators in early stages is in agreement with a solid body rotation driven by Skumanich's relationship. The relationship between {$v\sin i$}~ and X-ray emission indicates that the transition between saturated X-ray emission and its steady decrease with rotation occurs at about $P=2$\,d or {$v\sin i$}$=25${\,km\,s$^{-1}$} for solar-type stars. About 10\% of Pleiades G dwarfs lie in the saturation domain. The comparison with older cluster such as M\,34 and the Hyades strongly suggests that angular momentum of slow and moderate rotators is transported from the fast rotating core to the convective enveloppe on a time scale of about 100--200\,Myr on the early main sequence. An alternative intepretation would call for intrinsic differences in the distribution of initial angular momenta from clusters to clusters, which is not currently supported by observations. The advent, in the next decade, of multi-fiber spectrographs on large telescopes will offer the possibility of determining the {$v\sin i$}~ distributions of more remote clusters of various ages and abundances as well as the rotational properties of very low-mass stars (0.1--0.5{$M_{\odot}$}). This will undoubtedly improve our understanding of the rotational evolution of young stars and provide new clues to the physical mechanisms responsible for angular momentum transport in stellar interiors. | 98 | 3 | astro-ph9803138_arXiv.txt |
9803 | astro-ph9803248_arXiv.txt | We have studied the variability of 6 low redshift, radio quiet `PG' quasars on three timescales (days, weeks, and months) using the ROSAT HRI. The quasars were chosen to lie at the two extreme ends of the ROSAT PSPC spectral index distribution and hence of the H$\beta$ FWHM distribution. The observation strategy has been carefully designed to provide even sampling on these three basic timescales and to provide a uniform sampling among the quasars We have found clear evidence that the X-ray steep, narrow H$\beta$, quasars systematically show larger amplitude variations than the X-ray flat broad H$\beta$ quasars on timescales from 2 days to 20 days. On longer timescales we do not find significant differences between steep and flat quasars, although the statistics are poorer. We suggest that the above correlation between variability properties and spectral steepness can be explained in a scenario in which the X-ray steep, narrow line objects are in a higher $L/L_{\rm Edd}$ state with respect to the X-ray flat, broad line objects. We evaluated the power spectrum of PG1440+356 (the brigthest quasar in our sample) between $2\times10^{-7}$ and $\sim10^{-3}$ Hz, where it goes into the noise. The power spectrum is roughly consistent with a 1/f law between $10^{-3}$ and $2\times10^{-6}$ Hz. Below this frequency it flattens significantly. | \bigskip While X-ray variability in Seyfert galaxies has been the subject of intensive study (see review by Mushotzky, Done \& Pounds 1993, and Green et al 1993, Nandra et al. 1997), the variability of quasars, being fainter, have received far less attention (the Zamorani et al. 1984 study remains the most extensive to date). Higher luminosity AGN had been expected to be physically larger and so have slower, and most likely lower amplitude, variations. A typical quasar is 100--1000 times more luminous than the highly variable Seyferts, such as NGC~4051, and so would vary at a significantly lower rate, i.e. a minimum of several days. It was probably this expectation that deterred extensive observing campaigns. This is unfortunate since we show here that X-ray variability is common, rapid and of quite large amplitude ($\gs$ factor of 2). The variability most likely originates in the innermost regions of quasar, and so can help unravel the basic parameters of the quasar central engine (mass, geometry, radiation mechanisms, radiative transfer) none of which are yet well constrained. These speculations are supported by compact galactic sources, for which the investigation of the ``variability properties vs. spectral shape'' and the ``variability properties vs. luminosity'' planes have brought a great improvement in our understanding (see e.g. van der Klis 1995). The same is likely to happen for quasars. Compact galactic sources are usually bright and variable on timescales as short as 1 ms, which means that their variability timescales can be probed by just a few observations of individual objects. For example, the Galactic black hole candidate (BHC) Cyg X-1 recently underwent two dramatic changes in its variability and spectral properties (see e.g. Cui et al. 1997), and it was possible to follow the whole cycle from the usual low and hard state, to a medium-high, softer state, and back to the low and hard state in the course of a few months. This is unlikely to happen in a luminous AGN, even if there were similarities between AGN and black hole candidates, because the variability timescales (and the sizes and luminosities) are probably much longer (greater and higher) in AGN. Instead, if the analogy between quasars and BHC holds, we will observe a population of quasars some in a `high and soft' state, and some in a `low and hard' state. Hence the analogous observational way forward in quasars is to investigate the variability properties of samples of quasars, selected according to their spectral properties and luminosity. The evidence for rapid, large amplitude, X-ray variability in a few AGN with unusually steep X-ray spectra and unusually narrow Balmer lines (mostly Narrow Line Seyfert 1 galaxies, NLSy1, given the strong correlation between these two quantities found by Laor et al., 1994, 1997, Boller, Brandt \& Fink 1996) has recently grown: \begin {enumerate} \item NGC4051 shows large variations (50 \%) on timescales of $\approx 100$ seconds and has very narrow optical and UV emission lines and steep 0.1-2 keV X-ray spectrum. It is also highly variable in the EUV (a factor of factor of 20 in 8 hours, Fruscione et al 1998). \item IRAS13224-3809 has a particularly steep 0.1-2 keV spectrum ($\alpha_X>3$) narrow H$\beta$ line, very strong FeII emission and shows X-ray variations of a factor 50 or more on timescales of a few days (Otani 1995, Brandt et al. 1995, Boller et al 1997) and a factor of 2 in less than 800 seconds (Boller et al. 1996); \item RE J 1237+264 showed one very large (factor of 50) variability event (Brandt, Pounds \& Fink, 1995) \item PHL1092 varied by a factor of 4 in 2 days (Forster \& Halpern 1996, Lawrence et al. 1997). PHL1092 is a high luminosity ($5\times10^{46}$ erg s$^{-1}$) very steep 0.1-2 keV spectrum and narrow emission line quasar. \item The relatively high luminosity quasar NAB0205+024 ($L_{0.5-10 keV}=8\times10^{44}$ erg s$^{-1}$) shows variations of a factor of 2 in less than 20 ks in an ASCA observation (Fiore et al. 1998a). \item Mark 478 (PG1440+356) shows factor of 7 variations in 1 day during a long EUVE monitoring (Marshall et al 1996) \item The extremely soft NLSy1 WPVS007 ($\alpha_{0.1-2keV}=7.3$) showed a huge variation (factor of 400) variation between the RASS observation and a follow-up PSPC pointed observation taken 2 years later (Groupe et al. 1995). \end{enumerate} At this point we believe that a systematic study of a sample of normal quasars spaning the observed range of properties is needed. A systematic study of the variability properties of AGNs is not an easy task, since it requires: \begin {enumerate} \item the selection of well defined and representative (i.e. unbiased) sample; \item the availability of numerous repeated observations; \item a carefully designed observational strategy. In particular, it is very important that the sampling time is regular and that it is similar for all objects in the sample, so that the results on each object can be easily compared with each other. Otherwise the differences in the observed variability may be induced just by differences in sampling patterns. \end{enumerate} Because of these stringent requirements systematic studies are still lacking (for example the Zamorani et al. 1984 study on quasars, and the Green et al. 1993 study on Seyfert galaxies do not fulfill any of the three criteria above). To explore in a systematic manner the possible relation between line width, soft X-ray spectrum and X-ray variability properties, we initiated a pilot program using the ROSAT HRI. In the following we present the results from a campaign of observations of six PG quasars which addresses the above points. | In our sample of six PG quasars we found clear evidence that X-ray steep quasars show larger amplitude variations than X-ray flat quasars on timescales from 2 days to 20 days. On longer timescales we do not find significant differences between steep and flat quasars, although the statistics are poorer. While the sample in this pilot study is small (and expanded monitoring of a larger sample of quasars is surely needed), the distinction is so clean cut that we feel justified in speculating on its origin. Large narrow band flux variability in sources with a steep spectrum could be induced by small changes in the spectral index without large variations of the spectrum normalization. If this is the case each source should show large spectral variability, with $\alpha_X$ anti-correlated (correlated) with the observed flux if the pivot is at energies lower (higher) than the observed band. Spectral variability of this kind has not been observed in the PSPC observations of these and other quasars (Laor et al. 1997, Fiore et al. 1994) and in NLSy1 (Boller et al 1997, Fiore et al. 1998a, Brandt et al. 1995). Although our HRI observations cannot directly rule out this possibility, we therefore conclude that the observed temporal variability pattern is not likely the result of complex spectral variability. Laor et al (1997) suggest that a possible explanation for the remarkably strong $\alpha_X$--$H_{\beta}$ FWHM correlation is a dependence of $\alpha_X$ on $L/L_{\rm Edd}$. The line width is inversely proportional to $\sqrt {L/L_{\rm Edd}}$ if the broad line region is virialized and if its size is determined by the central source luminosity (see Laor et al. 1997, \S 4.7). So narrow-line, steep (0.1-2 keV) spectrum AGNs emit close to the Eddington luminosity and have a relatively low mass black hole. A similar dependence of spectral shape on $L/L_{\rm Edd}$ is seen in Galactic black hole candidates (BHC) as they change from the `soft-high' state to the `hard-low' state. A physical interpretation for this effect, as described by Pounds et al. (1995), is that the hard X-ray power-law is produced by Comptonization in a hot corona and that as the object becomes more luminous in the optical-UV, Compton cooling of the corona increases, the corona becomes colder, thus producing a steeper X-ray power-law. In BHC in `soft-high' states this power law component emerges above $\sim 10$ keV, while the spectrum below this energy is dominated by softer emission often associated with optically thick emission from an accretion disk. In this model quasar disc emission is at a too low an energy to observe, since the disc temperature scales with the mass of the compact object as $M_{BH}^{-1/4}$, leaving the Comptonized power law to dominate the 2-10 keV spectrum. The 0.2-2 keV quasar emission could be due to Comptonization (see e.g. Czerny \& Elvis 1987 and Fiore et al. 1995) by a second cooler gas component. The 0.2-2 keV and 2-10 keV spectral indices might well be correlated with each other, if the emitting regions are connected or if the emission mechanisms know about each other. We have undertaken a campaign of observations of the quasars in the Laor et al. (1997) sample with ASCA and BeppoSAX to clarify this point. Preliminary results (Fiore et al. 1998b) shows that this is indeed the case: steep $\alpha_X$(PSPC) quasars tends to have a steeper hard (2-10 keV) X-ray power-law (although the spread of the 2-10 keV indices seems smaller than that of the PSPC indices (as also found by Brandt et al. 1997). In a sample of quasars with similar luminosities, those emitting closer to the Eddington luminosity will also be those with the smaller black hole and hence smaller X-ray emission region. Thus light travel time effects would smear intrinsic X-ray variability up to shorter time scales in high $L/L_{Edd}$ objects, compared with low $L/L_{Edd}$ objects. Based on this interpretation, and on figures \ref{stru_med} and \ref{stru_mean} it appears that the emission region of the steep soft X-ray quasars is a factor of $\approx 10$ smaller than that of flat soft X-ray quasars (about $\approx 10^{16}$ cm and $10^{17}$ cm respectively). Alternatively, the higher variability of steep spectrum objects may be a true intrinsic property, which could be induced by some increased instability in high $L/L_{Edd}$ objects. The range of luminosity in our sample is too small to tell if the variability amplitude appears to be better correlated with black hole mass, as expected for light travel time smearing, or with $L/L_{Edd}$, which would indicate an intrinsic mechanism. It is interesting to note that a completely different analysis, based on the interpretation of the optical to X-ray spectral energy distribution in terms of emission from accretion disks also suggest a small black hole mass and an high arretion rate in two NLSy1 (Siemiginowska et al. 1998). It is also interesting to note that Cyg X-1 in the ``high and soft'' state (Cui et al. 1997) shows a total root mean square variability higher than that measured during periods of transitions and in the ``low and hard'' state. This is due to strong 1/f noise, extending down to at least a few $10^{-3}$ Hz, when the source is the ``high and soft'' state (Cui et al. 1997). When the source is in the ``low and hard'' state this 1/f noise is not present (see e.g. the review of van der Klis 1995). Ebisawa (1991) found in a systematic study of Ginga observations of 6 BHC that the time scales of variability for the soft and hard components are often different. The soft component is usually roughly stable on time scales of 1 day or less, while the hard component exhibits large variations down to msec time scales. If the time scales of BHC and quasars scale with the mass of the compact object the above two time scales translate to $10^4$ years and 0.1 day respectively for our quasar sample. For a sample of quasars with similar redshifts and luminosities this predicts a rather small scatter in the soft component and a bigger scatter in the hard component. ROSAT results go in this direction. Laor et al. (1994, 1997) find that (for their sample of 23 low--z PG quasars) the scatter in the normalized 2 keV luminosity is significantly larger than that in the 0.3 keV luminosity. We conclude that the analogy between AGN and Galactic BHC seems to hold qualitatively for their X-ray variability properties. An alternative and intriguing possibility to explain the correlation between X-ray variability amplitude and spectral shape is that a component generated closer to the black hole dominates the emission of steep $\alpha_X$ quasars, as in the spherically converging optically thick flow proposed by Chakrabarti and Titarchuk (1995) to characterize BHC in the high and soft state. If this is the case then we would again expect that the spectrum of the steep $\alpha_X$ quasars remain steep above 2 keV (and up to $m_ec^2$ according to Chakrabarti and Titarchuk). Observations with the ASCA and BeppoSAX satellites instruments, which are sensitive up to 10 keV, (Brandt et al 1997, Fiore et al 1998b, Comastri et al 1998) suggest that this is indeed the case. Schwartz and Tucker (1988) suggested that AGN emission can be produced by an ensemble of acceleration sites (i.e. shock waves) with different electron spectral indices and therefore emitting power laws with different photon indices. In this picture the spectrum is a quadratic function in log E and the mean index at a given energy arises from the greatest number of individual acceleration sites. Variability would be greatest for both steepest energy index and flattest energy index sources, each of which are dominated by fewer individual regions. This is contradicted by the present observations, unless the Soft X-ray flat quasars become still flatter (and more variable) above 2 keV. High energy X-ray spectroscopy and variability studies are again needed to obtain a definitive answer. A RossiXTE monitoring campaign of 4 PG quasars, with a sampling similar to that used in the HRI campaign, is in progress and could allow us to clarify this point. | 98 | 3 | astro-ph9803248_arXiv.txt |
9803 | astro-ph9803295_arXiv.txt | We have searched for molecular gas towards the nucleus of four galaxies known to harbor a water vapor megamaser. CO(1\raw0) emission of NGC 2639 and NGC 5506 was strong enough to allow us to map their inner regions. Weak emission from Mrk 1210 was detected and Mrk 1 was not detected at all. We report the tentative detection of the CO(2\raw1) line in NGC 5506. After this work, 12 of the 18 known galaxies harboring a water vapor megamaser have been observed in CO. The molecular gas content in the inner regions of water megamaser galaxies ranges from 5$\times$$10^7$ to 6$\times$$10^9~\Msun$. The circumnuclear molecular gas surface density also extends over nearly two orders of magnitude. The maser luminosity is correlated neither with the total amount of molecular gas found in the inner few kpc of these galaxies nor with global properties of the molecular gas such as surface density or filling factor; it is also independent of the infrared and optical luminosities. The only significant correlation we have found involves the maser luminosity and the low frequency radio continuum flux density. We conclude that the maser activity is intrinsically related to the energy of the active galactic nucleus whereas the intensity and even the presence of a water megamaser is independent of the molecular gas global properties such as the molecular gas content and surface density in the inner galactic regions. We have also found a pos\-sible anti\-cor\-re\-lation between the molecular gas surface density and the rate of the megamaser variations. A higher molecular gas abundance in the inner region could lead to higher maser variability because of larger nuclear flux variations due to the more variable gas infall, and/or because of more frequent interactions of the pumping agent with molecular gas condensations. | The first water megamaser was discovered towards the nucleus of NGC 4945 (dos Santos \& Lepine 1979). Strong emission at 22 GHz was detected, with a luminosity about 100 times higher than that of W49, the most powerful galactic water maser. Until 1985 four additional water megamasers were discovered, towards the nuclei of the Circinus Galaxy, NGC 1068, NGC 4258 and NGC 3079. During the following decade no more extragalactic water megamasers were found. In 1994 a survey towards active galactic nuclei was carried out by Braatz et al. (1994), leading to the discovery of five new water megamasers, doubling the number known to that date. Recently, Braatz et al. (1996) have reported the discovery of 6 additional water megamasers towards active galactic nuclei. The last ones up to now have been found in NGC 5793 (Hagiwara et al. 1997) and in NGC 3735 (Greenhill et al. 1997b). \begin{table*}[t] \begin{center} \caption{Adopted parameters for the observed water megamaser galaxies} \begin{tabular}{lcccc} \hline Parameter & NGC 2639 & NGC 5506 & Mrk 1 & Mrk 1210 \\ \hline \hline $\alpha_{2000}$$^a$ & \hms[8 43 38.0] &\hms[14 13 14.9] &\hms [1 16 7.25] & \hms[8 4 6.0 ] \\ $\delta_{2000}$$^a$ & \gms[50 12 20 3] &\gms[$-$3 12 26 7] &\gms[33 5 22 2] & \gms[5 6 50 4] \\ Morphological type$^a$ & SA(r)a & SA pec sp & S & S \\ Activity$^a$ & LINER & Sy 2 &Sy 2 &Sy 2 \\ Heliocentric velocity (\kms)$^a$ & 3236 &1816 & 4824 & 4043 \\ \VLSR (\kms) & 3235 &1825 & 4825 & 4030 \\ Distance (Mpc)$^b$ & 44 & 24 &64 &54 \\ Position angle (deg)$^c$ & 140 & 91 & --- & ---\\ Inclination (deg) & 58 & 80 & 56 & 3 \\ $D_{25}$ (arcmin)$^c$ & 1.8 & 2.8 & 0.8 & 0.8 \\ Linear scale (pc arcsec$^{-1}$) &209 &118 &312 & 260\\ $\log L_{\rm IR}$ (\Lsun)$^b$& 10.21 &10.35 & --- & 10.58\\ $\log L_{\rm FIR}$ (\Lsun)$^b$& 9.95 & 9.85 & ---& 9.85 \\ \hline \multicolumn{5}{l}{$^a$ Obtained from the NASA Extragalactic Database (NED)}\\ \multicolumn{5}{l}{$^b$ See more information in the text (Sect. 2)}\\ \multicolumn{5}{l}{$^c$ Obtained from the RC3 catalog (de Vaucouleurs et al. 1991)}\\ \end{tabular} \end{center} \end{table*} Water megamasers, unlike the galactic and normal extragalactic masers, have been detected at the nuclei of distant galaxies. Interferometric observations (Claussen \& Lo 1986; Greenhill et al. 1995a; Greenhill et al. 1996) have shown that megamaser emission comes from within a region around the galactic nucleus whose radius is in general smaller than 1 pc. Another important point is that all the galaxies harboring a water megamaser present some level of activity. This fact has led to suppose that the maser emission mechanism is identical to galactic masers. The huge difference in the energy involved is explained in terms of the pumping source: the energy source of the megamasers is the central object of the active nucleus, a much more powerful source than the central star powering the galactic masers. Braatz et al. (1997) have recently examined the conditions for detectability of water megamasers in terms of a variety of properties of the active galaxies, but not including the molecular gas content. It is known (Heckman et al. 1989) that Seyfert 2 have abnormally large quantities of dust and gas if compared with Seyfert 1 galaxies. On the other hand LINER galaxies are thought to be simply extensions of Seyfert 2 galaxies to lower luminosities, photoionized by a weaker AGN spectrum (Osterbrock et al. 1993). Thus, the fact that no water megamasers have been found in Seyfert 1 nuclei, while all of them are in either Seyfert 2 or LINER galaxies seems to indicate that a high concentration of molecular gas in the inner regions of the galaxies is a key parameter for the megamaser emission to be produced. To check this hypothesis and to find if the properties of the masers are related to those of the molecular gas, we have carried out a study of the molecular gas content and its properties in several of the water vapor megamasers for which no previous data existed or were incomplete. The megamasers we observed are four of those discovered by Braatz el al. in 1994: NGC 2639, NGC 5506, Mrk 1 and Mrk 1210. In our analysis we have used the available CO data for other water megamaser galaxies found in the literature (Heckman et al. 1989; Planesas et al. 1989; Sahai et al. 1990; Aalto et al. 1991; Wang et al. 1992 and Young et al. 1995). \begin{figure*}[t] \vspace{6cm} \special{psfile=ms7133f1.ps hoffset=-40 voffset=-456} \caption{CO(1\raw0) emission profiles towards the central region ($23\as$ diameter) of the three detected water megamaser galaxies. Velocity resolution is $21 \kms$. The two NGC objects show double peak spectra, which may indicate the existence of a molecular gas ring surrounding the central AGN at a kpc scale. The spectrum of Mrk 1210 is single-peaked, a possible indication that the expected molecular gas ring is seen face-on. This spectrum was obtained in May 1997} \end{figure*} | We have searched for molecular gas towards the nucleus of four galaxies known to harbor a water vapor megamaser, and detected the CO(1\raw0) emission in three of them and the CO(2\raw1) emission in one. With this work 12 of the 18 known water megamaser galaxies have been observed in CO, and only the most distant of the observed ones, Mrk 1, has not been detected yet. \begin{enumerate} \item The 12 water megamaser galaxies with molecular gas data available are not an homogeneous set regarding their molecular gas properties. The amount of H$_2$ in their circumnuclear regions ranges from 5$\times$$ 10^7$ to 6$\times$$ 10^9\ \Msun$. The extreme values of the H$_2$ surface density, $\Sigma_{\rm H_2}$, for the central kpc are 6 and 280 \Msun\ pc$^{-2}$. This parameter extends over a range of 2 orders of magnitude, a range similar to that of Seyfert galaxies, starburst galaxies or luminous infrared galaxies. The maser luminosity, $L_{\rm maser}$, is not correlated to the total molecular gas mass. Therefore it seems that the total amount of molecular gas in the inner few kpc is not a fundamental parameter on which depends the existence and the average intensity of the water megamaser. \item $L_{\rm maser}$ is not correlated with $\Sigma_{\rm H_2}$ or to the filling factor of giant molecular clouds. Apparently the maser luminosity does not depend on the content of molecular gas in the inner kpc. However, the accumulation of clouds of dense molecular gas is believed to be necessary for the generation of a water megamaser. Observations with higher angular resolution of the molecular gas in the inner regions would help to solve the issue. \item The only correlation we have found involving the maser emission and molecular gas parameters is between the rate of relative variation of the maser intensity and $\Sigma_{\rm H_2}$. This fact may indicate that a high abundance of molecular gas in the inner regions could lead to higher variability in the maser emission, on the one hand, due to the higher variability of the central pumping source produced by wider variations in the gas infall; on the other hand, due to the more frequent interactions of the pumping agent with molecular gas condensations. \item $L_{\rm maser}$ is not correlated to any other luminosity (infrared, optical, X-ray, blue). However, we have found some correlation between $L_{\rm maser}$ and the global radio continuum flux density at 1.4 and 8.4 GHz. This fact supports the idea that $L_{\rm maser}$ is a property related to the galactic nucleus (characterized by the radio luminosities) rather than to the inner galactic regions (characterized by the infrared luminosity and the molecular gas content). \item Mrk 1210 stands as a peculiar object, having the highest star formation efficiency among water megamaser galaxies is spite of its relatively low molecular gas content. \end{enumerate} | 98 | 3 | astro-ph9803295_arXiv.txt |
9803 | astro-ph9803156_arXiv.txt | We study the FIR and UV-visible properties of star forming galaxies in the nearby Universe. This comparison is performed using the local luminosity functions at UV and FIR wavelengths and on individual starburst galaxies for which photometric data from UV to NIR and FIR are available. The FIR and UV luminosity functions have quite different shapes~: the UV function exhibits a strong increase for low luminosity galaxies whereas the FIR tail towards ultra luminous galaxies ($\rm L > 10^{11} L\odot$) is not detected in UV. The comparison of the FIR and UV local luminosity densities argues for a rather moderate extinction in nearby disk galaxies. The galaxies selected to be detected in FIR and UV are found to be located in the medium range of both luminosity functions. An emphasis is made on starburst galaxies. For a sample of 22 of these objects, it is found that the UV (912-3650 $\rm \AA$), the visible (3600-12500 $\rm \AA$) and the NIR (12500-22000 $\rm \AA$) wavelength range contribute $\sim 30\%$, $\sim 50\%$ and $\sim 20\%$ respectively to the total emerging stellar emission (for a subsample of 12 galaxies for the NIR and visible light). The mean ratio of the dust to bolometric luminosity of these galaxies is 0.37$\pm$0.22 similar to the ratio found for normal spiral galaxies. Only 4 out of the 22 galaxies exhibit a very large extinction with more than 60$\%$ of their energy emitted in the FIR-submm range. The mean extinction at 2000$\rm \AA$ is found to be $\sim 1.2$ mag although with a large dispersion. The UV, visible and NIR emissions of our sample galaxies are consistent with a burst lasting over $\sim 1$ Gyr. The conversion factor of the stellar emission into dust emission is found to correlate with the luminosity of the galaxies, brighter galaxies having a higher conversion factor. Since our sample appears to be representative of the mean properties of the galaxy population in FIR and UV, a very large conversion of the stellar light into dust emission can no longer be assumed as a general property of starburst galaxies at least in the local Universe. Instead a larger amount of energy emerging from the present starburst galaxies seems to come from the stars rather than from the dust. We compare the UV properties of our local starburst galaxies to those of recently detected high redshift galaxies. The larger extinction found in the distant galaxies is consistent with the trend we find for the nearby starburst galaxies namely the brighter the galaxies the lower the escape fraction of stellar light. \keywords { Galaxies: starburst --Galaxies: stellar content -- Infrared: galaxies-- Ultraviolet: galaxies -- dust, extinction} | One of the most important challenge of modern astronomy is the detection of young primeval galaxies. Indeed, very significant progress has been made with the detection of very high redshift galaxies either from ground based observations (Steidel et al. 1996b) or in the Hubble Deep Field (e.g. Steidel et al. 1996a, Lowenthal et al. 1997). In order to understand the properties of high redshift galaxies and study the cosmological evolution of star-forming galaxies, it is crucial to properly characterize the properties of starburst galaxies in the local Universe. These nearby galaxies are forming stars with a very high rate and it it actually important to analyse their emission over the entire spectral range (from UV to FIR-submm) to study the efficiency of the dust extinction and know what is the spectral range (UV, visible, NIR or FIR) where most of their energy is emitted. If high redshift star forming galaxies are similar to their low-z counterparts studying the latter will bring some clues to detect the former. We can wonder whether the observation of the rest-frame UV continuum is the best way to detect high redshift galaxies or if the high obscuration from dust makes them emit more energy in the FIR. IRAS discovered infrared bright galaxies with prodigious star formation rates, a very high extinction and therefore a low optical flux (Sanders \& Mirabel 1996). The most luminous FIR galaxies are produced by strong interactions or merging of molecular gas-rich galaxies which induce enormous starbursts. Such objects might be the progenitors of elliptical galaxies (Kormendy \& Sanders 1992). In a "bottom-up" scenario of galaxy formation, numerous starbursts induced by merging are expected and the bulk of their emission would be in the FIR redshifted in the submm (e.g. van der Werf \& Israel 1996). Mazzei et al. (1994) predict that more than 90 \% of the energy emitted by a starburst is in the FIR range during the first Gyr. This percentage rapidly drops to reach $\sim 30\%$ for $\sim 5$ Gyr- old objects. Models aimed at explaining the galaxy counts in optical and FIR predict that during intense phases of star formation the quasi totality of the stellar light is absorbed and re-radiated in the FIR wavelength range (Franceschini et al. 1994, Pearson \& Rowan-Robinson 1996). Conversely, considerable effort has been carried out in the UV-optical study of star forming galaxies for some years. Calzetti, Kinney and co-workers have extensively used the IUE spectra of star-forming galaxies complemented with optical and IR data to characterize the star formation history and the extinction occurring in the central regions of these objects (Calzetti et al. 1994, Calzetti et al. 1995, Calzetti 1997a). Meurer et al. (1995) have used Faint Object Camera on board the {\it Hubble Space Telescope} HST-FOC observations to study the morphology of some starburst galaxies. From these studies, a foreground distribution of dust and a rather grey extinction curve seems to be able to explain the spectral distribution of the central regions of starburst galaxies. The extinction found by Meurer et al. (1995) for nearby starburst galaxies is rather low: at 2000 $\rm \AA$ it lies between 0.08 and 1.9 mag (excluding NGC7552 at 3.13 mag). Nevertheless these studies deal with the central parts of starburst galaxies and may well not be valid for the global emission of these objects at least when longer wavelengths than UV are concerned (Buat et al. 1997). Analyses of the UV-optical and FIR global emissions of nearby spiral and irregular galaxies selected to be observed both in UV and in FIR led to a rather low extinction (Xu \& Buat 1995, Buat \& Xu 1996, Wang \& Heckman 1996). An important result of these studies is that the UV non-ionizing stellar emission is likely to be the major cause of dust heating. The contribution of OB stars to the dust heating is estimated to amount to about 20$\%$ of the total FIR emission in the Milky Way (Cox \& Mezger 1989), almost the same contribution of the ionizing radiation to the dust heating is found for spiral galaxies (Xu \& Buat, 1995) and for starburst objects (Calzetti et al. 1995). The comparison between the FIR emission (dust re-radiation) and the UV and optical emission (escaped stellar light) constrains the extinction. As a consequence the FIR to UV continuum ratio is a powerful indicator of the extinction occurring in galaxies (Meurer et al. 1995, Buat \& Xu 1996, Wang \& Heckman 1996). Recently, HST imaging of very high redshift galaxies complemented when possible by spectroscopic observations with the Keck Telescope have led to the discovery at high redshift (z$\sim 2-3$) of compact star forming galaxies with a moderate size and a strong rest-frame UV emission (Steidel et al. 1996a) with sometimes more diffuse extended structures Lowenthal et al. 1997). Depending on the intrinsic UV spectrum adopted, the average extinction estimated at 1600 $\rm \AA$ from the rest-frame UV spectral energy distribution of these galaxies is of the order of 1.7 to 3 mag (Meurer et al. 1997, Calzetti 1997b). These significant average extinctions are therefore larger than the values estimated for nearby starburst galaxies. However, it must be noted that these high redshift galaxies are very luminous ($\rm M_B < -21$) when compared to the mean luminosity of nearby starburst galaxies studied by Meurer et al. (1995) ($\rm <M_B> = -18.6$) and the extinction is known to correlate with the luminosity of galaxies (Giovanelli et al. 1995, Wang \& Heckman 1997). Moreover, the selection biases are very strong towards very luminous galaxies with a strong UV continuum and it cannot be excluded that high redshift galaxies almost entirely hidden by the dust are missing from these observations in the rest-frame UV (Mobasher et al. 1996, Burigana et al. 1997). The selection biases in the recent detections of high z galaxies are difficult or even impossible to quantify in the absence of similar observations at other wavelengths corresponding to longer than UV rest frame emissions (NIR or FIR). A first step is to estimate the importance of such a bias in the local Universe. Such a study is also crucial to compare the properties of high redshift galaxies to those in the nearby Universe. At this aim we will adopt a global approach to study the local Universe which consists in comparing the luminosity functions and the luminosity densities in UV and FIR. The UV wavelength range is particularly interesting since it is a tracer of the recent star formation rate as already mentioned and observations in the visible range of high z ($>2$) galaxies correspond to their UV rest frame emission. More specifically we will compare the amount of energy locked up in FIR to the amount of energy directly emitted in UV in the local Universe. The comparison of these global values with individual galaxies selected to be observed both in UV and FIR will allow to discuss how such samples of individual galaxies are representative of the mean properties of the local Universe. We will also investigate the specific case of a sub sample of nearby starburst galaxies detected in UV, visible, NIR and FIR in order to compare their global dust and stellar emission and to estimate what fraction of the emission of stars is converted into dust emission as well as the relative contribution of the UV, visible and NIR spectral ranges to the observed stellar emission. Such estimates will lead to predict what spectral range is more favorable for the detection of high redshift starbursts under the hypothesis that they are similar to their nearby counterparts. The main limitation to this approach is that we deal with global fluxes integrated over the galaxies whereas the starburst often occurs in the central parts. Moreover the galaxies at high redshift so far detected seem to have compact morphologies. Nevertheless as it will be shown in section 4 a large fraction of the UV emission of a starburst galaxy is likely to come from the starburst itself making valid a study on the global fluxes as soon as this wavelength is concerned. It is also the case for the FIR emission since the UV (ionizing and non ionizing) emissions is the major contributor to the dust heating), especially in starbursting objects. Obviously, more care must be taken when dealing with the visible and NIR emission: at these wavelengths the contribution of the underlying old stellar population present in local starburst galaxies is very large even dominant. Endly, available FIR fluxes on large samples are integrated over the galaxies due to the poor resolution of the IRAS satellite and dealing with global fluxes allows a reliable comparison of the emission of galaxies at different wavelengths. Beyond the detection of high redshift star forming galaxies it is necessary to estimate a quantitative star formation rate (SFR) for these galaxies. The deduction of such a quantity from the observed rest-frame UV continuum relies almost entirely on the amount of the extinction with only a moderate dependence on the star formation history (e.g. Meurer et al. 1997, Calzetti 1997). More specifically after $\sim 5~10^7$ years of constant star formation rate the UV flux (912-3650 $\rm \AA$) reaches 80 $\%$ of its stationnary value calculated for $\rm 10^{10}$ years of constant star formation (from Bruzual \& Charlot 1993). From a global energetic budget, we will try to constrain the amount of extinction and bring some clues to this difficult problem. | \subsection{ The local Universe} We have considered galaxies selected to be detected photometrically both in UV and FIR. An analysis of the local luminosity functions at both wavelengths shows that the galaxies selected in this way have a FIR and a UV emission representative of the mean population of galaxies in the local Universe. From a sample of 22 starburst galaxies, the mean escape fraction of the stellar light (ratio of the stellar luminosity to the bolometric one) is found to be $63\%\pm 22\%$, i.e. similar to what is found for normal galaxies. This escape fraction exhibits a strong decrease with increasing galaxy luminosity ranging from $\sim 80\%$ at faintest B magnitudes ($\rm M_B~>-17$) to $\sim 10\%$ for $\rm M_B~\sim~ -20$. This result is quite different to that of Pearson \& Rowan-Robinson (1996) who found that the escape fraction of the stellar light in starburst galaxies cannot exceed 5-10$\%$ from an analysis of deep counts in visible and NIR and assuming a very strong cosmological evolution for the starburst population. Such a low fraction of escaped light is consistent with the studies of FIR bright galaxies but seems not to be a generic property of starbursting objects, at least in the local Universe. \subsection {Comparison with high redshift galaxies} We can compare our results found for nearby starbursts to high redshift galaxies recently detected. A limitation to this comparison might be that we deal with integrated fluxes on nearby galaxies which contain an underlying stellar population pre-existing to the starburst. To our knowledge, there is no evidence for the presence or not of such an older population in high redshift starburst galaxies (z $\sim$ 3). Nevertheless, as discussed in the paper, the UV range is largely dominated by the emission of the newly formed stars and the comparison of the high redshift galaxies to present-day starburst ones in this wavelength range is justified. From an analysis of the slope of the rest-frame UV continuum, an extinction has been estimated for high z galaxies (Meurer et al. 1997, Calzetti 1997b, Pettini et al. 1997). A mean extinction of 1.65 mag at 1650 $\rm \AA$ is found by Calzetti for a star formation rate constant over $\sim 1$ Gyr which implies an unreddened slope for the UV continuum $\rm \beta=-2$. This value is in agreement with Pettini et al.'s estimates. Meurer et al. have adopted a different star formation law with a starburst lasting 10 $\rm Myr$, leading to a larger proportion of very young stars and a steeper UV slope ($\rm \beta=-2.5$). Such a very extreme scenario gives an extinction as large as 3 mag. at 1600 $\rm \AA$ which can be considered as a very upper limit. The extinction found for nearby starburst galaxies is lower than the value found by Calzetti or Pettini et al. at high redshift. But the galaxies observed at high redshift are much more luminous that the nearby starburst galaxies studied by us or by Meurer et al. (1995) (Meurer et al. 1997, Lowenthal et al. 1997, Pettini et al. 1997). As an example, an extinction of 1.6 mag at 1600 $\rm \AA$ gives $\rm F_{dust}~/~F_{bol} = 0.6$ and $\rm F_{dust}~/~F_{UV} = 4.5$ using the extinction curve of Calzetti et al. (1994), a foreground screen and a star formation rate constant over 1 Gyr. Extrapolating Fig.~5, these ratios correspond to the brightest galaxies with $\rm M_B\le -20$. If the trends found in Fig.5 are mainly due to the extinction, the extinction estimated in high redshift star forming galaxies is in agreement with the values found for nearby starburst galaxies when the galaxy luminosity is accounted for. We must point out, however, that low luminosity galaxies ( $\rm M_B \sim -16$) would be detectable with a NGST-type telescope in a reasonable amount of time as far as $\rm z\sim 9$ (if they exist). In a hierarchical scenario for the formation of the galaxies, it is likely that such galaxies would outnumber the large luminous galaxies presently detected (Ellis 1997). Obviously the knowledge of the local Universe properties in terms of the extinction occurring in galaxies as well as their spectral energy distribution, especially in the UV range, is essential to interpret the observations of the Universe at high redshift. It is also crucial to predict the best way to detect young galaxies during a phase of intense star formation. We show that a comparison of the FIR and UV emissions in nearby galaxies can bring some clues since these two emissions are very sensitive to the current star formation rate and to the extinction. A more straightforward method would be a systematic comparison of deep fields in UV and FIR. Such a preliminary comparison was already performed (Buat \& Xu 1996) using FAUST (Deharveng et al. 1994) and IRAS observations on two 7.6$\rm ^o$-wide fields in the central region of the Virgo cluster. Although the UV observations of FAUST were not very deep (limiting flux $\rm 10^{14} - 10^{15}~ erg/cm^2/s/\AA$ at 1600 $\rm \AA$), we have searched for galaxies detected by IRAS without any UV detected counterpart. Given the low sensitivity of the FAUST experiment, the lower limits obtained for the ratio of the FIR to UV luminosity are within the range of values found for the galaxies detected at both wavelengths (fig. 4 in Buat \& Xu 1996). This is in agreement with the results found in this paper but needs to be confirmed with deeper observations. The UV observations carried out with the large field ($\rm \sim 2^o$) FOCA telescope (e.g. Donas et al. 1990) reach the magnitude 18 at 2000$\rm \AA$. Unfortunately up to now the FIR observations (made by the IRAS satellite) are not deep enough to be compared to these UV observations and we have to wait for new FIR large field instruments like WIRE to perform such a comparison. | 98 | 3 | astro-ph9803156_arXiv.txt |
9803 | astro-ph9803010_arXiv.txt | Metal line ratios in a sample of 13 quasar spectra obtained with the HIRES spectrograph on the KeckI telescope have been analyzed to characterize the evolution of the metagalactic ionzing flux near a redshift of 3. The evolution of \ion{Si}{4}/\ion{C}{4} has been determined using three different techniques: using total column densities of absorption line complexes, as in Songaila \& Cowie\markcite{sc96} (1996); using the column densities of individual Voigt profile components within complexes; and using direct optical depth ratios. All three methods show that \ion{Si}{4}/\ion{C}{4} changes abruptly at $z \sim 3$, requiring a jump in value of about a factor of 3.4, and indicating a significant change in the ionizing spectrum that occurs rapidly between $z = 2.9$\ and $z = 3$, just above the redshift at which Reimers et al.\markcite{reim} (1997) detected patchy \ion{He}{2} ${\rm Ly}\alpha$\ absorption. At lower redshifts, the ionization balance is consistent with a pure power law ionizing spectrum but at higher redshifts the spectrum must be very soft, with a large break at the He$^+$\ edge. An optical depth ratio technique is used to measure the abundances of ions whose transitions lie within the forest and \ion{C}{3}, \ion{Si}{3} and \ion{O}{6} are detected in this way. The presence of a significant amount of \ion{O}{6} at $z > 3$\ suggests either a considerable volume of \ion{He}{3} bubbles embedded in the more general region where the ionizing flux is heavily broken, or the addition of collisional ionization to the simple photoionization models. | \label{intro} Whereas we know from the absence of any significant Gunn-Peterson effect even in the highest redshift quasars that hydrogen reionization of the intergalactic gas must have taken place at $z > 5$, we have much less information about the period at which the bulk of singly ionized helium converted to doubly ionized helium. Since late He$^+$\ ionization may significantly change the temperature of the intergalactic gas it is critical to understand this heating if we are to correctly model the growth of structure in the IGM, and determine the mapping of the baryon density to observable quantities such as observations of the neutral hydrogen ${\rm Ly}\alpha$\ forest and the He$^+$\ ${\rm Ly}\alpha$\ opacity. Phenomenological modelling of this event depends critically on the softness of the composite spectrum of the ionizing sources (e.g.\ Miralda-Escud\'e \& Rees\markcite{mr93} 1993; Madau \& Meiksin\markcite{mm} 1994), amplified by the subsequent radiative transfer, and such models cannot be considered reliable in predicting the high energy ($E > 54~{\rm eV}$) metagalactic ionizing spectrum above the He$^+$\ ionization edge, which determines the fraction of singly ionized helium. Our most direct information on the He$^+$\ opacity is through observations of quasars whose spectra extend to the He$^+$\ ${\rm Ly}\alpha$\ wavelength. Despite the extreme difficulty of these measurements, successful observations of the He$^+$\ ${\rm Ly}\alpha$\ absorption have been made toward $z > 2.8$\ quasars with HST (Jakobsen et al.\markcite{jak} 1994; Hogan et al.\markcite{hog} 1997; Reimers et al.\markcite{reim} 1997) and of the $z = 2.72$\ quasar HS1700+6416 with HUT (Davidsen et al.\markcite{dkz} 1996; Zheng et al.\markcite{zdk} 1998). The He$^+$\ ${\rm Ly}\alpha$\ opacity shows a marked decrease from a value of $\tau = 3.2^{+\infty}_{-1.1}$\ in Q0302$-$003 at $z = 3.29$\ to $\tau = 1.0 \pm 0.07$\ in HS1700+6416. More remarkably, the Reimers et al.\markcite{reim} observation of the intermediate redshift quasar HE2347$-$4342 at $z = 2.89$\ shows both `troughs' and `voids' in the \ion{He}{2} ${\rm Ly}\alpha$\ observations, suggesting that at $z \sim 2.8$\ we are seeing fully ionized \ion{He}{3} bubbles interspersed among as yet unionized He$^+$\ regions and that it is at this point that the porosity of \ion{He}{3} regions is approaching unity. The recent discovery that the bulk of ${\rm Ly}\alpha$\ forest absorption with $N({\rm H~I}) > 3\times 10^{14}~\rm cm^{-2}$\ contains associated metals (Songaila \& Cowie\markcite{sc96} 1996, hereafter SC) gives us an alternative approach to the problem since ionization balance in these forest metals provides a diagnostic of the shape of the metagalactic flux in the neighborhood of the He$^+$\ edge. As was first noted in Songaila et al.\markcite{shc} (1995), the value of \ion{Si}{4}/\ion{C}{4} in high ionization systems is critically dependent on the He$^+$\ ionization edge break strength. This has subsequently been investigated in more detail by SC, Savaglio et al.\markcite{sav} (1997) and Giroux \& Shull\markcite{gs97} (1997), among others. As will be discussed further here, the bulk of the forest metal line systems at the currently observed redshifts ($z \sim 2 - 4$) are high ionization (\ion{C}{2}/\ion{C}{4} $\ll 0.1$) and so, unless there is a strong break at the edge, they will have \ion{Si}{4}/\ion{C}{4} $\ll 0.1$\ even for the higher Si/C abundances characteristic of low metallicity systems. Therefore, once the He$^+$\ is fully ionized and the IGM becomes relatively transparent to the integrated quasar spectrum (e.g.\ in the metagalactic spectrum of Haardt \& Madau\markcite{hm} 1996), the observed \ion{Si}{4}/\ion{C}{4} values should fall in this low range. SC and Savaglio et al.\markcite{sav} (1997) have shown that, whereas this is generally true at $z < 3$, much higher \ion{Si}{4}/\ion{C}{4} values are regularly seen at $z > 3$, suggesting a significant change above this redshift, which would be consistent with the interpretation of the Reimers et al.\markcite{reim} (1997) observations as showing the redshift at which \ion{He}{3} bubbles begin to overlap. Boksenberg\markcite{bok} (1998) has recently questioned this result, based on an analysis of the the redshift evolution of the ion ratios in the separate Voigt profile components in complex systems rather than in the integrated column densities of the complexes. However, his analysis is based on rather a small sample of systems without clear selection criteria. In this paper, I shall use the largest sample to date to demonstrate unambiguously that, irrespective of the method of analysis, there is indeed a rapid jump in the value of \ion{SI}{4}/\ion{C}{4} at a redshift just below 3, and that the ionization stages in the metals are consistent with this being, for most systems, the point at which they change from being ionized by a metalgalactic spectrum that is, at $z > 3$, heavily broken above 54~eV to one that is only mildly broken at the lower redshifts. The sample and the data reduction are described in \S 2 and the reader who is primarily interested in the results could safely skip this section. The evolution of the \ion{C}{2}/\ion{C}{4} and \ion{Si}{4}/\ion{C}{4} values with redshift is described in \S 3 where I show that the presence of a jump in \ion{Si}{4}/\ion{C}{4} values at $z$\ just under 3 is highly significant and does not depend on the method of analysis, whether by total column densities of complex, by the column densities of individual Voigt components, or by directly analyzing the distribution of optical depth ratios. In \S 4 I consider the overall ionization balance including intermediate ionization stages such as \ion{C}{3} and \ion{Si}{3} and high ions such as \ion{N}{5} and \ion{O}{6} using, for those lines that lie primarily within the forest, the optical depth distributions of the ensembles of such lines, which provides a new and robust technique for determining their properties. The overall ionization balance is broadly consistent with unbroken power law photoionization at $z < 3$; however, the observation of significant amounts of \ion{O}{6} at $z > 3$\ requires either the presence of a considerable volume of \ion{He}{3} bubbles permeating regions where the ionizing flux is heavily broken, or the addition of collisional ionization to the simple photoionization models. Finally, the conclusions are briefly summarised in \S 5. | \label{conc} I summarise the results of the paper by noting that, irrespective of the analysis methodology adopted, there is a significant change in the ionization balance of forest metal lines which occurs just below a redshift of 3. At lower redshifts, the ionization balance in the forest lines is fully consistent with a pure power law ionization spectrum with an index of $-1.8$\ but at higher redshifts the high values of \ion{Si}{4}/\ion{C}{4} seen in most of the forest clouds despite generally low \ion{C}{2}/\ion{C}{4} values implies that the ionizing flux must be very soft, with a large break at the He$^+$\ edge. The change occurs quite rapidly between $z = 2.9$\ and $z = 3$, just above the redshift at which highly patchy \ion{He}{2} ${\rm Ly}\alpha$\ absorption is seen in the quasar HE~2347$-$4342 (Reimers et al.\markcite{reim} 1997). The simplest explanation seems to be that we are seeing the redshift at which \ion{He}{2} ionizes completely to \ion{He}{3} as the \ion{He}{3} Str\"omgren spheres overlap. | 98 | 3 | astro-ph9803010_arXiv.txt |
9803 | astro-ph9803226_arXiv.txt | The 6.4~s X-ray pulsar \src\ was observed by \sax\ in 1997 May. This source belongs to the class of ``anomalous'' pulsars which have pulse periods in range 5--11~s, show no evidence of optical or radio counterparts, and exhibit long-term increases in pulse period. The phase-averaged 0.5--10 keV spectrum can be described by an absorbed power-law and blackbody model. The best-fit photon index is 2.5$\pm$0.2 and the blackbody temperature and radius are 0.64$\pm$0.01~keV and $0.59 \pm 0.02$~km (for a distance of 3~kpc), respectively. The detection of blackbody emission from this source strengthens the similarity with two of the more well studied ``anomalous'' pulsars, 1E\,2259+586 and 4U\,0142+614. There is no evidence for any phase dependent spectral changes. The pulse period of $6.45026 \pm 0.00001$~s implies that \src\ continues to spin-down, but at a slower rate than obtained from the previous measurements in 1994 and 1996. | \label{sec:introduction} \src\ is an unusual pulsar, with a period of 6.4~s and a soft spectrum, discovered during {\it Einstein}\/ observations of the Carina nebula (Seward et al.\ 1983). The source has been spinning down for at least the last 17 years (Mereghetti 1995; Corbet \& Mihara 1997). The spectrum has been modeled by an absorbed power-law with a photon index, $\alpha$, of $\sim$2--3 (Seward et al.\ 1983; Corbet \& Mihara 1997). This spectral shape is softer than those of typical high-luminosity X-ray pulsars. Despite a small error box, no optical counterpart has been identified, with a limiting magnitude of $m_{V}\sim20$ which excludes the presence of a massive companion (Mereghetti et al.\ 1992). A recent observation with the {\it RossiXTE}\/ satellite provides a strong upper limit to the projected X-ray semi-major axis of 0.06~lt-s for orbital periods between 200~s and $\sim$1~day (Mereghetti et al.\ 1997). The lack of an optical counterpart and orbital Doppler shifts argue against a binary model for \src, unless the companion has a low mass. Mereghetti et al.\ (1997) find that a probable upper limit to the mass of a Roche lobe filling main sequence companion is $\sim$0.3 \Msun. Masses up to $\sim$0.8 \Msun\ are allowed in the case of a helium-burning companion filling its Roche lobe. The properties described above are similar to those of a small number of other X-ray pulsars with spin periods in the 5--11~s range (Mereghetti \& Stella 1995), such as 1E\,2259$+$586 and 4U\,0142+614. These form a class of so-called ``anomalous'' pulsars, with clearly different properties from the majority of systems. Although accretion from a very low mass companion cannot be excluded, the lack of evidence for a binary nature from any of these systems has stimulated models where the X-ray emission originates from a compact object that is not in an interacting binary system. While an isolated, massive, white dwarf powered by the loss of rotational energy, as originally proposed for 1E\,2259$+$586 (Paczynski 1990; Usov 1994), has been ruled out by the detection of a large increase in the spin-down rate of \src\ (Mereghetti 1995), other single object models, such as loss of magnetic energy of a strongly magnetized neutron star (Thompson \& Duncan 1993), or an isolated neutron star accreting from a circumstellar disk (Corbet et al. 1995; van Paradijs et al. 1995), may be applicable. We present a study of \src\, based on data obtained with the \sax\ satellite. We focus on the X-ray spectrum at energies $<$10 keV and on the pulse period history. As with some of the other ``anomalous'' pulsars, we find evidence for the presence of a blackbody spectral component. Since this component is not observed from the majority of other accreting X-ray pulsars, this strengthens the similarity between \src\ and the other better studied ``anomalous'' pulsars. | The 0.5--10~keV spectrum of \src\ can be described by the sum of an absorbed power-law and blackbody models. The existence of a blackbody component in \src\ was first suggested by the ASCA results of Corbet \& Mihara (1997). However, these authors are unable to clearly discriminate between this two component model, which gives a \rchisq\ of 0.94 for 332 dof, and an absorbed power-law which also provides an acceptable description of the ASCA data with a \rchisq\ of 1.02 for 337 dof. The \sax\ results reported here clearly require that the \src\ spectrum differs significantly from an absorbed power-law. This deviation in consistent with a blackbody, but we cannot exclude the possibility that it has another form. A discussion about the physical interpretation of the alternate models can be found in White et al.\ (1996). This discussion is applicable to \src, since its spectral shape is roughly similar to that of 4U\,0142+614. The reason for the significant \sax\ detection of spectral complexity is probably related to the combination of good energy resolution and extended low energy coverage of the LECS, together with the longer \sax\ exposure. 1E\,2259+586 (Corbet et al.\ 1995) and 4U\,0142+614 (White et al.\ 1996) have also been successfully fit with absorbed power-law and blackbody spectral models. Since the majority of X-ray pulsar spectra are fit by absorbed power-law models in the 0.5--10~keV energy range (e.g., White et al. 1983), our results strengthen the similarity between \src\ and the other ``anomalous'' pulsars. Table \ref{tab:bbprops} is a compilation of the spectral properties of these ``anomalous'' pulsars. The blackbody component in these systems has been interpreted as evidence for quasi-spherical accretion onto an isolated neutron star formed after common envelope evolution and spiral-in of a massive X-ray binary (White et al.\ 1996). In this case, the accretion flow results from the remaining part of the massive star's envelope and may consist of two components (Ghosh et al. 1997). A low-angular momentum component gives rise to the blackbody emission from a considerable fraction of the neutron star surface, while a high-angular momentum component forms an accretion disk and is responsible for the power-law emission and the long term spin-down evolution. The area for the blackbody emitting surface obtained for \src\ ($\sim$0.59 d$_{\rm 3kpc}$ km) is smaller than those of the other ``anomalous'' systems. However, the distance to \src\ is poorly constrained and the assumed distance of 3~kpc could be considered as a lower limit since the measured ${\rm N_H}$ implies that it lies behind the Carina Nebula at 2.8~kpc (Seward et al.\ (1986). White et al.\ (1996) propose that the low pulsed fraction observed from 4U\,0142+614 results from the large polar cap area in this system. This is consistent with the small polar cap area and high pulsed fraction reported here for \src\, but not with 1E\,2259+586 and 1RXS{\thinspace}J170849.0$-$400910, which both have large radii and a moderate pulsed fraction (see Table \ref{tab:bbprops}). Interestingly, the best-fit blackbody temperature of 0.64~keV is somewhat higher than for 1E\,2259+586 and 4U\,0142+614 which may be related to the smaller area of the polar caps. Summarizing, we find that for \src\ (i) the blackbody temperature is higher, (ii) the blackbody radius is smaller, (iii) the power-law index is smaller (i.e. the spectrum is harder), and (iv) the pulsed fraction is higher than for the other ``anomalous'' pulsars (with the exception of RX{\thinspace}J0720.4$-$3125 where the blackbody parameters are obtained from a one component fit). The phase-resolved spectra are not well fit with a constant contribution from the blackbody component. This is unsurprising given the large pulse amplitude ($\sim$70\%), together with the large blackbody contribution (55\%; 2--10~keV) to the phase averaged flux. The probable constancy of the pulsed fraction over the 1--10~keV energy range and the lack of any spectral dependence on pulse phase (see Sect.\ \ref{subsec:src_pulsetiming}) imply that the {\it whole} spectrum is varying in a similar manner; i.e. the pulsed component cannot be attributed solely to either the blackbody or the power-law components. Furthermore, the long term variations in source flux (Fig.\ \ref{fig:fluxhistory}) and the approximate constancy of the pulsed fraction (Table~\ref{tab:pdata}) during this time again implies that either the luminosities of the two components are correlated, or that the underlying spectral shape is a more complex ``single component'' that happens to mimic a power-law and a blackbody. The spin-down rate of \src\ obtained from ROSAT and ASCA observations (Mereghetti 1995; Corbet \& Mihara 1997), showed an 80\% increase with respect to the value of $\sim5 \times 10^{-4}$ s yr$^{-1}$ measured before 1988 with \ginga\ and EXOSAT. This increasing trend was further extended by the {\it RossiXTE}\/ observation of 1996 July ($P$ = $6.449769 \pm 0.000004$ s, Mereghetti et al.\ 1997), yielding a $\dot{P}$ of 13$\times10^{-4}$ s yr$^{-1}$ after the ASCA measurement. The pulse period obtained with \sax\ lies significantly below the linear extrapolation from the previous two measurements, indicating, for the first time in \src, a decrease in the overall spin-down rate. Table \ref{tab:pdata} and Fig.\ \ref{fig:fluxhistory} indicate that there is no clear correlation between the observed long term spin-down rate ($\dot{P}$) and the 2--10 keV source flux. At first sight this argues against an accretion hypothesis, but since $\dot{P}$ is a long-term average and the flux an instantaneous measurement this comparison may not be valid. A much better comparison would be between the average flux and $\dot{P}$ during the same time interval. Unfortunately, due to the faintness of the source, no such comparison is available. In view of the uncertainties in the average flux a good measure of the time variability of the source on timescales of months to years would be useful. | 98 | 3 | astro-ph9803226_arXiv.txt |
9803 | astro-ph9803082_arXiv.txt | The EROS and MACHO collaborations have each published upper limits on the amount of planetary mass dark matter in the Galactic Halo obtained from gravitational microlensing searches. In this paper the two limits are combined to give a much stronger constraint on the abundance of low mass MACHOs. Specifically, objects with masses $10^{-7}~\msun \simlt m \simlt 10^{-3}~\msun$ make up less than $25$\% of the halo dark matter for most models considered, and less than $10$\% of a standard spherical halo is made of MACHOs in the $3.5\ee{-7}~\msun < m < 4.5\ee{-5}~\msun$ mass range. | 98 | 3 | astro-ph9803082_arXiv.txt |
||
9803 | astro-ph9803177_arXiv.txt | We present a simple method for determining the (correlated) uncertainties of the light element abundances expected from big bang nucleosynthesis, which avoids the need for lengthy Monte Carlo simulations. Our approach helps to clarify the role of the different nuclear reactions contributing to a particular elemental abundance and makes it easy to implement energy-independent changes in the measured reaction rates. As an application, we demonstrate how this method simplifies the statistical estimation of the nucleon-to-photon ratio through comparison of the standard BBN predictions with the observationally inferred abundances. | Big bang nucleosynthesis is entering the precision era \cite{Sc97}. On the one hand, there has been major progress in the observational determination of the abundances of the light elements D \cite{Bu97,We97}, $^3$He \cite{Ba94,Gl96}, $^4$He \cite{Ol97a,Iz97}, and $^7$Li\cite{Ry96,Bo97}, although the increasing precision has highlighted discrepancies between different measurements (see Refs.\cite{Mo97,Ho97,Le97} for recent assessments). Secondly, we have a sound analytical understanding of the physical processes involved \cite{Be89,Es91} and the standard BBN computer code \cite{Wa73,Ka92} which incorporates this physics is robust and can be easily altered to accomodate changes in the input parameters, e.g.\ nuclear reaction rates \cite{Sm93}. The comparison of increasingly accurate observationally inferred and theoretical abundances will further constrain the values of fundamental parameters, such as the nucleon density parameter (see, e.g., Ref.\cite{Ol97b}) or extra degrees of freedom related to possible new physics beyond the Standard Model (see, e.g., Ref.\cite{Sa96}). It goes without saying that error evaluation represents an essential part of such comparisons. Because of the complex interplay between different nuclear reactions, it is not straightforward to assess the effect on a particular elemental yield of the uncertainties in the experimentally determined reaction rates. The authors of Ref.\cite{Kr90} first employed Monte Carlo methods to sample the error distributions of the relevant reaction cross-sections which were then used as inputs to the standard BBN computer code. This enables well-defined confidence levels to be attached to the theoretically predicted abundances, e.g. the abundance range within which say 95\% of the computed values fall correspond to 95\% C.L. limits on the expected abundance. It was later realized that error correlations are also relevant, and can be estimated with the same technique \cite{Ke94,KeKr}. The Monte Carlo (MC) approach has since become the standard tool for comparing theory and data \cite{Sm93,Kr94,Co95,Ha95,Ol97c}. However, although it can include refinements such as asymmetric or temperature-dependent reaction rate uncertainties \cite{Sm93}, it requires lengthy calculations which need to be repeated each time (any of) the input parameters are changed or updated. Since we may expect continued improvement in the determination of the relevant parameters, it is desirable to have a faster method for error evaluation and comparison with observations. In this work we propose a simple method for estimation of the BBN abundance uncertainties and their correlations which requires little computational effort. The method, based on linear error propagation, is described in Sec.~II. A concrete application is given in Sec.~III, where theory and observations are compared using simple $\chi^2$ statistics to obtain the best-fit value of the nucleon-to-photon ratio. In Sec.~IV we study with this method the relative importance of different nuclear reactions in determining the synthesized abundances. Conclusions and perspectives for further work are presented in Sec.~V. | We have shown that a simple method based on linear error propagation allows us to quantify the uncertainties associated with the elemental abundances expected from big bang nucleosynthesis, in excellent agreement with the results obtained from Monte Carlo simulations. This method makes transparent which nuclear reaction rate is mainly responsible for the uncertainty in the abundance of a given element. If determinations of the primordial abundances improve to the point where the observational errors become smaller than the theoretical uncertainties (say for $^7$Li), this will enable attention to be focussed on the particular reaction rate whose value needs to be experimentally better known. We have also demonstrated that for standard BBN, our method enables the use of simple $\chi^2$ statistics to obtain the best-fit value of $\eta$ from the comparison of theory and observations. At present there are conflicting claims regarding the primordial abundances of, particularly, D and $^4$He, and different choices of input data sets imply values of $\eta$ differing by a factor of $\sim\,3$. However this quantity can also be determined through measurements of the angular anisotropy of the cosmic microwave background (CMB) on small angular scales. Within a decade the forthcoming all-sky surveyors MAP and PLANCK are expected to pinpoint the nucleon density to within $\sim5\%$ \cite{cmb}. Such measurements probe the acoustic oscillations of the coupled photon-matter plasma at the (re)combination epoch and will thus provide an independent check of BBN, assuming $\eta$ did not change significantly between the two epochs.% \footnote{New physics beyond the Standard Model can change $\eta$, e.g. by increasing the photon number through massive particle decay \cite{decay} or, more exotically, by {\em decreasing} the photon number through photon mixing with a shadow sector \cite{Ba91}. However such possibilities are strongly constrained by the absence of distortions in the Planck spectrum of the CMB \cite{El92} and also, in the latter case, by the absence of Sakharov oscillations in the power spectrum of large-scale structure \cite{Bi97}.} Nevertheless precise measurements of light element abundances, particularly $^4$He, are still crucial because they provide a unique probe of physical conditions, in particular the expansion rate at the BBN epoch. To illustrate, if $\eta$ was determined by the CMB measurements to be $\approx\,2\times10^{-10}$ (consistent with data set ``A''), but the abundance of $^4$He was established to be actually closer to its higher value of $\approx\,24\%$ in data set ``B'', this would be a strong indication that the expansion rate during BBN was higher than in the standard case with $N_\nu=3$ neutrinos. Although the number of $SU(2)$ doublet neutrinos is indeed 3, there are many light particles expected in extensions of the Standard Model, e.g. singlet neutrinos, which can speed up the expansion rate during nucleosynthesis \cite{Sa96}. The generalization of our method to such non-standard cases is straightforward and we intend to present these results in a future publication \cite{us}. It is clear that BBN analyses will continue to be important in this regard for both particle physics and cosmology. | 98 | 3 | astro-ph9803177_arXiv.txt |
9803 | astro-ph9803207_arXiv.txt | Hubble Space Telescope observations of the gravitational lens PG~1115+080 in the infrared show the known $z_l=0.310$ lens galaxy and reveal the $z_s=1.722$ quasar host galaxy. The main lens galaxy G is a nearly circular (ellipticity $\epsilon < 0.07$) elliptical galaxy with a de Vaucouleurs profile and an effective radius of $R_e = 0\farcs59\pm0\farcs06$ ($1.7 \pm 0.2 h^{-1}$ kpc for $\Omega_0=1$ and $h = H_0/100$ \kmm). G is part of a group of galaxies that is a required component of all successful lens models. The new quasar and lens positions (3 milliarcsecond errors) yield constraints for these models that are statistically degenerate, but several conclusions are firmly established. (1) The principal lens galaxy is an elliptical galaxy with normal structural properties, lying close to the fundamental plane for its redshift. (2) The potential of the main lens galaxy is nearly round, even when not constrained by the small ellipticity of the light of this galaxy. (3) All models involving two mass distributions place the group component near the luminosity-weighted centroid of the brightest nearby group members. (4) All models predict a time delay ratio $r_{ABC}\simeq 1.3$. (5) Our lens models predict $H_0=44\pm4$ \kmm\ if the lens galaxy contains dark matter and has a flat rotation curve, and $H_0=65\pm5$ \kmm\ if it has a constant mass-to-light ratio. (6) Any dark halo of the main lens galaxy must be truncated near $1\farcs5$ ($4 h^{-1}$ kpc) before the inferred \Ho\ rises above $\sim 60$ \kmm. (7) The quasar host galaxy is lensed into an Einstein ring connecting the four quasar images, whose shape is reproduced by the models. Improved NICMOS imaging of the ring could be used to break the degeneracy of the lens models. | Gravitational lens time delays offer a means of determining the Hubble constant that is purely geometrical and hence completely avoids the complications of the local distance scale (Refsdal 1964). The time delay for Q~0957+561 is now well-measured (Schild \& Thomson 1997; Kundi\'c et al. 1997; Haarsma et al. 1997), but significant systematic uncertainties remain due to the degeneracy between the mass of the primary lens galaxy and its host cluster (e.g. Grogin \& Narayan 1996; Bernstein et al. 1997; Romanowsky \& Kochanek 1998). No single lens is likely to be completely free of systematic uncertainties, so a reliable estimate of $H_0$ should rely on an ensemble of lenses. There are now three more systems with time delay estimates: PG~1115+080 (Schechter et al. 1997), B~1608+656 (Fassnacht et al. 1996), and B~0218+357 (Corbett et al. 1996), which need detailed exploration of their lens models to examine the systematic uncertainties. PG~1115+080 was the second gravitationally lensed quasar to be discovered (Weymann et al. 1980). The source is an optically selected, radio-quiet quasar at redshift $z_s=1.722$. Hege et al. (1981) first resolved the four quasar images (a close pair A1/A2, B and C), confirming the early model of Young et al. (1981) that the lens was a five-image system, one image being hidden in the core of the lens galaxy. Henry \& Heasly (1986) detected the lens galaxy, followed by gradual improvements in the astrometry by Kristian et al. (1993; hereafter K93), and Courbin et al. (1997). The redshift of the lens galaxy was determined by Angonin-Willaime, Hammer \& Rigaut (1993) and confirmed by Kundi\'c et al. (1997) and Tonry (1998) to be $z_l=0.310$. Tonry also determined the central velocity dispersion of the lens galaxy: $\sigma = 281\pm25$ km s$^{-1}$. The spatial resolution of published data has always been insufficient to perform any surface photometry on the lens galaxy. Young et al. (1981) noted that the lens seemed to be part of a small group centered to the southwest of the lens, with a velocity dispersion of approximately $270\pm70$ km s$^{-1}$ based on only four galaxy redshifts (Kundi\'c et al. 1997). The group is an essential component of any model that successfully fits the lens constraints (Keeton, Kochanek \& Seljak 1997; Schechter et al. 1997). Finally, Schechter et al. (1997) successfully determined two time delays between the images, which were reanalyzed by Barkana (1997) to give $\Delta\tau_{BC}=25.0^{+1.5}_{-1.7}$ days and the time delay ratio $r_{ABC}=\Delta\tau_{AC}/\Delta\tau_{BA}=1.13\pm0.18$. These results were analyzed by Keeton \& Kochanek (1997) and Courbin et al. (1997) to deduce $H_0=53_{-7}^{+15}$ km s$^{-1}$ Mpc$^{-1}$, with comparable contributions to the uncertainties from the time delay measurement and the models. The extreme variations are given in non-parametric form by Saha \& Williams (1997), although some of these models may not be physical. We present new near-infrared observations of the PG~1115+080 system obtained with the Hubble Space Telescope (HST) NICMOS camera. These are the first results of the CfA-Arizona Space Telescope Lens Survey (CASTLES).\footnote{A summary of gravitational lens data and model results, including CASTLES data, is available at the URL http://cfa-www.harvard.edu/castles.} After summarizing the observations in \S2, we present improved astrometry in \S2.1, the first surface photometry of the lens galaxy in \S2.2, a discussion of lens models and the Hubble constant in \S2.3, photometry of the nearby group in \S2.4, and comments on the quasar host galaxy in \S2.5. In \S3 we comment on the strengths and limitations of this system as a cosmological tool. | New infrared data on PG~1115+080 affirms multiple-component gravitational lens systems as powerful cosmological tools. The major puzzle remaining in the PG~1115+080 system is the anomalous A1/A2 flux ratio. Our observations rule out differential extinction as an explanation, and microlensing is ruled out by its lack of variability. Since a flux ratio near 0.9 is a generic feature of the large scale potential near a fold caustic, only a potential perturbation intermediate between that produced by isolated stars (microlensing) and by the overall galaxy can explain the flux ratio. The potential of PG~1115+080 must be perturbed either by a satellite galaxy or a globular cluster. Mao \& Schneider (1998) showed that such perturbations alter the time delay -- and so the inferred value of the Hubble constant -- fortunately by no more than 2-3\%. Our improved astrometry greatly reduces some of the degeneracies in early models of the system. The group position is now well constrained and located near the luminosity centroid of the four bright group galaxies, and the lens galaxy is constrained to be nearly circular. Unfortunately, the degeneracies in the $H_0$ estimate have been exacerbated because with the revised astrometry the models no longer favor dark matter models over constant $M/L$ models for the main lens galaxy. Because all 4 images are located at nearly the same radial distance from the center of the lens galaxy, we do not expect the models to be sensitive to the radial mass profile of the lens galaxy (Kochanek 1991; Wambsganss \& Pa\'czynski 1994). We find \Ho\ ranging from $44\pm4$ \kmm\ if the lens galaxy is modeled as a singular isothermal ellipsoid and the group as a singular isothermal sphere, to $65\pm5$ ($72\pm5$) \kmm\ for $\Omega_0 = 1$ ($0.1$) if the lens galaxy has a constant $M/L$. Note that we find evidence for dark matter in our high value of $M/L$. A model with an adjustable truncation radius shows that the halo must be truncated on scales comparable to the ring diameter for $H_0$ to exceed $60$ \kmm. Such a halo seems smaller than physically plausible given that the velocity dispersions of the group and the lens galaxy are comparable. Further progress in reducing the uncertainties depends on improving the time delay measurements and on making more detailed studies of the Einstein ring formed by the quasar host galaxy. First, for any given mass profile, most of the current errors in \Ho\ are due to the uncertainties in the time delays. Second, all the best fit models predict time delay ratios near $r_{ABC}=1.3$, consistent with the current measurement of $1.13 \pm 0.18$. If nothing else, a more accurate measurement of the delay ratio than is now available would be a powerful test of the models. For any given lens mass profile, the ratio constraint would further reduce the parameter space for the position and mass of the group, or could be used to constrain more complicated models (e.g., Saha \& Williams 1997). Deep new observations to determine the surface brightness of the ring accurately, combined with direct measurement of the point spread function at the time of the observations would probably permit us directly to break the degeneracy of the models. Only NICMOS, however, has the ability to make these difficult observations within a decade. No single technique or observation can tie down the Hubble constant -- the long history of unrecognized or underestimated systematic errors in this subject encourages humility. Nevertheless, PG~1115+080 demonstrates the potential of the gravitational lens approach. With recent reductions in age estimates of the oldest globular clusters (Chaboyer et al. 1998), and the likelihood that the mass density of the universe is lower than the Einstein-de Sitter case (e.g., Garnavich et al. 1998), the possibility of an age conflict in the big bang model has receded. The modeling of PG~1115+080 gives a plausible upper bound on the Hubble constant if we accept that the group is not a point mass and that the lens galaxy is unlikely to have a mass distribution that is more concentrated than its light distribution. This bound is $H_0 < 67$ ($72$) \kmm\ for $\Omega_0 = 1$ ($0.1$). The most recent result of the HST Extragalactic Distance Scale Key Project is $H_0 = 72 \pm5$ (random) $\pm12$ (systematic) km s$^{-1}$ Mpc$^{-1}$ (Madore et al. 1998). Our upper limit is inconsistent with the upper end of the range from the Key Project, although it is consistent with the lower end of the range. There appears to be satisfactory concordance among the basic parameters of the big bang model, and between direct and indirect measures of the distance scale. Gravitational lenses can be expected to play an increasing role as versatile cosmological tools. \vskip 1truecm | 98 | 3 | astro-ph9803207_arXiv.txt |
9803 | astro-ph9803031_arXiv.txt | In this paper, we measure the ellipticities of 30 LSB dI galaxies and compare the ellipticity distribution with that of 80 dEs (Ryden \& Terndrup 1994; Ryden et al.\ 1998)\markcite{ryden 1994; 1998} and 62 BCDs (Sung et al.\ 1998).\markcite{sung1998} We find that the ellipticity distribution of LSB dIs is very similar to that of BCDs, and marginally different from that of dEs. We then determine the distribution of intrinsic shapes of dI galaxies and compare to those of other type dwarf galaxies under various assumptions. First, we assume that LSB dIs are either all oblate or all prolate, and use non-parametric analysis to find the best-fitting distribution of intrinsic shapes. With this assumption, we find that the scarcity of nearly circular LSB dIs implies, at the 99\% confidence level, that they cannot be a population of randomly oriented oblate or prolate objects. Next, we assume that dIs are triaxial, and use parametric analysis to find permissible distributions of intrinsic shapes. We find that if the intrinsic axis ratios, $\beta$ and $\gamma$, are distributed according to a Gaussian with means $\beta_0$ and $\gamma_0$ and a common standard deviation of $\sigma$, the best-fitting set of parameters for LSB dIs is $(\beta_0,\gamma_0,\sigma) = (0.66,0.50,0.15)$, and the best fit for BCDs is $(\beta_0,\gamma_0,\sigma) = (0.66,0.55,0.16)$, while the best fit for dEs is $(\beta_0,\gamma_0,\sigma) = (0.78,0.69,0.24)$. The dIs and BCDs thus have a very similar shape distribution, given this triaxial hypothesis, while the dEs peak at a somewhat more spherical shape. Our results are consistent with an evolutionary scenario in which the three types of dwarf galaxy have a close relation with each other. | Although the faint end of the galaxy luminosity function is not well determined, recent studies indicate that low-surface-brightness (LSB) dwarf galaxies are by far the most numerous type of galaxy, and contribute a significant fraction of the mass of the universe (Reaves 1983; Binggeli, Sandage, \& Tammann 1985; Phillipps et al.\ 1987)\markcite{reaves1983, sandage1985, phillipps1987}. Morphologically, dwarf galaxies, like their counterpart bright galaxies, are classified into several types. The most common type of dwarf galaxy ($\sim 80\%$ of the total) is the dwarf elliptical (dE). These galaxies have regular elliptical isophotes and roughly exponential surface brightness profiles; they are often found in groups and clusters (Davies et al.\ 1988).\markcite{davies1988} The second type of dwarf galaxy is the blue compact dwarf (BCD) galaxy. In contrast to gas-poor dEs, BCDs contain giant HII regions surrounding O and B stars within a massive HI reservoir; BCDs exhibit spectra slowly rising toward the blue, implying that they are undergoing intense star formation (du Puy 1970; Searle \& Sargent 1972).\markcite{dupuy1970, searle1972} Most BCDs have regular isophotes in the outer region, like dEs, but the inner isophotes are frequently distorted from ellipses, due to the presence of bright HII regions (Loose \& Thuan 1986).\markcite{loose1986} The final type of dwarf galaxy is the LSB dwarf galaxy. LSB dwarfs include both irregular (dI) and more regular spiral (dS) galaxies. Like BCDs, they contain a large amount of HI, often with small OB associations, and have blue colors ($B-V\sim 0.5\ {\rm mag}$), indicating a significant level of recent star formation (Staveley-Smith, Davies, \& Kinman 1992)\markcite{staveley1992}. However, they are distinguished from BCD galaxies by their amorphous shapes even in the outer region. Additionally, in contrast to dEs, these galaxies are more likely to be found outside of clusters (Bingelli, Tarenghi, \& Sandage 1990).\markcite{bingelli1990} The evolutionary connections among the three different types of dwarf galaxies remain both elusive and confusing. There are two major competing hypotheses for the evolutionary connection between BCDs and dEs. The first hypothesis claims that BCDs are basically a different population from dEs, as evidenced by the spectroscopic and spectrophotometric differences. According to this scenario, BCDs are truly young systems, in which the present star burst is the first in the galaxy's lifetime. The second hypothesis suggests that BCDs, like dEs, are mainly composed of old stellar populations, and that their observed spectroscopic features and spectral energy distributions are the result of a recent burst of star formation (Staveley-Smith et al.\ 1992).\markcite{stavely1992} As an evidence for the second scenario, it is argued that the near-infrared emission in the vast majority of BCDs is attributable to old K and M giants, which are the major component of dEs (Thuan 1983; Hunter \& Gallagher 1985).\markcite{thuan1983, hunter1985} Similarly, there exist two competing hypotheses to explain the evolutionary connection between dIs and dEs. The first hypothesis states that dE galaxies are the faded remnants of previously actively star-forming dI galaxies whose gas has been lost. There exists circumstantial evidence that dEs have in fact evolved directly from dIs. Faber \& Lin (1983)\markcite{faber1983} and Kormendy (1985)\markcite{kormendy1985} have used the similarity in the surface brightness profiles of dIs and dEs, which are mostly exponential, to argue that gas-rich dIs are the progenitors of dEs. The second hypothesis for the relation between dIs and dEs states that they represent parallel sequences of dwarf galaxies, fundamentally separated by the intrinsic difference in their structure. The observational evidence for this hypothesis is based mostly on the differences in appearance between the two types of dwarf galaxies; for instance, dIs have a more diffuse light distribution than dEs, and lack the bright nucleation which is frequently found in dEs. In addition to these differences, there is a dissimilarity in the flattening distribution of dEs and dIs; the apparent flattening of a galaxy is customarily given either by the apparent axis ratio $q$ or by the ellipticity $\epsilon \equiv 1 - q$. Bothun et al.\ (1986) and Impey \& Bothun (1997)\markcite{bothun1986, impey1997} presented the results of Ichkawa, Wakamatsu, \& Okamura (1986)\markcite{ ichikawa1986} and Caldwell (1983)\markcite{caldwell1983} as evidence for the different flattening distributions between dEs and dIs. However, the analysis of Ichikawa et al.\ was based on the comparison between the flattening distributions of dEs and bright (non-dwarf) spiral galaxies; the situation is similar for Caldwell's analysis. In addition, contrary to the claims of Bothun et al.\ and Impey \& Bothun, both Ichikawa et al.\ and Caldwell showed that the flattening distribution of dEs is similar to that of bright irregular galaxies. There have been previous attempts to compare the apparent axis ratio distributions between LSB dI galaxies and other types of dwarf galaxies. For example, Staveley-Smith et al.\ (1992)\markcite{stavely1992} constructed the axis ratio distribution for 438 Uppsala Galaxy Catalogue (hereafter UGC, Nilson 1973)\markcite{nilson1973} LSB galaxies, and compared it to that of BCDs whose ellipticities were measured from Palomar Observatory Sky Survey (POSS) plates by Gorden \& Gottesman (1981).\markcite{gorden1981} However, previous studies of axis ratio distributions suffer from large uncertainties for several reasons. First, owing to the small dimensions and low surface brightness of dwarf galaxies, estimating their axis ratio is difficult and leads to large uncertainties. Second, the UGC sample used by Staveley-Smith et al.\ (1992)\markcite{stavely1992} is known to be inhomogeneous, containing galaxies ranging from true dwarf galaxies to more luminous very low surface brightness systems (Thuan \& Seitzer 1979; McGaugh, Schombert, \& Bothun 1995).\markcite{thuan1979, mcgaugh1995} Finally, previous determinations of LSB dI axis ratios have been based on photographic plates; for comparison with recent CCD observations of other types of dwarf galaxy, it is essential to have measurements of the axis ratios of a homogeneous sample of LSB dIs based on modern CCD observations. In this paper, we measure the ellipticities of 30 LSB dI galaxies and compare the ellipticity distribution with that of 80 dEs (Ryden \& Terndrup 1994; Ryden et al.\ 1998)\markcite{ryden 1994; 1998} and 62 BCDs (Sung et al.\ 1998).\markcite{sung1998} We find that the ellipticity distribution of LSB dIs is very similar to that of BCDs, and marginally different from that of dEs. We then determine, under various assumptions, the distribution of intrinsic shapes of dI galaxies and compare it to that of other types of dwarfs. First, we assume that LSB dIs are either all oblate or all prolate, and use non-parametric analysis to find the best-fitting distribution of intrinsic shapes. With this assumption, we find that the scarcity of nearly circular LSB dIs implies, at the 99\% confidence level, that they cannot be a population of randomly oriented oblate or prolate objects. Next, we assume that dIs are triaxial, and use parametric analysis to find permissible distributions of intrinsic shapes. We find that if the intrinsic axis ratios, $\beta$ and $\gamma$, are distributed according to a Gaussian with means $\beta_0$ and $\gamma_0$ and a common standard deviation of $\sigma$, the best-fitting set of parameters for LSB dIs is $(\beta_0,\gamma_0,\sigma) = (0.66,0.50,0.15)$, and the best fit for BCDs is $(\beta_0,\gamma_0,\sigma) = (0.66,0.55,0.16)$, while the best fit for dEs is $(\beta_0,\gamma_0,\sigma) = (0.78,0.69,0.24)$. The LSB dIs and BCDs thus have a very similar shape distribution, given this triaxial hypothesis, while the dEs peak at a somewhat more spherical shape. Our results are consistent with an evolutionary scenario in which the three types of dwarf galaxy have a close relation with each other. | We measure the ellipticities for a sample of 30 LSB dIs and compare the distribution of ellipticities with those for the samples of 62 BCDs and 80 dEs. From this comparison, we find that the axis ratio distribution of LSB dIs is very similar to that of BCDs. Compared to dEs, LSB dIs are slightly flatter, but the difference is marginal. We also determine the intrinsic shape of LSB dIs from the distribution of apparent axis ratios. From the non-parametric analysis, we find the hypothesis that our sample LSB dIs are randomly oriented oblate or prolate objects is rejected with strong confidence level. On the other hand, the shape of LBS dI galaxies are well described by triaxial spheroids if their axis ratios, $\beta$ and $\gamma$, have a Gaussian distribution. From the parametric analysis, we determine the best-fitting parameters are $(\beta_0,\gamma_0,\sigma) = (0.66, 0.50, 0.15)$. These results directly contradict the long-standing belief that LSB dIs have very flattened disky shapes, quite different from the spheroidal shapes of dEs and BCDs. Therefore, our results are consistent with the scenario that the three major types of dwarf galaxies have very close evolutionary connections. | 98 | 3 | astro-ph9803031_arXiv.txt |
9803 | astro-ph9803025_arXiv.txt | The 48~ms radio pulsar PSR~B1259$-$63 is in a highly eccentric 3.4~yr orbit with a Be star (Johnston et al. 1992). Near periastron, for a sufficiently strong Be star wind, PSR~B1259$-$63 could possibly make a transition to an accretion regime, exhibiting X-ray pulsations of luminosity $>$10$^{35}$~erg~s$^{-1}$. {\it ASCA} X-ray observations at three epochs around the 1994 periastron passage (Kaspi et al. 1994; Hirayama 1996) found the source to be unpulsed, with modest luminosity $\sim$10$^{34}$~erg~s$^{-1}$, for a distance of 2~kpc. These properties imply that shock emission produces the observed X-rays, rather than accretion (Tavani \& Arons 1997). However radio timing observations suggested sudden, brief accretion events near periastron that could not be ruled out by the X-ray observations (Manchester et al. 1995). Indeed the pulsar's anomalously low period suggests it may have undergone occasional accretion episodes in the past. {\it ASCA} observations during the 1997 periastron passage were impossible due to solar constraints. The ASM on {\it RXTE} has been observing bright celestial X-ray sources regularly since early 1996. The instrument consists of three ``scanning shadow cameras,'' each containing a position-sensitive proportional counter that is mounted below a wide-field collimator, covered by a coded mask. The instrument provides roughly 5 celestial scans per day, with diminished exposure in directions toward the Sun. Further information on the ASM is given by Levine et al. (1996). While PSR~B1259$-$63 is not routinely monitored for the ASM source-history database, we extracted an X-ray light curve as an archival analysis project using standard techniques. The derived light curve (Figure~1a) covers the time interval of MJD 50178--50762 (1996 Apr 5 to 1997 Nov 10). The overall mean intensity of PSR B1259$-$63 during this time interval was $(0.5 \pm 2.4)\times 10^{-11}$~erg~s$^{-1}$~cm$^{-2}$ at 2--10 keV. This result is given with consideration of both the systematic bias and uncertainty in the ASM light curves for faint, yet fairly isolated X-ray sources. Figure~1b shows $3\sigma$ upper limits obtained in the periastron vicinity in 2-day averages. It shows that in the 2 weeks following periastron, the source was not observed to be brighter than $\sim$10 times the 1994 periastron luminosity. The $3\sigma$ upper limits to the X-ray emission from PSR~B1259$-$63 from the ASM thus argue that the pulsar did not undergo even brief episodes of accretion during the 1997 periastron passage. This is consistent with the shock emission model, as well as with new conclusions based on timing that rule out the sudden ``spin-ups'' that were claimed previously (Wex et al. 1998). \begin{figure}[t] \centerline{ \psfig{figure=psr1259.ps,height=5.5cm} \psfig{figure=psr1259_upper.ps,height=5.5cm} } \caption{(a) ASM light curve for PSR~B1259$-$63. (b) 3$\sigma$ upper limits to the ASM flux near periastron from (a). The horizontal dashed line represents the brightest {\it ASCA} 1994 periastron detection (Kaspi et al. 1994), and the vertical dotted line indicates periastron.} \end{figure} | The 48~ms radio pulsar PSR~B1259$-$63 is in a highly eccentric 3.4~yr orbit with a Be star (Johnston et al. 1992). Near periastron, for a sufficiently strong Be star wind, PSR~B1259$-$63 could possibly make a transition to an accretion regime, exhibiting X-ray pulsations of luminosity $>$10$^{35}$~erg~s$^{-1}$. {\it ASCA} X-ray observations at three epochs around the 1994 periastron passage (Kaspi et al. 1994; Hirayama 1996) found the source to be unpulsed, with modest luminosity $\sim$10$^{34}$~erg~s$^{-1}$, for a distance of 2~kpc. These properties imply that shock emission produces the observed X-rays, rather than accretion (Tavani \& Arons 1997). However radio timing observations suggested sudden, brief accretion events near periastron that could not be ruled out by the X-ray observations (Manchester et al. 1995). Indeed the pulsar's anomalously low period suggests it may have undergone occasional accretion episodes in the past. {\it ASCA} observations during the 1997 periastron passage were impossible due to solar constraints. The ASM on {\it RXTE} has been observing bright celestial X-ray sources regularly since early 1996. The instrument consists of three ``scanning shadow cameras,'' each containing a position-sensitive proportional counter that is mounted below a wide-field collimator, covered by a coded mask. The instrument provides roughly 5 celestial scans per day, with diminished exposure in directions toward the Sun. Further information on the ASM is given by Levine et al. (1996). While PSR~B1259$-$63 is not routinely monitored for the ASM source-history database, we extracted an X-ray light curve as an archival analysis project using standard techniques. The derived light curve (Figure~1a) covers the time interval of MJD 50178--50762 (1996 Apr 5 to 1997 Nov 10). The overall mean intensity of PSR B1259$-$63 during this time interval was $(0.5 \pm 2.4)\times 10^{-11}$~erg~s$^{-1}$~cm$^{-2}$ at 2--10 keV. This result is given with consideration of both the systematic bias and uncertainty in the ASM light curves for faint, yet fairly isolated X-ray sources. Figure~1b shows $3\sigma$ upper limits obtained in the periastron vicinity in 2-day averages. It shows that in the 2 weeks following periastron, the source was not observed to be brighter than $\sim$10 times the 1994 periastron luminosity. The $3\sigma$ upper limits to the X-ray emission from PSR~B1259$-$63 from the ASM thus argue that the pulsar did not undergo even brief episodes of accretion during the 1997 periastron passage. This is consistent with the shock emission model, as well as with new conclusions based on timing that rule out the sudden ``spin-ups'' that were claimed previously (Wex et al. 1998). \begin{figure}[t] \centerline{ \psfig{figure=psr1259.ps,height=5.5cm} \psfig{figure=psr1259_upper.ps,height=5.5cm} } \caption{(a) ASM light curve for PSR~B1259$-$63. (b) 3$\sigma$ upper limits to the ASM flux near periastron from (a). The horizontal dashed line represents the brightest {\it ASCA} 1994 periastron detection (Kaspi et al. 1994), and the vertical dotted line indicates periastron.} \end{figure} | 98 | 3 | astro-ph9803025_arXiv.txt |
|
9803 | astro-ph9803163_arXiv.txt | We describe a concept for an imaging spectrograph for a large orbiting observatory such as NASA's proposed Next Generation Space Telescope (NGST) based on an imaging Fourier transform spectrograph (IFTS). An IFTS has several important advantages which make it an ideal instrument to pursue the scientific objectives of NGST. We review the operation of an IFTS and make a quantitative evaluation of the signal-to-noise performance of such an instrument in the context of NGST. We consider the relationship between pixel size, spectral resolution, and diameter of the beamsplitter for imaging and non-imaging Fourier transform spectrographs and give the condition required to maintain spectral modulation efficiency over the entire field of view. We give examples of scientific programs that could be performed with this facility. | The Association of Universities for Research in Astronomy (AURA), with NASA support, recently appointed a committee to ``study possible missions and programs for UV-Optical-IR astronomy in space for the first decades of the twenty-first century.'' The report urged the development of a general-purpose, near-infrared observatory equipped with a passively cooled primary mirror ($T \le 70$~K) with a minimum diameter of 4 meters (Dressler 1996). To enhance its performance, the report recommended that the observatory be placed as far from the Earth-Moon system as possible to reduce stray light and to maintain the telescope's relatively low temperature. With such a facility, it should be possible to learn in detail how galaxies formed, measure the large-scale curvature of space-time by measuring distant standard-candles, trace the chemical evolution of galaxies, and study nearby stars and star-forming regions for signs of planetary systems. A detailed discussion of the next generation space telescope (NGST) and its scientific potential given by Stockman (1997). For NGST to attain these scientific objectives, it must have an instrument which is designed to execute panchromatic observations over the critical 1--15 $\mu$m wavelength range of the faintest detectable objects. With nJy sensitivity levels attainable at near-infrared (1--5 $\mu$m) (NIR) and mid-infrared (5--15$\mu$m) (MIR) wavelengths, NGST will be able to study the well-calibrated rest-frame optical diagnostics in distant ($z=3-10$) galaxies, thus probing for the first time, their stellar content, star-formation history and nuclear activity. At the longer wavelengths, NGST can investigate these properties in $z=3-5$ galaxies using diagnostics that are unaffected by dust extinction and reddening, and also study the dust properties directly. At the flux limits characteristic of NGST, the confusion limit is likely to be approached, with virtually every pixel having significant information (e.g., by extrapolation from counts in the Hubble Deep Field (Williams et al. 1996)). As a result, one of the best ways to maximize the scientific output from NGST is to provide a wide-field imaging spectrograph that is efficient in this limit. An imaging Fourier transform spectrometer (IFTS) provides these capabilities in a low-cost, high throughput, compact design. It provides the only efficient means of conducting {\it unbiased} spectroscopic surveys of the high-$z$ Universe, i.e., without object preselection (e.g., using broad band colors) and without the restrictions imposed by spectrometer slit geometry and placement. An IFTS also allows spectroscopy over a wide bandpass, affords flexibility in choice of resolution, is easy to calibrate, and is ideal for wide-field spectroscopic surveys. Bennett et al. (1993) and Bennett, Carter, \& Fields (1995) describe the operating principles of imaging Fourier transform spectrographs and compare their performance with alternative imaging spectrometers. A comprehensive review of the application of interferometers and the techniques of Fourier spectroscopy to astrophysical problems is given by Ridgway \& Brault (1984), and a recent summary of the field, including a description of an astronomical IFTS is given by Maillard (1995). Spaceborne Fourier transform spectrometers have been responsible for spectacular results in the fields of planetary exploration and cosmology. Infrared FT spectrometers developed at the Goddard Space Flight Center (GSFC) flew on board the Mariner 9 mission to Mars, and were carried to the outer planets by the Voyager spacecraft (Hanel et al. 1992). The instruments provided superb data revealing, for the first time, the composition of the atmospheres of the giant gaseous planets (e.g. Jupiter, Hanel et al. 1979). The Composite Infrared Spectrometer (CIRS), currently traveling to Saturn onboard the Cassini spacecraft, is another instrument developed at GSFC. CIRS is the first step towards an imaging FTS as it has a linear array of detectors, rather than a single element detector, in order to map the temperature and composition of the atmospheres of Saturn and Titan as a function of altitude during limb soundings. (Kunde et al. 1996). The definitive measurement of the spectrum of the cosmic microwave background radiation (CMBR) was one of the most dramatic experimental measurements of this decade (Mather et al. 1990; Gush, Halpern \& Wishnow 1990). The FIRAS instrument onboard the NASA satellite COBE which first performed this measurement, and the COBRA rocket experiment conducted by the University of British Columbia which confirmed it a few months later, were both liquid-helium cooled, differential Fourier transform spectrometers. These instruments used a dual-input, dual-output configuration where one input viewed the sky and the other viewed a blackbody calibrator (Mather, Fixen \& Shafer 1993; Gush \& Halpern 1992). Absolute photometric measurements were obtained by reference to the blackbody calibrator, and the CMBR was observed to have an undistorted Planck spectrum corresponding to a temperature of $2.728\pm0.004$ K (Fixen et al. 1996). The IFTS proposed here can be thought of as an extension of these experiments where focal plane detector arrays yield simultaneous imaging and spectral information. In the next decade, missions such as WIRE, AXAF, and SIRTF will expand astrophysical horizons, possibly unveiling entirely new populations of objects. An IFTS offers the flexibility (e.g., spectral resolution) that may prove essential in investigating the nature of these sources. Due to its flexibility and its ability to provide simultaneous imaging and spectroscopy of every object in the field of view (FOV), an IFTS is a {\it necessary} instrument for the NGST mission. \section {IFTS CONCEPT} \label{concept} An IFTS (Fig. \ref{fts}) is axis-symmetric, and the optical path difference (OPD) is the same for all the points of the image with the same angle of incidence from the axis of the interferometer. Hence, the FOV is circular. On the object side, an entrance collimator illuminates the interferometer with parallel light. The interfering beams are collected by the output camera, creating a stigmatic relation between the object and image planes. By placing a detector array in the output focal plane the entrance field is imaged on the array and each pixel works as a single detector matched to a point on the sky. \begin{figure}[thb] \centerline{\psfig{figure=fig1.eps,width=3.3in,angle=-90}} \caption{\small A sketch of the optics of a simple single-beam imaging Fourier transform spectrograph consisting of a collimating lens, a beam splitter, two mirrors (one movable), and a camera lens. The optical path difference is $x$.} \label{fts} \end{figure} Retrieving spectral information involves recording the interferogram generated by the source imaged onto the focal plane array (FPA). The OPD is scanned in discrete steps since FPAs are integrating detectors. Scanning in this way generates a data cube of two-dimensional interferograms. The signal from the same pixel in each frame forms an independent interferogram. These interferograms are Fourier transformed individually yielding a spectral data cube composed of the same spatial elements as the image. \subsection{A Perfect Match to NGST Science} \label{perfect} The features of an IFTS which make it the instrument of choice for NGST are efficiency, flexibility, and compactness. The most compelling reason for choosing an IFTS is that in the dual port design (see Fig. \ref{optical-layout}) virtually every photon collected by the telescope is directed towards the focal plane for detection. Other solutions are inefficient, inflexible, and wasteful of mass, power, and volume. Cameras equipped with filters admit only a restricted bandpass at low spectral resolution. To compete with the spectral multiplex advantage of an IFTS, a camera system needs multiple dichroics and FPAs. The additional mass and thermal load is a severe penalty. Classical dispersive spectrographs have slit losses, grating inefficiencies due to light lost in unwanted orders, and limited free spectral range (the same is true for a Fabry-Perot). {\em An IFTS acquires full bandpass imaging simultaneously with higher spectral resolution data.} Therefore, a high SNR broad-band image always accompanies full spectral sampling of the FOV with no penalty in integration time. An IFTS is a true imaging spectrograph and measures a spectrum for every pixel in the FOV. It is not necessary to choose which regions in the image are most deserving of spectroscopic analysis. Overheads are eliminated because no additional observing time is needed for imaging prior to object selection, and there is no delay in positioning slit masks, fibers, or image slicing micro-mirrors. Thus, an IFTS will produce a rich scientific legacy with tremendous potential for serendipity. \begin{figure}[thb] \centerline{ \psfig{figure=fig2.ps,width=3.3in,angle=-90}} \caption{\small Schematic optical layout of a 60$^\circ$ dual-input, dual-output Michelson interferometer. } \label{optical-layout} \end{figure} Table \ref{capabilities} details the capabilities of a putative IFTS suitable for NGST. We use the instrument described by this table to illustrate the potential of an IFTS. Two points in Table \ref{capabilities} must be stressed: 1) An IFTS is spectrally multiplexed, therefore all spectral channels are obtained simultaneously within the stated integration time. 2) The free spectral range of an IFTS is limited only by the band-pass filter and the detector response. Consequently, the usual definition of resolution, $R = \lambda/\delta \lambda$, is of limited use. It is conventional to scan the OPD of an IFTS in equal steps so that the resolution is constant in wavenumber, $k$. Thus, we use $M$ to denote the number of spectral channels. For example, in the NIR with a 1-5 $\mu$m band-pass, $M=5$ means that $\delta k = (k_{max} - k_{min})/M = 1600$~cm$^{-1}$, and a scan yields 5 bands centered 1.1, 1.3, 1.7, 2.3, \& 3.6 $\mu$m. \begin{deluxetable}{lll} \tablewidth{0pt} \tablecaption{Capabilities of a Putative NGST IFTS} \tablehead{ \colhead{} & \colhead{NIR Channel} &\colhead{MIR Channel}} \startdata Design & Dual-port & Dual-port \\ Bandpass & 1-5 $\mu$m & 5-15 $\mu$m \\ Resolution & 1 cm$^{-1}$ & 1 cm$^{-1}$ \\ FOV & $200''$ & $100''$ \\ Pixel size & $0.''05$ & $0.''1$ \\ Array format & 4k$\times$4k & 1k$\times$1k \\ Detector & InSb & HgCdTe \\ Throughput & $> 0.5$ & $> 0.5$ \\ Sensitivity\tablenotemark{a} \\ ~$M=1$ & 200 pJy & 13 nJy \\ ~$M=5$ & 1 nJy & 65 nJy \\ ~$M=100$ & 35 nJy & 1.3 $\mu$Jy \\ \enddata \tablenotetext{a}{SNR = 10 for a $10^5$~s integration over the entire spectral band for a point source. $M$ is the number of simultaneous spectral channels in the band-pass --- see \S \ref{perfect}. Note that all spectral channels are obtained simultaneously. The spectrum is assumed to be flat in $F_\nu$ and the SNR is quoted at 2 $\mu$m for the NIR channel, and at 10 $\mu$m for the MIR channel.} \label{capabilities} \end{deluxetable} The throughput of an IFTS with ideal optics is only limited by the efficiency of the beam splitter. In a dual-input, dual-output port design no light is wasted and the throughput approaches 100\%. An IFTS has no loss of light or spatial information because there is no slit, hence an IFTS is perfectly adapted to doing multi-object spectroscopy in crowded or confusion limited fields. A IFTS uses every photon whereas traditional cameras and spectrographs throw away photons (either spectrally with a filter or spatially with a slit), so at a very fundamental level an IFTS is superior. On blaze, a good grating is 80\% efficient, but averaged over the free spectral range this drops to about 65\%. An IFTS is not optimized for single-object spectroscopy because the broad-band photon shot noise is associated with every frame in the interferogram. Hence, for a single object a slit spectrograph is $\eta_g \eta_s M$ times faster than an IFTS of the same resolution in background limited operation, where $\eta_g$ is the grating efficiency averaged over the blaze function, and $\eta_s$ is the slit loss, where typically the product $\eta_g \eta_s \approx 0.3$. This disadvantage is more than compensated for by the spatial-multiplexing capability of an IFTS. A typical deep background limited exposure of an IFTS will reach $K=29.5$, SNR=10 and will contain at least 3500 and possibly, depending on cosmology, up to 11,000, objects per field (see Fig. \ref{number-counts}). A grating spectrograph with a fiber feed or multi-slit capability can perhaps record spectra for only a few percent of these objects at a time, requiring hundreds of pointings to make an unbiased survey of a single field, as opposed to the single IFTS imaging-cum-spectroscopic observation. \begin{figure}[thb] \centerline{\psfig{figure=fig3.ps,width=3.4in,angle=90}} \caption{\small $K$-band number counts (Djorgovski et al. 1995; Gardner et al. 1993, 1996; Glazebrook et al. 1994; Huang et al. 1997; Mobasher et al. 1986; Moustakis et al. 1997; McLeod et al. 1995; \& Metcalfe et al. 1996) together with models of the luminosity function modeled using the formalism of Gardner (1998), which has been used to extrapolate the number counts into the NGST domain. The solid lines include the effects of passive evolution, while the dashed lines include only K-corrections. The upper line in each case is for $q_0 = 0.1$, and the lower lines are for $q_0 = 0.5$. Current number counts imply at least 3500 objects per $3.'3$ NGST field, while the extrapolations shown here suggest as many as 11,000 to $K=29.5$. } \label{number-counts} \end{figure} An IFTS is tolerant of detector noise because it always operates under photon limited conditions due to the broad spectral bandpass transmitted to the FPA. This is illustrated in Table \ref{readoutexamples} which shows a break-down of the noise sources in the NIR and MIR channels corresponding to the performance listed in Table \ref{capabilities}. Table \ref{readoutexamples} also shows that the read-out rates required to avoid saturation are modest ($1-10$~mHz), since typical well depths for NIR InSb or HgCdTe arrays are a few $10^5$ e$^-$ and $10^7$ e$^-$ for MIR Si:As arrays. \begin{deluxetable}{llll} \tablewidth{0pt} \tablecaption{Signal and Noise Budget\tablenotemark{a}} \tablehead{ \colhead{} & \colhead{} & \colhead{NIR Channel} &\colhead{MIR Channel}} \startdata & & $F$ = 1 nJy & $F$ = 65 nJy \\ & & $t$ = 1000 s & $t$ = 100 s \\ & & $\Delta\lambda = 1-5\mu$m & $\Delta\lambda = 5-15\mu$m \\ \hline Signal \\ & Source & 610 & 2709 \\ & Background\tablenotemark{b} & 3724 & 463669 \\ & & & \\ Total Signal & & 4334 & 466378 \\ \\ Noise\\ & Signal Shot Noise & 24.7 & 52.0 \\ & Background Shot Noise & 60.8 & 680.9 \\ & Dark Shot Noise & 5.5 & 10.0 \\ & Read Noise & 5 & 5 \\ Total Noise & & 66.0 & 683.0 \\ \enddata \label{readoutexamples} \tablenotetext{a}{In electrons} \tablenotetext{b} {Background includes zodiacal foreground and thermal emission from the telescope as calculated as described in \S \ref{snrcal}.} \end{deluxetable} Similarly, orders of magnitude higher thermal emission from the instrument, or thermal radiation leaks from outside the instrument bay, can be tolerated compared to the case for dispersive spectrometers or fixed filter cameras. As a pragmatic demonstration of this principle, the IFTS instruments LIFTIRS and HIRIS are routinely operated with ambient temperature optics in the 8-14 $\mu$m band (Bennett et al. 1995), whereas dispersive spectrometers, like SEBASS (Bennett, {\it private communication}), operating in the same spectral region must have the slit and all following optics cooled far below ambient temperatures. The reason is that in a dispersive spectrometer the thermal emission of all the elements and optics downstream of the slit reach the detector at full spectral range determined by the bandpass limiting element at or near the coldstop, whereas only the narrow spectral range corresponding to the width of a spectral channel for the signals of interest reach the detector pixels. For the IFTS, both the signals of interest and the thermal emission are seen over the full spectral range determined by the bandpass limiting filter, and thus it is only necessary that the thermal emission of the optical elements along the optic axis integrated over the bandpass of interest, either 1-5 $\mu$m or 5-15 $\mu$m, be somewhat less than that of the integral of the zodiacal foreground, telescope emission, and source signal level integrated over the same broad spectral region. An IFTS is potentially immune to cosmic ray hits because the ``energy'' of a single upset pixel in one OPD frame appears as a sinusoidal signal divided among all bins in the spectral transform of the interferogram for that pixel. We can ignore cosmic ray hits only if the counts generated are at or below our noise level. A minimum ionizing cosmic ray proton ($E \simeq 1$ GeV) has ionization losses of $dE/dx \simeq 400$ eV $\mu$m$^{-1}$ in Si. Assuming that 3.6 eV is required to produce an electron-hole pair, a cosmic ray will yield at least a few thousand events, since typical pixels have sensitive layers that are tens of $\mu$m thick. We would obtain similar number for a hybrid device, i.e., InSb or HgCdTe on a Si multiplexer. If a cosmic ray hit produces a significant signal in a certain number of pixels, those pixels must be ``repaired'' by interpolating the interferogram between the previous few ``good'' frames, and the following few ``good'' frames which are not contaminated by cosmic ray hits. The same sort of processing would be needed for any other system as well, be it an imager or a spectrometer. Comparison with the noise sources listed in Table \ref{readoutexamples} indicates that cosmic ray hits will have to be repaired in the NIR channel, while the MIR channel will be more tolerant. A dual port design (Fig. \ref{optical-layout}) delivers the complementary symmetric and antisymmetric interferograms. In this dual-input dual-output design, the field of the complementary input (labeled ``Calibration Input'' in Fig. \ref{optical-layout}) is also imaged and superimposed on each image of the ``Primary Input''. This property is often used to cancel the sky emission. In operation, when observing the sky in the primary input, the secondary input would be fed with a cold blackbody load, having negligible radiance. The final interferogram is constructed from the difference between the two outputs (which is therefore also immune to common mode electrical noise) while the normalized ratio reveals systematic variation due to detector drifts. The wavelength scale and the instrumental line shape (a $sinc$ function if there is no apodizing) are precisely determined and are independent of wavelength. Absolute wavelength calibration is done by counting fringes of an optical single-mode laser. Compared to a dispersive system the broad-band operation of an IFTS means that there are $M$ times more photons for flat fielding and determining signal-dependent gain (linearity). Hence, high signal-to-noise calibration images can be acquired faster or with lower power internal sources. \subsection{Pixel Size, Spectral Resolution, and Field of View} \label{pixelsize} Spatial multiplexing renders the performance of an IFTS equal to that of an ideal multi-slit spectrograph (Bennett 1995). Hence, even if we ignore slit losses and blaze inefficiency the other advantages of an IFTS are overwhelming. The spectral resolution can be varied arbitrarily from the coarsest case of a small number of bands up to a spectral resolution limit determined only by the maximum OPD characteristic of the instrument. The proposed instrument has a maximum OPD of 1 cm and hence can operate over a range of resolutions from full band up to $M$=8000 in the NIR The spectral resolution limit, $R = k/\delta k$, of a Michelson interferometer is \begin{equation} R = 8(d/ \phi D)^2, \end{equation} \noindent where $\phi$ is the angular diameter of the FOV, $d$ is the diameter of the beam splitter, and $D$ the telescope primary mirror diameter (e.g., Jacquinot 1954; Maillard 1995). Classically, $\phi$ refers to the entire field, but in the case of an IFTS, $\phi$ is the FOV of an on-axis pixel. Although convenient if a single fringe fills the FPA, just as with imaging Fabry-Perots, there is no reason why each pixel should record the same apparent wavenumber. Fringes crowd together with increasing field angle. Therefore, the need to maintain modulation efficiency over the entire field of view requires that the spatial separation of the fringes at the edge of the FPA, for a given retardance, is significantly greater than the pixel spacing. If $x$ is the OPD for a normally incident beam with wavenumber $k$, and $\theta$ is the field angle of off-axis rays at the beam splitter, then the path difference at $\theta$ is $x_\theta = x cos\theta$ and the apparent wavenumber of this beam is \begin{equation} k_\theta = k/cos\theta. \end{equation} \noindent The angles $\theta$ and $\phi$ are related by the angular magnification, $D/d$. If $\delta \theta$ is the angular width, also at the beam splitter, corresponding to a single pixel, the spectral resolution limit for off-axis points can be found by differentiating Eq (2), \begin{equation} 1/R_\theta = \delta k_\theta/k_\theta = tan\theta \delta \theta, \end{equation} \noindent Fig. \ref{pixel_fov} shows the pixel size for a given field of view for a range of resolutions. For example, for an 8~m diameter primary aperture and a beam splitter of diameter 10~cm, a FOV of $200''$, and a pixel size of $0.''05$, the corresponding angles at a beam splitter of diameter 10~cm are $2.^\circ 2$ and $4''$ respectively, leading to a resolution limit of $R = 1.3 \times 10^6$. Since this resolution is two orders of magnitude greater than we are proposing, it is clear that spectral resolution is not the principal factor determining pixel size. An alternative way to view this constraint is that $d$, i.e., the size of the optics, is determined not by spectral resolution, but by the requirement that there be no vignetting over the field of view. Thus, the optics for an IFTS are similar to that of a simple re-imaging camera, and are smaller and slower than those of an equivalent dispersive spectrograph. \begin{figure}[thb] \centerline{\psfig{figure=fig4.ps,width=3.4in,angle=90}} \caption{\small Pixel size as a function of the field of view required to spatially fully sample fringes at the edge of the FPA, and hence maintain modulation efficiency. Curves are plotted for resolutions $R = k/\delta k = 10^4$, $10^5$ and $10^6$ assuming an 8~m diameter primary aperture and a beam splitter of diameter 10~cm. } \label{pixel_fov} \end{figure} We therefore have broad freedom to choose the pixel size by trading off field of view and spatial sampling. Given that NIR arrays of $4096 \times 4096 $ pixels are likely to be available in the near future, a pixel size of $0.''05$ yields a $3.'3$ field of view and $\lambda / 2D$ sampling at 4 $\mu$m. This choice of pixel size does not preclude diffraction limited imaging at shorter wavelengths. If pixels have sharp boundaries, then it is possible to extract information at spatial frequencies above the cut-off in the pixel-sampling modulation transfer function if the spacecraft can offset and track at the sub-pixel level (cf. Fruchter and Hook 1998). Similar reasoning suggests $0.''1$ pixels would be a satisfactory compromise for the MIR channel. | An IFTS instrument can perform a wide variety of NGST science. The advantages of the IFTS concept are: \begin{enumerate} \item Deep imaging acquired simultaneously with higher spectral resolution data over a broad wavelength range. \item ``Hands-off'', unbiased, multi-object, slitless spectroscopy (ideal for moving objects). Efficient in confusion limit. \item Flexible resolution ($M=1-10,000$). \item High throughput (near 100\%) dual-port design. \item Tolerant of cosmic rays, read-noise, dark current, and light leaks. \item Simple and reliable calibration. High SNR determination of flat-fields and detector non-linearity. \item Compact, lightweight design. Slow reimaging optics. \end{enumerate} | 98 | 3 | astro-ph9803163_arXiv.txt |
9803 | astro-ph9803119_arXiv.txt | We present a study of 581 Hz oscillations observed during a thermonuclear X-ray burst from the low mass X-ray binary (LMXB) 4U 1636-54 with the Rossi X-ray Timing Explorer (RXTE). This is the first X-ray burst to exhibit both millisecond oscillations during the rising phase as well as photospheric radius expansion. We measure an oscillation amplitude within 0.1 s of the onset of this burst of $75 \pm 17 \%$, that is, almost the entire thermal burst flux is modulated near onset. The spectral evolution during the rising phase of this burst suggests that the X-ray emitting area on the neutron star was increasing, similar to the behavior of bursts from 4U 1728-34 with 363 Hz oscillations reported recently. We argue that the combination of large pulsed amplitudes near burst onset and the spectral evidence for localized emission during the rise strongly supports rotational modulation as the mechanism for the oscillations. We discuss how theoretical interpretation of spin modulation amplitudes, pulse profiles and pulse phase spectroscopy can provide constraints on the masses and radii of neutron stars. We also discuss the implications of these findings for the beat frequency models of kHz X-ray variability in LMXB. | Large amplitude millisecond oscillations have now been observed during thermonuclear X-ray bursts from six low mass X-ray binary (LMXB) systems with the Rossi X-ray Timing Explorer (RXTE) (see \markcite{SZS}Strohmayer, Zhang \& Swank 1997; \markcite{SMB}Smith, Morgan \& Bradt 1997, \markcite{Z96}Zhang {\it et al.} 1996; \markcite{Swank97}Swank {\it et al.} 1997; and \markcite{Stroh97}Strohmayer {\it et al.} 1997). The thermonuclear instability which triggers an X-ray burst burns in a few seconds the nuclear fuel which has been accumulated on the neutron star surface over several hours. This $>$ 10$^3$ difference between the accumulation and burning timescales means that it is extremely unlikely that the conditions required to trigger the instability will be achieved simultaneously over the entire stellar surface. This realization, first emphasized by \markcite{Joss78}Joss (1978), led to the study of lateral propagation of the burning instability over the neutron star surface (see \markcite{FW82}Fryxell \& Woosley 1982, \markcite{NIF84}Nozakura, Ikeuchi \& Fujimoto 1984, and \markcite{B95}Bildsten 1995). The subsecond risetimes of thermonuclear X-ray bursts suggests that convection plays an important role in the physics of the burning front propagation, especially in the low accretion rate regime which leads to large ignition columns (see \markcite{B98}Bildsten (1998) for a review of thermonuclear burning on neutron stars). \markcite{B95}Bildsten (1995) has shown that pure helium burning on neutron star surfaces is in general inhomogeneous, displaying a range of behavior which depends on the local accretion rate. Low accretion rates lead to convectively combustible accretion columns and standard type I bursts, while high accretion rates lead to slower, nonconvective propagation which may be manifested in hour long flares. These studies emphasize that the physics of thermonuclear burning is necessarily a multi-dimensional problem and that {\it localized} burning is to be expected, especially at the onset of bursts. There is now good evidence that the oscillations seen during the rising phase of bursts from 4U 1728-34 are produced by spin modulation of such a localized thermonuclear hotspot on the surface of the neutron star, and that the observed oscillation frequency is a direct measure of the neutron star spin frequency (see \markcite{SZS}Strohmayer, Zhang \& Swank 1997). These observations provide the most compelling evidence to date that neutron stars in LMXB are rotating with near millisecond periods. In this Letter we present new burst data from the LMXB 4U 1636-54 which provides further evidence in support of the spin modulation hypothesis for the millisecond burst oscillations. We present data from a thermonuclear burst from 4U 1636-54 which reveals a strong transient oscillation during the burst rise at 1.723 ms with an initial amplitude of $75 \pm 17 \%$. We also discuss the implications of these findings for the spin modulation interpretation and how they can be used to place constraints on the mass and radius of neutron stars. Finally, we discuss some implications of our observations for the current theories of kilohertz quasiperiodic oscillations (QPO) in LMXB. | If the millisecond oscillations seen during bursts are in fact due to spin modulation, then detailed study and modelling of the oscillation amplitudes, pulse profiles and spectral variability with pulse phase during X-ray bursts can provide a wealth of information on the mass and radius of the neutron star. For example, the maximum modulation amplitude that can be obtained from a hotspot of a given angular size on a rotating neutron star is set by the strength of general relativistic light bending. For the case that rotation is not rapid enough to substantially distort the exterior spacetime from the Schwarzchild spacetime, and this is the case even for spin periods of a few milliseconds (Lamb \& Miller 1995), then the maximum amplitude depends only on the compactness of the neutron star, that is, the ratio of stellar mass to radius $GM/c^2R$. Stars which are more compact produce lower amplitudes due to flattening of the pulse by light bending (see \markcite{PFC}Pechenick, Ftaclas \& Cohen 1983; \markcite{S92}Strohmayer 1992; and \markcite{ML97}Miller \& Lamb 1997). Since the intrinsic rotational modulation amplitude can only be decreased by other effects such as photon scattering (see \markcite{MLP}Miller, Lamb \& Psaltis 1997; \markcite{BL}Brainerd \& Lamb 1987; and \markcite{KP}Kylafis \& Phinney 1989) or the viewing geometry of the spot, the maximum observed oscillation amplitude represents a lower limit to the intrinsic amplitude. Thus an observed amplitude can be used to place an upper limit on the compactness of the neutron star, that is, if the star were more compact than some limit it would not be able to produce a modulation amplitude as large as that observed. In principle, stellar rotation will also play a role in the observed properties of spin modulation pulsations. For example, assuming the oscillation frequency of 581 Hz represents the spin frequency of the neutron star in 4U 1636-54, then for a 10 km radius neutron star the spin velocity is $v_{spin}/c = 2\pi \nu_{spin} R \approx 0.12$ at the rotational equator. The motion of the hotspot produces a Doppler shift of magnitude $\Delta E / E \approx v_{spin}/c = 0.12$, thus the observed spectrum is a function of pulse phase (see \markcite{CS}Chen \& Shaham 1989). Measurement of a pulse phase dependent Doppler shift in the X-ray spectrum would provide additional evidence supporting the spin modulation model and would also provide a means of constraining the neutron star radius. The rotationally induced velocity also produces an aberration which results in asymmetric pulses, thus the pulse shapes also contain information on the spin velocity and therefore the stellar radius (\markcite{CS}Chen \& Shaham 1989). The component of the spin velocity along the line of site is proportional to $\cos\theta$, where $\theta$ is the lattitude of the hotspot measured with respect to the rotational equator. The modulation amplitude also depends on the lattitude of the hotspot, as spots near the rotational poles produce smaller amplitudes than those at the equator. Thus we expect a correlation between the observed oscillation amplitude and the size of any pulse phase dependent Doppler shift. Dectection of such a correlation in a sample of bursts would definitively confirm the rotational modulation model in our opinion. We will present calculations of mass - radius constraints for 4U 1636-54 and 4U 1728-34 including the effects of light bending, rotation and angle dependent emission, based on the observed properties of burst oscillations as well as spectroscopy of Eddington limited bursts in a subsequent paper. Several models for the kilohertz QPO seen in 13 LMXB systems (see \markcite{vdk}van der Klis 1997 for a recent review) invoke some sort of beat-frequency interpretation for the twin kHz peaks seen in many of the sources (see \markcite{MLP}Miller, Lamb \& Psaltis 1997; \markcite{S96}Strohmayer {\it et al.} 1996). So far only in 4U 1728-34 does the frequency difference between the twin kHz peaks match the frequency observed during X-ray bursts (see \markcite{S96}Strohmayer {\it et al.} 1996). In two other sources (KS 1731-26 and 4U 1636-54) the separation of the twin kHz peaks is closer to 1/2 the frequency of oscillations observed during bursts (see \markcite{WV}Wijnands \& van der Klis 1997; and \markcite{Z96}Zhang {\it et al.} 1996). For example, \markcite{W97}Wijnands {\it et al.} (1997) report a frequency difference for the twin kHz QPO in 4U1636-54 of $276 \pm 10$ Hz. The effort to reconcile these observations with a beat-frequency interpretation has led to speculation that the oscillation frequency observed during bursts may sometimes be twice the spin frequency of the star, although in 4U1636-54 the difference frequency appears to be a bit less than 1/2 the burst oscillation frequency, and in some Z sources the frequency difference is not constant (\markcite{van97}van der Klis {\it et al.} 1997). If this scenario is correct it implies the existence of two antipodal spots on the neutron star surface during X-ray bursts. In addition to the daunting requirement of initiating the thermonuclear flash nearly simultaneously on opposite sides of the neutron star, the observation of large oscillation amplitudes shortly after burst onset in 4U1636-54 places severe constraints on the two hotspot scenario. To see this one can ask the following question. What maximum amplitude can be produced by a star with antipodal hotspots? For two antipodal spots light bending strongly constrains the amplitudes that can be achieved (see \markcite{PFC}Pechenick, Ftaclas \& Cohen 1983). Calculations using the Schwarzchild spacetime and isotropic emission from the stellar surface indicate that even a neutron star with an implausibly small compactness of $M/R = 0.1$, recall that rotational modulation amplitude increases with decreasing compactness, can only achieve a maximum amplitude of about 30 \%, whereas we measured an amplitude of 75\% from the burst described here. We note that with $M/R = 0.1$ a 1.4 $M_{sun}$ neutron star would have a radius of 21 km. This is far stiffer than any neutron star equation of state that we are familiar with. Rapid rotation could in principle modify this result, but at the modest inferred spin period of 290.5 Hz for the two spots to produce the observed 581 Hz frequency one would still require implausibly large neutron star radii to make a significant rotational correction to the amplitude. Another process which can increase the amplitude is beaming of radiation at the stellar surface. For example, to the extent that electron scattering is the dominant opacity process in neutron star atmospheres then one should expect a specific intensity distribution which approximates that from a grey atmosphere. Such a distribution is proportional to $\cos\delta + 2/3$, where $\delta$ is the angle from the normal to the stellar surface (see \markcite{Mihal}Mihalas 1978), so this will modestly increase the rotational modulation amplitude. In addition, it is likely that the emergent spectrum will also be a weak function of $\delta$ (\markcite{ML97}Miller \& Lamb 1997). With more detailed modelling of the emission from hotspots of finite size it will be possible to place strong constraints on the antipodal hotspot interpretation for the burst oscillations. Finally, if further analysis such as pulse phase spectroscopy continues to support the interpretation of the burst oscillation frequency as the neutron star spin frequency in 4U1636-54, then kHz QPO frequency separations near 1/2 the spin frequency would suggest that the beat frequency interpretation may be untenable. | 98 | 3 | astro-ph9803119_arXiv.txt |
9803 | astro-ph9803096_arXiv.txt | { To examine the evidence for hierarchical evolution on mass scales of $\sim 10^{13}$-$10^{14} \Mdot$, we apply a statistic that measures correlations between galaxy velocity and projected position (Dressler \& Shectman 1988) to data for six poor groups of galaxies, HCG 42, HCG 62, NGC 533, NGC 2563, NGC 5129, and NGC 741. Each group has more than 30 identified members (Zabludoff \& Mulchaey 1998ab). The statistic is sensitive to clumps of galaxies on the sky whose mean velocity and velocity dispersion deviate from the kinematics of the group as a whole. The kinematics of galaxies within $\sim 0.1$\lith\inv\ Mpc of the group center do not deviate from the global values, supporting our earlier claim that the group cores are close to virialization or virialized. We detect significant substructure (at $\geq 99.9$\% confidence) in the two groups with the most confirmed members, HCG 62 and NGC 741, that is attributable mostly to a subgroup lying $\sim 0.3$-0.4\lith\inv\ Mpc outside of the core. We conclude that at least some poor groups, like rich clusters, are evolving via the accretion of smaller structures from the field. With larger poor group surveys, the incidence of such accretion and the distribution of subgroup masses are potential constraints of cosmological models on mass scales of $\simless 10^{13}$-$10^{14} \Mdot$ and on physical scales of $\simless 0.5$\lith\inv\ Mpc. \vskip 0.5cm \noindent{\it Subject headings}: galaxies: clustering --- cosmology: large-scale structure of Universe } \vfill\eject | The evolution of structure on different mass scales is one of the outstanding issues in cosmology. For example, although galaxy clusters of $\sim 10^{15} \Mdot$ (including Virgo (Binggeli 1993; Bohringer \etal 1994), Coma (Mellier \etal 1988; Briel \etal 1992; White \etal 1993), and Abell 754 (Zabludoff \& Zaritsky 1995)) are clearly evolving from the accretion of smaller groups, it is uncertain whether poor groups of $\sim 10^{13}$-$10^{14} \Mdot$ also evolve hierarchically. There is some indirect evidence that the evolution of poor groups is similar to rich clusters; the galaxies and hot gas in poor groups follow the same relationships found among the X-ray temperature, X-ray luminosity, and galaxy velocity dispersion for rich clusters (Mulchaey \& Zabludoff 1998, hereafter MZ98). Historically, however, the number of known poor group members has been too small to examine individual groups for direct evidence of hierarchical evolution. Multi-object spectroscopy now makes it possible to obtain ``cluster-size'' samples of galaxies in poor groups and to identify substructure, if it exists, in the same manner as for rich clusters. Substructure in clusters was not detected until the number of spectroscopically-confirmed cluster members exceeded $\sim 30$-50 galaxies. Recent poor group surveys have reached this membership level (Zabludoff \& Mulchaey 1998ab; hereafter ZM98a and b). The discovery of substructure in poor groups would provide new cosmological constraints by establishing that hierarchical evolution is occurring on mass scales of $\sim 10^{13}$-$10^{14} \Mdot$ and on physical scales of $\sim 0.5$\lith\inv\ Mpc. The detection of substructure would also support the picture in which at least some poor groups evolve as low-mass analogs to rich clusters. In this Letter, we describe the results from applying a substructure statistic (Dressler \& Shectman 1988; hereafter DS88) to the six best sampled poor groups in ZM98ab, the most detailed spectroscopic survey of poor groups to date. | To search for substructure in the six poor groups, we use the method applied by DS88 to rich clusters. This test identifies a fixed number of nearest neighbors on the sky around each galaxy, calculates the local mean velocity and velocity dispersion of the subsample, and compares these values with the mean velocity and velocity dispersion of the entire group. The kinematic deviations of the subsamples from the global values are summed. This sum is larger for a group with a kinematically distinct subgroup than for a similar group without substructure. For each galaxy $i$, the deviation of its nearest projected neighbors from the kinematics of the group as a whole is defined as ${\delta_i} \equiv (n^{1/2}/ \sigma_r) \lbrack (\upsilon_{loc} - \overline \upsilon)^2 + (\sigma_{r,loc} - \sigma_r)^2 \rbrack ^{1/2}$, where $\overline \upsilon$ is the mean velocity for the group, $\sigma_r$ is the group velocity dispersion, and $n$ is the number of nearest neighbors (including the galaxy) used to determine the local mean velocity $\overline \upsilon_{loc}$ and local velocity dispersion $\sigma_{r,loc}$. The total deviation for the group is defined as the sum of the local deviations, $\Delta \equiv \sum |\delta_i|$ for all $i \leq N_{grp}$, the number of group members. As pointed out in DS88, the $\Delta$ statistic is similar to the $\chi^2$ statistic, except that the $\delta_i$'s are not squared before summation in order to reduce the contributions of the largest, rarest deviations. If the galaxy velocity distribution of the group is close to Gaussian, and the local variations are only random fluctuations, $\Delta \simeq N_{grp}$. To calculate $\delta$, we choose $n = 11$ (as in DS88). This choice allows robust determinations of $\overline \upsilon_{r,loc}$ and $\sigma_{r,loc}$. Silverman (1986) argues that using $n \sim {N_{grp}}^{1/2}$ nearest neighbors (= 6-8 for these groups) maximizes the sensitivity of such a test to small scale structures while reducing its sensitivity to fluctuations within the Poisson noise (also see Bird 1994b). To check the robustness of the $n=11$ assumption, we compare the results below with those for $n=6$. The conclusions drawn from the $n=6$ and $n=11$ cases are the same. Calibration of the $\Delta$ statistic for each group is required because 1) the $\delta_i$'s are not statstically independent and 2) the velocity distribution may not be intrisically Gaussian even if there are no subgroups ({\it e.g.}, the group members may follow predominantly circular or radial orbits). We determine the significance of the observed $\Delta$ by comparing it with the results of 1000 Monte Carlo trials in which galaxy velocities are drawn randomly from the observed distribution and assigned to galaxy positions. This scrambling technique effectively destroys any substructure (DS88). If the probability is low that a group without substructure has a $\Delta$ value at least as large as that observed, then we consider the substructure detection significant. In two groups, HCG 62 and NGC 741, the observed value of $\Delta$ is significant at the $\geq 99.9\%$ confidence level. Such high $\Delta$ values might arise from substructure, but also could result from smooth variations in the group's velocity field ({\it e.g.}, rotation or velocity shear (Malumuth \etal 1992)) and/or from a dependence of $\sigma_r$ on radius (Bird 1994b). ZM98a show that $\sigma_r$ is constant out to radii of $\sim 0.5$\lith\inv\ Mpc in the combined velocity dispersion profile for the sample groups. To determine whether substructure is in fact responsible for the high $\Delta$ values in HCG 62 and NGC 741, we examine the local deviation $\delta$ for each group member. A concentration of large $\delta$ values on the sky indicates a kinematically distinct subgroup. Figure 1 shows the projected spatial distribution of group members for each group (top panel). The second panel shows this distribution with the radii of the circles weighted by $e^\delta$ (as in DS88). Because each point is not statistically independent, a few very deviant galaxies can boost the $\delta$ values for a large number of nearby points. Therefore, a visual comparison with the Monte Carlo simulations is required to assess the significance of any structures. The third panel shows the results of the Monte Carlo trial (out of 1000) with the largest $\Delta$ value, or greatest total deviation. The results of the trial with the median $\Delta$ value are in the bottom panel. The significance of the seven large clustered circles to the northeast of HCG 62 and five large clustered circles to the south of NGC 741 is high. In each case, the large $\delta$ values show that a subgroup not in equilibrium with the global group potential is the principal source of the significant $\Delta$ value. Each of the two subgroups lies a projected distance of $\sim 0.3$-0.4\lith\inv\ Mpc outside of the group core. On the other hand, the clustering of small circles within $\sim 0.1$\lith\inv\ Mpc of all of the group centers indicates that the core mean velocity and velocity dispersion are similar to the global values for the group. This result suggests that the group cores are close to virialization or virialized and is consistent with the conclusions from our earlier studies of group dynamics (ZM98a, MZ98). | 98 | 3 | astro-ph9803096_arXiv.txt |
9803 | astro-ph9803089_arXiv.txt | Aperiodic variability and Quasi Periodic Oscillations (QPOs) are observed from accretion disks orbiting white dwarfs, neutron stars, and black holes, suggesting that the flow is universally broken up into discrete blobs. We consider the interaction of these blobs with the magnetic field of a compact, accreting star, where diamagnetic blobs suffer a drag. We show that when the magnetic moment is not aligned with the spin axis, the resulting force is pulsed, and this can lead to resonance with the oscillation of the blobs around the equatorial plane; a resonance condition where energy is effectively pumped into non--equatorial motions is then derived. We show that the same resonance condition applies for the quadrupolar component of the magnetic field. We discuss the conditions of applicability of this result, showing that they are quite wide. We also show that realistic complications, such as chaotic magnetic fields, buoyancy, radiation pressure, evaporation, Kelvin--Helmholtz instability, and shear stresses due to differential rotation do not affect our results. In accreting neutron stars with millisecond periods, we show that this instability leads to Lense--Thirring precession of the blobs, and that damping by viscosity can be neglected. | The interaction between the magnetic field anchored to rotating bodies and matter orbiting around them plays an important role in a variety of astrophysical situations, ranging from the Jupiter-Io system (Stern and Ness 1982) to the Jovian ring (Burns \etal\/ 1985), to protoplanetary disks around young magnetic stars (Bodenheimer 1995) and accreting degenerate stellar remnants in binary systems, such as white dwarfs in intermediate polars or neutron stars in X-ray binaries (Frank, King and Raine 1995). Straightforward applications of the laws of electrodynamics are possible in cases, such as the Jovian ring, in which the orbiting matter is made of solid, dust particles. Due to a variety of poorly known MHD and plasma effects, modelling this interaction in the case of gaseous disks that orbit different classes of magnetic stars is far more uncertain and difficult. In particular, theoretical descriptions of viscous accretion disks around magnetic rotating stars have been largely based on a number of simplifying assumptions. Two of these are especially relevant to the present work: (a) that the magnetic field (assumed dipolar) and rotation axes of the star are coaligned; (b) that the disk is continuous, smooth, azimuthally symmetric and lies in the equatorial plane of the rotating star. For example Ghosh and Lamb (1979) adopted these assumptions and introduced an {\it ad hoc} effective diffusivity of the disk plasma to derive a stationary disk solution, in which the star magnetic field lines thread the disk and slip across its material. Models of this kind proved very useful for evaluating the basic properties of the disk-magnetic field interaction, such as the size of the magnetosphere and the balance of the material and non-material torques. It has always been recognised that the coalignement of the magnetic field and rotation axes of the star is an unrealistic hypothesis as a finite magnetic colatitude is required for the generation of the periodic signals at the star spin frequency that are often observed in these systems. On the other hand, the pronounced aperiodic (or, in some cases, quasi-periodic) variability that is frequently detected on a variety of timescales ranging from days to milliseconds in disk accreting compact stars of all classes (intermediate polars: King and Lasota 1990; accreting neutron stars and black holes: van der Klis 1995, van der Klis 1997) testify that the disk cannot be regarded as smooth, continuous and azimuthally symmetric. In a number of cases the quasi-periodic timing signature of blobs orbiting over a limited range of radii close to the compact star is clearly seen. A highly inhomogeneous and clumpy disk, possibly comprising two distinct and coexisting phases (hot and cold), is also envisaged as the end product of the instabilities predicted by current models of viscous disks (e.g the so-called secular and viscous instabilities of radiation-pressure dominated $\alpha$-disks, Lightman and Eardley 1974, Shakura and Sunyaev 1976, or the magnetic amplification and buoyancy of flux tubes in dynamo-driven disks, Vishniac and Diamond 1992), as discussed by Krolik (1998). In all accreting objects endowed with a strong magnetic field, the presence of inhomogeneities and/or blobs that can be regarded as discrete entities suggests that a novel mode of interaction between the compact star and the disk is possible. This occurs because individual blobs are most likely strongly diamagnetic (see for instance King 1993, Wynn and King 1995 and references therein); when moving through a magnetic field, strong surface currents develop on the blob the main effect of which is the generation of a drag opposite to the component of the blob velocity perpendicular to the field (Drell, Foley and Ruderman 1965). The acceleration acting on each blob is thus \begin{equation} \vec{a} = -\frac{\vec{v}_\perp^{(rel)}}{t_d}\;; \end{equation} the drag time--scale $t_d$ is given by \begin{equation} t_d = \frac{c_A m}{B^2 l^2}\;, \end{equation} where $c_A$ is the Alfv\`en speed in the magnetic field $B$, and $m$ and $l$ are the mass and characteristic radius of the blob. Here $\vec{v}^{(rel)}$ is the relative velocity between the blob and a magnetic field line. It is easy to see (Fig.1) that this acceleration has a component along the disk axis which tends to lift the blob off the equatorial plane where it is, at least initially, lying. It is the purpose of this paper to study this dynamical interaction, and to show how this alters the conventional view of accretion disks onto magnetized compact stars. In order to bring out the physical meaning of the instability most clearly, we shall at first idealize these plasma blobs as point masses. In the next section, it will be shown that this interaction leads to the lifting of blobs off the equatorial plane, at a resonance radius; the conditions under which this result applies are discussed in Section 3. In Section 4, we relax the hypothesis that blobs are point masses, and establish that a variety of effects, all related to the blobs having a finite size, do not modify our results. As the only concrete application of this instability, we discuss in Section 5 the generation of modulation of the X--ray flux in Low Mass X--ray Binaries (LMXBs) exhibiting millisecond QPOs at frequencies comparable to the single--particle Lense--Thirring (1918) precession frequency. The last section summarizes the results. | The processes described above are generic: they apply to every accretion disk broken up into discrete units, surrounding a magnetized non--aligned rotator provided the ordering of time--scales (Eq. \ref{ordering}) holds and the blobs have density contrasts $(\rho_b-\rho_d)/ \rho_d \ga 1$. The instability is independent of the detailed diamagnetic properties of the blobs (here enshrined in the parameter $t_d$ which is just modulated at the relative frequency $\omega_K-\omega_s$), of their masses and densities, their interactions with radiation emitted by the accreting source, and the exact form of the viscosity. Provided the hierarchy of timescales $t_K \ll t_d \ll t_v$ (Eq. \ref{ordering}) is established and blobs have non--negligible density contrasts, it should apply to both accreting white dwarfs and neutron stars. The existence of resonances is essentially due to the fact that a spatially periodic magnetic field, such as that due to a single multipole, will produce a temporally--variable electromagnetic force on a single blob, as it passes through the various frequency components of the field. Minor corrections to the exact locations of the resonances might derive from effects we opted not to consider, such as an axially offset field, or a rotation rate different for different multipoles, as it happens to the non--dipolar part of the magnetic field of the Earth. Another phenomenon we neglected to investigate, and which might generate further resonances, is the non--linear coupling of radial, latitudinal and longitudinal motions. It should however be borne in mind that the exact locations of resonances might be irrelevant since blobs are expected to have a finite size. The evolution of the accretion disk after the resonance of Eq. \ref{resonance} is very difficult to predict; it seems clear enough that blobs will tend to move closer to the magnetic equatorial plane, and oscillate around it, but the further evolution is uncertain because the details depend on a very poorly known quantity, the $z$--axis viscosity. This is the reason why we did not push the analysis performed in this paper into the non--linear regime. It seems quite clear that the disk must puff--up, but whether, and when, viscosity will manage to bring it back to the equatorial plane remains an open question. Very interesting consequences of the above--discussed instability occur in a specific class of accretors, \ie\/ neutron stars exhibiting millisecond QPOs. Blobs that have acquired $z$~axis motions will be acted upon by the torque which causes Lense--Thirring (1918, LT) precession. We have shown here that, in both Atoll and Z--type sources, viscosity is surely unable to bring blobs back down to the equatorial plane. For motions in the plane, the net effect of the instability is to induce epicyclic motions, or, given that in $1/r$ potentials all orbits close, to perturb the blobs' motions into ellipses. Then periastron precession induced by general relativistic effects ought to ensue. A related mechanism for the generation of off--plane motions has been proposed for the Jovian dust ring (Burns \etal\/ 1985). It differs from the present one in that the individual units are not blobs of plasma, but single dust grains with nonzero electric charge, which are then acted upon by the normal Lorenz force $q \vec{v}^{(rel)}\wedge\vec{B}/c$. It is easy to see that this differs from our case because the forces differ: in our case, the acceleration (Eq. 1) is in the plane of the two vectors $\vec{v}^{(rel)}$ and $\vec{B}$, while in the Jovian case the force is perpendicular to this plane. This difference has as a consequence that, in the Jovian case, each magnetic field multipole has its own, single resonance, while in our case the two multipoles we explored have the same array of resonances, and the array is an infinite one. In short, in this paper we have shown that a simple interaction lifts blobs off the equatorial plane of a viscous accretion disk, provided the accreting object is not an aligned rotator. This occurs in a thin radial annulus, identified by Eq. \ref{resonance}. This conclusion is rather general, being independent of the exact form of the drag time--scale, of the blobs' diamagnetic properties, and also of the assumed form for the disk viscosity. We have also shown why viscosity is unable to damp these motions (Section \ref{lt}), and that this leads to modulation of the X--ray flux at the single particle precession frequency, or possibly twice as much. Thus, the proposed identification of some QPOs in the spectra of both Atoll and Z sources as due to LT--precession (Stella and Vietri 1998) is made more likely by the existence of a relevant mechanism exciting blobs' motions off the equatorial plane. Helpful comments from F.K. Lamb and J. Imamura are gratefully acknowledged. This work was begun while one of us (M.V.) was visiting the Institute for Advanced Study; its Director, John Bahcall, is thanked for the warm hospitality. The work of L.S. was partially supported through ASI grants. | 98 | 3 | astro-ph9803089_arXiv.txt |
9803 | astro-ph9803040_arXiv.txt | With the new generation of instruments for Cosmic Microwave Background (CMB) observations aiming at an accuracy level of a few percent in the measurement of the angular power spectrum of the anisotropies, the study of the contributions due to secondary effects has gained impetus. Furthermore, a reinvestigation of the main secondary effects is crucial in order to predict and quantify their effects on the CMB and the errors that they induce in the measurements. \par In this paper, we investigate the contribution, to the CMB, of secondary anisotropies induced by the transverse motions of clusters of galaxies. This effect is similar to the Kaiser--Stebbins effect. In order to address this problem, we model the gravitational potential well of an individual structure using the Navarro, Frenk \& White profile. We generalise the effect of one structure to a population of objects predicted using the Press-Schechter formalism. We simulate maps of these secondary fluctuations, compute the angular power spectrum and derive the average contributions for three cosmological models. We then investigate a simple method to separate this new contribution from the primary anisotropies and from the main secondary effect, the Sunyaev-Zel'dovich kinetic effect from the lensing clusters. \par | During the next decade, several experiments are planned to observe the Cosmic Microwave Background (CMB) and measure its temperature fluctuations (Planck surveyor, Map, Boomerang, ...). Their challenge is to measure the small scales anisotropies of the CMB (a few arcminutes up to ten degrees scale) with sensitivities better by a factor 10 than the COBE satellite (Smoot et al. 1992). These high sensitivity and resolution measurements will tightly constrain the value of the main cosmological parameters (Kamionkowski et al. 1994). However, the constraints can only be set if we are able to effectively measure the {\it primary} temperature fluctuations. These fluctuations, present at recombination, give an insight into the early universe since they are directly related to the initial density perturbations which are the progenitors to the cosmic structures (galaxies and galaxies clusters) in the present universe; but which are first and foremost the relics of the very early initial conditions of the universe.\\ Between recombination and the present time, the CMB photons could have undergone various interactions with the matter and structures present along their lines of sight. Some of these interactions can induce additional temperature fluctuations called, {\it secondary} anisotropies because they are generated after the recombination. Along a line of sight, one measures temperature fluctuations which are the superposition of the {\it primary } and {\it secondary} anisotropies. As a result, and in the context of the future CMB experiments, accurate analysis of the data will be needed in order to account for the foreground contributions due to the secondary fluctuations. Photon--matter interactions between recombination and the present time are due to the presence of ionised matter or to variations of the gravitational potential wells along the lines of sight. \par The CMB photons interact with the ionised matter mainly through Compton interactions. In fact, after recombination the universe could have been re-ionised globally or locally. Global early re-ionisation has been widely studied (see Dodelson \& Jubas 1995 for a recent review and references therein). Its main effect is to either smooth or wipe out some of the primary anisotropies; but the interactions of the photons with the matter in a fully ionised universe can also give rise to secondary anisotropies through the Vishniac effect (Vishniac 1987). This second order effect has maximum amplitudes for a very early re-ionisation. The case of a late inhomogeneous re-ionisation and its imprints on the CMB fluctuations has been investigated (Aghanim et al. 1996) and found to be rather important. In this case, the secondary anisotropies are due to the bulk motion of ionised clouds with respect to the CMB frame. When the re-ionisation is localised in hot ionised intra-cluster media the photons interact with the free electrons. The inverse Compton scattering between photons and electrons leads to the so-called Sunyaev-Zel'dovich (hereafter SZ) effect (Sunyaev \& Zel'dovich1972, 1980). The Compton distortion due to the motion of the electrons in the gas is called the thermal SZ effect. The kinetic SZ effect is a Doppler distortion due to the peculiar bulk motion of the cluster with respect to the Hubble flow. The SZ thermal effect has the unique property of depressing the CMB brightness in the Rayleigh-Jeans region and increasing its brightness above a frequency of about 219 GHz. This frequency dependence makes it rather easy to observe and separate from the kinetic SZ effect. In fact, the latter has a black body spectrum which makes the spectral confusion between kinetic SZ and primary fluctuations a serious problem. The SZ effect has been widely studied for individual clusters and for populations of clusters. For full reviews on the subject we refer the reader to two major articles: Rephaeli 1995 and Birkinshaw 1997. These investigations have clearly shown that the SZ effect in clusters of galaxies provides a powerful tool for cosmology through measurements of the Hubble constant, the radial peculiar velocity of clusters and consequently the large scale velocity fields. \par Besides the interactions with the ionised matter, some secondary effects arise when the CMB photons traverse a varying gravitational potential well. In fact, if the gravitational potential well crossed by the photons evolves between the time they enter the well and the time they leave it, the delay between entrance and exit is equivalent to a shift in frequency, which induces a temperature anisotropy on the CMB. This effect was first studied by Rees \& Sciama (1968) for a potential well growing under its own gravity. Numerous authors have investigated the potential variations due to collapsing objects and their effect on the CMB (Kaiser 1982, Nottale 1984, Martinez-Gonz\'alez, Sanz \& Silk 1990, Seljak 1996). Similarly, a gravitational potential well moving across the line of sight is equivalent to a varying potential and will thus imprint secondary fluctuations on the CMB. This effect was first studied for one cluster of galaxies by Birkinshaw \& Gull (1983) (Sect. 2). Kaiser \& Stebbins (1984) and Bouchet, Bennett \& Stebbins (1988) investigated a similar effect for moving cosmic strings. Recent work (Tuluie \& Laguna 1995, Tuluie, Laguna \& Anninos 1996) based on N-body simulations has pointed out this effect in a study of the effect of varying potential on rather large angular scales ($\simeq 1^{\circ}$). A discussion of some of these results and a comparison with ours will follow in the next sections. \par In this paper, following the formalism of Birkinshaw \& Gull (1983) and Birkinshaw (1989), we investigate the contribution of secondary anisotropies due to a population of collapsed objects moving across the line of sight, these objects range from small groups to rich clusters in scale ($10^{13}$ to $10^{15}$ $M_{\sun}$). In section 2., we first study in detail the case of a unique collapsed structure. We use a structure model to compute in particular the deflection angle and derive the spatial signature of the moving lens effect. We then account (Section 3.) for the contribution, to the primordial cosmological signal, of the whole population of collapsed objects using predicted counts and we simulate maps of these secondary anisotropies. In section 4., we analyse the simulated maps and present our results. We give our conclusions in section 5. | In our work, we investigate the secondary fluctuations induced by moving lenses with masses ranging from those of groups of galaxies to those of clusters of galaxies in a simple way, based on predicted structure counts and simulated maps. This method allows us to explore a rather wide range of scales ($>10$ arcseconds) in various cosmological models. The analysis, in terms of angular power spectra, show the scales for which the primary fluctuations are dominant (Fig. \ref{fig:pspec}). In the standard and lambda CDM models, the primary anisotropies are dominant respectively for scales $l<4000$ and $l<4500$ whereas in the Open CDM model they are dominant for $l<6000$. In practice, it is thus impossible to detect the secondary anisotropies due to moving lenses in the open model. The standard CDM model shows the smallest cut-off scale with an intermediate SZ kinetic pollution, compared to the other two models. It is therefore the ``best case'' framework for making an analysis and predicting the detection of fluctuations and the contributions that they induce. One must keep in mind that the results quoted in this particular case represent the ``best'' results we get from the analysis. \par The results of our analysis are obtained under the assumption of a universe that never re-ionises, which is of course not the case. The re-ionisation, if it is homogeneous, is supposed to somewhat ease the task of extraction of the pattern. In fact, its main effect is to damp the angular power spectrum of the primary anisotropies on small scales, shifting the cut-off towards larger scales. In this case, the effect of moving lenses dominates over the CMB fluctuations, and the SZ kinetic is not as high as it is on very small scales. However, if the re-ionisation is late and inhomogeneous, it generates additional SZ kinetic-type secondary fluctuations (Aghanim et al. 1996) without damping the power spectrum by more than a few percent. Here, the re-ionisation might worsen the analysis at small scales. In any case, there could be some other additional secondary fluctuations principally due to the Vishniac effect, that arise in a re-ionised universe. Our work thus gives a ``best case'' configuration of the problem, with all other effects tending to worsen the situation. \par We found that the secondary fluctuations induced by the moving gravitational lenses can be as high as $1.5\,10^{-5}$; with {\it rms} contributions of about 5 to $3.\,10^{-7}$ in the three cosmological models. Even if the moving lens fluctuations have a particular dipolar pattern and even if they are ``perfectly'' located through their SZ thermal effect, the detection of the moving lens effect and its separation from the SZ kinetic and primary fluctuations are very difficult because of the very high level of confusion, on the scales of interest, with the point--like SZ kinetic anisotropies and because of spectral confusion. \par We nevertheless analysed the simulated maps using an adap\-ted wavelet technique in order to extract the moving lens fluctuations. We conclude that {\it the contribution of the secondary anisotropies due to the moving lenses is thus negligible whatever the cosmological model. Therefore it will not affect the future CMB measurements except as a background contribution. We have highlighted the fact that the moving lens fluctuations have a very significant spatial signature but we did not succeed in separating this contribution from the other signals.} | 98 | 3 | astro-ph9803040_arXiv.txt |
9803 | astro-ph9803276_arXiv.txt | Images recorded through broad ($J, H, K$), and narrow (CO, and $2.2\mu$m continuum) band filters are used to investigate the photometric properties of bright ($K \leq 13.5$) stars in a $6 \times 6$ arcmin field centered on the SgrA complex. The giant branch ridgelines in the $(K, J-K)$ and $(K, H-K)$ color-magnitude diagrams are well matched by the Baade's Window (BW) M giant sequence if the mean extinction is A$_K \sim 2.8$ mag. Extinction measurements for individual stars are estimated using the M$_K$ versus infrared color relations defined by M giants in BW, and the majority of stars have A$_K$ between 2.0 and 3.5 mag. The extinction is locally high in the SgrA complex, where A$_K \sim 3.1$ mag. Reddening-corrected CO indices, CO$_o$, are derived for over 1300 stars with $J, H$, and $K$ brightnesses, and over 5300 stars with $H$ and $K$ brightnesses. The distribution of CO$_o$ values for stars with $K_o$ between 11.25 and 7.25 can be reproduced using the M$_K -$ CO$_o$ relation defined by M giants in BW. The data thus suggest that the most metal-rich giants in the central regions of the bulge and in BW have similar photometric properties and $2.3\mu$m CO strengths. Hence, it appears that the central region of the bulge does not contain a population of stars that are significantly more metal-rich than what is seen in BW. \vspace{0.3cm} \noindent{Key words: Galaxy: center -- stars: abundances -- stars: late-type} | The stars in the central regions of the Galaxy have been the target of numerous photometric and spectroscopic investigations. The brightest, best-studied, objects belong to a young population (e.g. reviews by Blum, Sellgren, \& DePoy 1996a, and Morris \& Serabyn 1996) that dominates the near-infrared light within 0.1 -- 0.2 parsec ($\sim 2.5 - 5.0$ arcsec) of SgrA* (Saha, Bicknell, \& McGregor 1996), and is concentrated within the central $\sim 1$ parsec (Allen 1994). However, there is also a large population of older stars near the Galactic Center (GC) belonging to the inner bulge, and it is only recently that efforts have been made to probe the nature of these objects. Minniti {\it et al.} (1995) reviewed metallicity estimates for a number of bulge fields, and a least squares fit to these data indicates that $\Delta$[Fe/H]/$\Delta$log(r) $= -1.5 \pm 0.4$ (Davidge 1997). The presence of a metallicity gradient suggests that the material from which bulge stars formed experienced dissipation, so it might be anticipated that the central regions of the bulge will contain the most metal-rich stars in the Galaxy. While the fields considered by Minniti {\it et al.} are at relatively large distances from the GC, and hence do not monitor trends at small radii, observations of other galaxies indicate that population gradients can extend into the central regions of bulges. This is clearly evident in spectroscopic studies of M31 (e.g. Davidge 1997), a galaxy that shares some morphological similarities with the Milky-Way (e.g. Blanco \& Terndrup 1990). It is not clear from the existing observational data if the bright stellar contents in the inner bulge and Baade's Window (BW), a field that is dominated by stars roughly 0.5 kpc from the GC, are similar. The $K$ luminosity function (LF) of moderately faint stars within an arcmin of SgrA* has a power-law exponent similar to that seen in BW (Blum {\it et al.} 1996a; Davidge {\it et al.} 1997). However, the significance of this result is low, as the LFs of bright giant branch stars are insensitive to metallicity (e.g. Bergbusch \& VandenBerg 1992). The brightest inner bulge stars, which are evolving on the asymptotic giant branch (AGB), have $2\mu$m spectroscopic properties reminiscent of bright giants in BW, although detailed measurements reveal that for a given equivalent width of near-infrared Na and Ca absorption, the inner bulge stars have deeper CO bands than giants in BW (Blum, Sellgren, \& DePoy 1996b). While suggestive of differences in chemical composition, it should be recalled that these objects are the brightest, most highly evolved members of the bulge, and Na can be affected by mixing (e.g. Kraft 1994). Hence, the spectroscopic properties of these bright red giants may not be representative of fainter objects. As the strongest features in the near-infrared spectra of cool evolved stars, the $2.3\mu$m first-overtone CO bands provide an important means of probing the stellar content of the inner bulge. Unfortunately, the crowded nature of the inner bulge, coupled with the low multiplex advantage offered by the current generation of cryogenically-cooled spectrographs, most of which use a single long slit, makes a $2\mu$m spectroscopic survey of a large sample of moderately faint objects a difficult task at present. Narrow-band imaging, using filters such as those described by Frogel {\it et al.} (1978), provides a highly efficient alternate means of measuring the strength of CO absorption in a large number of objects. In the current paper $J, H, K$, CO and $2.2\mu$m continuum measurements of moderately faint ($K \leq 13.5$) stars are used to measure the strength of CO absorption in stars within 3 arcmin of the GC. To the best of our knowledge, this is the largest survey of stellar content in the central regions of the bulge conducted to date. The observations and reduction techniques are described in \S 2, while the photometric measurements are discussed in \S 3. In \S 4 the line-of-sight extinction to these sources is estimated by assuming that they follow the same M$_K -$ color relations as M giants in BW. Reddening-corrected CO indices are derived, and the distribution of CO indices is compared with that predicted if stars in the inner bulge and BW follow similar M$_K -$ CO relations. A summary and brief discussion of the results follows in \S 5. | Moderately deep near-infrared images have been used to probe the bright ($K \leq 13.5$) stellar content in the central $6 \times 6$ arcmin field of the Galaxy. The ridgeline of the bulge giant branch on the $(K, J-K)$ and $(K, H-K)$ CMDs is well matched by the BW M giant sequence, reddened according to the Rieke \& Lebofsky (1985) extinction curve. This similarity in photometric properties suggests that the extinctions to individual stars in the inner bulge can be estimated by adopting the M$_K -$ color relations defined by M giants in BW. The mean extinction outside of the SgrA complex, where A$_K = 3.1$ mag, is A$_K = 2.8$ mag. The extinction estimates for individual stars have been used to generate reddening-corrected CO indices, and the histogram distribution of CO$_o$ values can be reproduced using the M$_K -$ CO$_o$ relation defined by M giants in BW. Therefore, M giants near the GC and in BW have similar $2.3\mu$m CO strengths. A potential source of systematic error in the procedure used to compute A$_K$ is that a single set of M$_K -$ color relations have been used, and no attempt has been made to allow for a dispersion in the metallicities of giants in the central regions of the bulge. There is a selection effect for magnitude-limited samples, in the sense that the brightest stars in a field containing an old composite population are likely the most metal-rich, so the current data likely do not sample the full range of metallicities near the GC. In any event, the CO$_o$ distribution does not change substantially when M$_K -$ color relations defined by red giants in 47 Tuc ([Fe/H] $\sim -0.7$) are used to estimate extinction, indicating that the main conclusions of this paper are insensitive to the adopted intrinsic colors of GC giants. Another source of systematic error is that the A$_K$ values derived in \S 4 require measurements in $J, H$, and $K$. This broad wavelength coverage introduces a bias against heavily reddened objects, which are faint, and hence may not be detected, in $J$. In fact, there are regions where the extinction is so high that stars are not detected in $K$. The tendency to miss the most heavily reddened stars skews the A$_K$ distribution to lower values. One way to reduce, but not entirely remove, this bias is to estimate extinction using only $H-K$ colors, for which the number of objects is over $3 \times$ greater than those with $J$ measurements. Following the procedure described earlier, $(H-K)_0$ was assigned to each star using the $M_K - (H-K)$ relation for BW M giants, and the results are shown in the lower panels of Figures 7 -- 12, while the radial distribution of CO$_o$ values is plotted in Figure 14. It is evident from the lower panel of Figure 8 that a number of sources with relatively high A$_K$ are added to the sample when measurements in $J$ are not required. Nevertheless, the impact on the CO$_o$ distribution is negligible, indicating that stars with higher than average obscuration near the GC have $2.3\mu$m CO strengths that are similar to less heavily reddened stars. It should be emphasized that the $2.3\mu$m CO bands provide only one diagnostic of chemical composition. Indeed, studies of the radial behaviour of the $2.3\mu$m CO bands in the bulges of other galaxies reveal weak or non-existant gradients (e.g. Frogel {\it et al.} 1978), even though many of these systems show line strength gradients at optical wavelengths. When interpreting this ostensibly contradictory result it should be recalled that the CO measurements are dominated by the brightest red stars which, in old populations, will be those that are the most metal-rich. If the most metal-rich population in the bulges of other galaxies follows a single M$_K -$ CO relation with no radial dependence, as appears to be the case in the inner regions of the Galactic bulge, then this would help to explain why the wide-aperture CO measurements of other systems do not show gradients. There are indications that the abundances of some species in the spectra of metal-rich giants may vary with distance from the GC. In particular, the infrared spectra obtained by Blum {\it et al.} (1996b) indicate that $2.3\mu$m CO absorption in GC giants is {\it stronger} than in BW stars having the same $2\mu$m Na and Ca absorption line strengths. Therefore, given that the CO line strengths in giants near the GC are similar to those in BW, then the Blum et al. measurements are suggestive of radial changes in [Na/Fe] and [Ca/Fe] among bulge stars. While this paper has concentrated on bulge stars, a modest disk population is also evident in the CMDs, and these data can be used to estimate the contribution disk objects make to the near-infrared light output near the GC. Objects with $(J-K) \leq 2.5$ and brightnesses between $K \sim 7.5$ (the approximate saturation limit of the CTIO data) and $K \sim 12$ (the faintest point at which the CTIO $K, J-K$ CMD is complete over a broad range of colors) account for 2.6\% of the total light from resolved sources within 3 arcmin of the GC. The relative contribution made by disk stars would be much lower if dust did not obscure the bulge. Assuming that (1) the disk stars are not heavily reddened, and (2) stars near the GC are obscured by A$_K \sim 2.8$ mag then, after correcting for this extinction, the contribution from disk stars drops to only 0.05\% when $K_0 \leq 8.5$. \vspace{0.3cm} Sincere thanks are extended to the referee, Jay Frogel, for providing comments that greatly improved the paper. \clearpage \begin{table*} \begin{center} \begin{tabular}{cccr} \tableline\tableline IRS \# & K$_{TD}$ & K$_{BSD}$ & $\Delta$ \\ \tableline 7 & 6.44 & 6.55 & $-0.11$ \\ 9 & 8.80 & 8.57 & 0.23 \\ 16NE & 8.56 & 9.01 & $-0.45$ \\ 28 & 9.41 & 9,36 & 0.05 \\ \tableline \end{tabular} \end{center} \caption{Comparison with $K$ measurements obtained by Blum {\it et al.} (1996a)} \end{table*} \clearpage | 98 | 3 | astro-ph9803276_arXiv.txt |
9803 | astro-ph9803106_arXiv.txt | Lithium is an excellent tracer of mixing in stars as it is destroyed (by nuclear reactions) at a temperature around $\sim 2.5\times 10^6$ K. The lithium destruction zone is typically located in the radiative region of a star. If the radiative regions are stable, the observed surface value of lithium should remain constant with time. However, comparison of the meteoritic and photospheric Li abundances in the Sun indicate that the surface abundance of Li in the Sun has been depleted by more than two orders of magnitude. This is not predicted by solar models and is a long standing problem. Observations of Li in open clusters indicate that Li depletion is occurring on the main sequence. Furthermore, there is now compelling observational evidence that a spread of lithium abundances is present in nearly identical stars. This suggests that some transport process is occurring in stellar radiative regions. Helioseismic inversions support this conclusion, for they suggest that standard solar models need to be modified below the base of the convection zone. There are a number of possible theoretical explanations for this transport process. The relation between Li abundances, rotation rates and the presence of a tidally locked companion along with the observed internal rotation in the Sun indicate that the mixing is most likely induced by rotation. The current status of non-standard (particularly rotational) stellar models which attempt to account for the lithium observations are reviewed. | Li\footnote{In this review I will use Li to represent $^7$Li, the isotope which is produced by big bang nucleosynthesis. $^7$Li accounts for $\sim 93\%$ of the total Li abundance in meteorites. Observers typically measure the total Li abundance, while theoretical models determine depletion factors for $^7$Li and $^6$Li The $^6$Li isotope is destroyed at much lower temperatures and $^7$Li. When making the comparison between the observations and theory, it is usually assumed that the $^6$Li contribution to the stellar Li content is neglible. However, see the discussion on halo stars (\S \ref{secthalo}) where observations of $^6$Li may be used to elucidate the mixing mechanism operating in these stars.} is a sensitive tracer of mixing in stellar radiative regions as it is easily destroyed at temperatures above $\sim 2.5 \times 10^6\,$K. For solar type stars, the Li destruction region is located below the surface convection zone in standard models. As a consequence, standard stellar models predict that Li should not be depleted at the surface of solar type stars. This is a rather robust prediction of stellar evolution theory, which has been known for 40 years \cite{schwar}. However, comparisons between the solar photospheric Li abundance and the Li abundance in meteorites show that the Sun has depleted a substantial amount of Li at its surface \cite{green51}. The solar Li depletion problem has posed a challenge to stellar evolution theory for 40 years, and the solution to this puzzle is still open to debate. The Sun is unique in that helioseismic observations allow us to probe the interior structure and rotation of the Sun. These observations can put constraints on possible solutions to the solar Li depletion problem, but by themselves solar observations cannot uniquely determine the cause of solar Li depletion. Observations of stellar Li abundances allow one to study the Li depletion problem as a function of age, metallicity and stellar mass. As such, they provide a powerful test for mechanisms which attempt to explain the solar Li depletion. The discovery of a large dip in Li abundances around 6600 K in the Hyades \cite{lidip} was not predicted by theorists, and remains a major challenge to theoretical stellar evolution models. There is increasing evidence that a dispersion in Li abundances exists among stars with similar ages, metallicities and masses (\citeauthor{soder93} \citeyear{soder93}; \citeauthor{boes98} \citeyear{boes98}). Such a dispersion suggests that another stellar property is important in determining the amount of Li which is depleted in stars. There is mounting observational evidence that rotation plays a key role in determining the amount of Li depletion in a star (\citeauthor{hyadbin} \citeyear{hyadbin}; \citeauthor{jones97} \citeyear{jones97}). In this review, I will discuss the relationship between mixing, rotation and Li abundances in stars. | } There is a wealth of data on Li abundances and rotation velocities in stars with a variety of ages, masses and metallicities. This data clearly indicates that Li depletion occurs on the main sequence for all stars with a surface convection zone. This is in direct contradiction with standard stellar evolution theory. A number of possible mechanisms which lead to extra Li depletion have been put forth. The dispersion in Li abundances at a given age, metallicity and temperature, the correlation between Li abundances and rotation velocities in Pleiades, the fact that tidally locked binary stars in the Hyades have an excess Li abundance as compared to single stars, and the detection of moderate Be deficiencies among Li dip stars with detectable Li abundances, all imply that that rotation induced mixing is leading to Li depletion on the main sequence. Helioseismic observations of the Sun support this hypothesis, for they show that slow form of slow mixing is operating below the base of the solar convection zone \cite{basu}. Current stellar models which incorporate rotation induced mixing explain many, by not all of the observations. Models which are able to account for all of the data are likely to include diffusion, rotation induced mixing, angular momentum transport by gravity waves and/or magnetic fields and modest stellar winds. | 98 | 3 | astro-ph9803106_arXiv.txt |
9803 | astro-ph9803330_arXiv.txt | We have used a coupled time-dependent chemical and dynamical model to investigate the lifetime of the chemical legacy left in the wake of C-type shocks. We concentrate this study on the chemistry of \HtwoO\ and \Otwo , two molecules which are predicted to have abundances that are significantly affected in shock-heated gas. Two models are presented: (1) a three-stage model of pre-shock, shocked, and post-shock gas; and (2) a Monte-Carlo cloud simulation where we explore the effects of stochastic shock activity on molecular gas over a cloud lifetime. For both models we separately examine the pure gas-phase chemistry as well as the chemistry including the interactions of molecules with grain surfaces. In agreement with previous studies, we find that shock velocities in excess of 10 km s$^{-1}$ are required to convert all of the oxygen not locked in CO into \HtwoO\ before the gas has an opportunity to cool. For pure gas-phase models the lifetime of the high water abundances, or ``\HtwoO\ legacy'', in the post-shock gas is $\sim 4 - 7 \times 10^{5}$ years, independent of the gas density. A density dependence for the lifetime of \HtwoO\ is found in gas-grain models as the water molecules deplete onto grains at the depletion timescale. Through the Monte Carlo cloud simulation we demonstrate that the time-average abundance of \HtwoO\ -- the weighted average of the amount of time gas spends in pre-shock, shock, and post-shock stages -- is a sensitive function of the frequency of shocks. Thus we predict that the abundance of \HtwoO , and to a lesser extent \Otwo , can be used to trace the history of shock activity in molecular gas. We use previous large-scale surveys of molecular outflows to constrain the frequency of 10 km s$^{-1}$ shocks in regions with varying star-formation properties and discuss the observations required to test these results. We discuss the post-shock lifetimes for other possible outflow tracers (e.g. SiO, \CHthreeOH) and show that the differences between the lifetimes for various tracers can produce potentially observable chemical variations between younger and older outflows. For gas-grain models we find that the abundance of water-ice on grain surfaces can be quite large and is comparable to that observed in molecular clouds. This offers a possible alternative method to create water mantles without resorting to grain surface chemistry: gas heating and chemical modification due to a C-type shock and subsequent depletion of the gas-phase species onto grain mantles. | The importance of molecular gas as the material from which stellar and planetary systems are made has led numerous investigators to examine the chemical processes that combine atoms into molecules. Over the past two decades these studies have gradually increased in both complexity and fidelity and can roughly be divided into three generations. The first generation chemical model solved the chemical rate equations at equilibrium, or steady-state, and demonstrated the importance of ion-molecule reactions in driving the gas-phase chemistry (\cite{HK73}). The second generation of models utilized the later availability of increased computing power to study the time evolution of chemical abundances (\cite{PH80}; \cite{GLF82}). These models, labeled as ``pseudo-time dependent'' -- ``pseudo'' because the chemistry evolves with fixed physical conditions -- used observed physical properties of dense molecular cores ($T_k \sim 10 - 30$ K, \nhtwo\ $= 10^{4-6}$ \cc ) to show that chemical equilibrium was reached in $\sim 10^{6-8}$ years. Pseudo-time dependent models represent the most prevalent chemical model and have been quite successful in describing the chemistry of quiescent regions in both dark and giant molecular cloud cores (\cite{Lee_etal96}; \cite{BGSL97}). Second generation chemical models operate under one simplifying assumption: that the molecular gas undergoes no dynamical evolution as it chemically evolves. However, molecular clouds are certainly dynamically evolving objects. The widespread occurrence of superthermal widths in molecular lines suggests that, overall, the evolution of molecular clouds is not simple quiescent evolution at a single gas temperature. Moreover, the formation of a low- or high-mass star from a molecular condensation involves a collapse that increases the density by many orders of magnitude. It has also been recognized that the birth of a protostar is associated with a period of intense mass loss, which manifests itself in energetic winds, bipolar jets, and large-scale outflows (c.f. \cite{Lada85}; \cite{Bachiller96}). The impact of energetic flows on surrounding quiescent gas can compress and heat the gas, in some cases providing enough energy to overcome endothermic barriers of chemical reactions, sputter or destroy grains, or even dissociate molecules. The result can be a considerably different chemical composition in the shocked gas than observed in quiescent material. Since molecules are the primary coolants of the gas, any chemical change induced by collapse or shocks can also alter the ensuing physical evolution of the cloud. To address such concerns, a third generation of chemical models has been constructed to build on the successes of previous generations by combining chemistry with dynamics. These models have been principally directed towards investigating the chemistry of core and star formation (c.f. \cite{PTVH87}; \cite{RHMW92}; \cite{BL97}), the physical and chemical structure of shocked gas (c.f. \cite{DM93}), or even complex scenarios in which gas is cycled between low and high density through recurrent episodes of low-mass star formation (\cite{CDHW88a}). One goal of coupled astrochemical models is to search for and isolate specific molecular species that serve as signposts of particular dynamical events, such as shocks. Coupled models of shocked molecular gas have isolated one molecule in particular, \HtwoO , which is predicted to form in large abundance in shock-heated gas (\cite{DRD83}; \cite{KN96a},b); if temperatures in the shocked gas exceed 400~K, the endothermic barriers of a few key chemical reactions are exceeded and all of the available oxygen not locked up in CO will be driven into \HtwoO . This led to the assertion that water emission is a ubiquitous tracer of shock-heated gas (\cite{NM87}). Unfortunately, due to constraints imposed by the earth's atmosphere, the detection of \HtwoO\ in interstellar clouds from ground-based observatories is a challenging task. Nevertheless, several studies have observed, and detected, transitions of isotopic water (\HtwoeiO), and even \HtwoO\ (c.f. \cite{Jacq_etal88}; \cite{KL91}; \cite{Zmuidzinas_etal94}). Quite recently, absorption from warm ($T_{gas} > 200$ K) water vapor has been unambiguously detected by the Infrared Space Observatory (ISO) towards hot stars (\cite{Helmich_etal96}; \cite{vDH96}), and in emission in HH54 (\cite{Liseau_etal96}). The inferred water abundance in these sources is $\sim 1 - 6 \times 10^{-5}$, much larger than predicted by chemical models run for cold quiescent conditions ($x$(\HtwoO ) $\sim$10$^{-7}$), and is consistent with water production in warm gas either though high-temperature chemistry -- as would apply in shock-heated gas or in the near vicinity of embedded sources -- or via evaporation of water-ice mantles. An important question yet to be addressed by coupled dynamical and chemical models of shocked gas in molecular clouds is how long the high abundances of water and other molecules persist following the passage of a shock. As we will demonstrate, the time needed for shock-heated gas to return to its pre-shock temperature is several orders of magnitude less than the time required for the gas to return to its pre-shock chemical composition. Thus, the enhanced abundances of water and other species could potentially exist long after the dynamical effects of a shock passage are dissipated. Therefore, the chemistry inside a cloud is reflective of, and can be used to probe, the physical shock history of the molecular gas. Motivated by these questions, we present the results of a coupled dynamical and chemical study of molecular gas that is subjected to shocks with velocities greater than 10 km s$^{-1}$. In particular, we follow the time dependence of the chemistry in a shock-heated gas layer as it cools and the quiescent time-dependent chemistry is ultimately re-established. We present models examining this evolution using pure gas-phase chemistry as well as chemistry which includes the interaction of molecules with grain surfaces. In Section 2 we discuss the combined chemical and dynamical model. In Section 3 we use published surveys of molecular outflows to investigate the average rate at which shocks with a minimally sufficient velocity to affect the water chemistry -- 10 km s$^{-1}$ -- pass a given region of a cloud. Section 4 presents the results from our combined models along with one example of quiescent chemical evolution. Two models are presented: (1) a three-stage model of pre-shock, shocked, and post-shock gas; and, (2) a Monte-Carlo cloud simulation in which the results from Section 3 are used to examine the effects of stochastic shock activity on molecular gas over a cloud lifetime. Section 5 discusses the importance of these results on chemical models of molecular gas and also reviews the observations required to test these models. In Section 6 we summarize our results. | \subsection{Comparison with Observations} Our results demonstrate that the {\em time-averaged} abundance of \HtwoO\ is a sensitive function of the rate at which molecular gas is subjected to shocks. In order to properly interpret these results it is useful to consider how the time-averaged abundances relate to observations of water in a giant molecular cloud (GMC). The gas inside a star-forming cloud can roughly be divided into three categories: (1) quiescent material that has been relatively unaffected by current or previous epochs of star formation; (2) gas that is being physically and/or chemically affected by current star formation; and, (3) as suggested here, gas that has been affected by a previous generation of star formation and is in the process of chemically and dynamically evolving back to a quiescent state. Observations of \HtwoO\ toward molecular material associated with each of these categories should find abundances varying from as low as $x(\rm{H_2O}) \sim 10^{-6}$ in (1), and ranging upwards to $\le 10^{-4}$ in (2) and (3). {\em A single pointed detection of H$_2$O emission in a GMC can therefore be represented by a single stage of the three-stage or Monte-Carlo models (quiescent, shock, post-shock). Because star formation will be spread throughout a cloud or core, the time-averaged abundance therefore refers to water abundances averaged over an area that contains both quiescent material and any associated gas currently being affected by local star formation. Thus the average abundance over a cloud (cloud-average) is comparable to the time-averaged abundances in the Monte-Carlo cloud model.} This average abundance might not necessarily apply to an entire GMC complex (e.g. Orion, Gem OB1), which can extend for several square degrees on the sky, but may apply to several dense cores within a single cloud. Regions with high star-formation rates, such as OMC-1 which is associated with the Trapezium cluster (see Section 3), might be expected to have a higher probability for strong shocks, and therefore a high cloud-averaged water abundance. Cores with lower star-formation rates, such as L1641 also in Orion, would have low cloud-averaged water abundances. {\em To test our model's ability to constrain the history of molecular clouds it is important to obtain maps of the water emission over large spatial scales in GMC cores.} Because of the strong absorption by atmospheric water vapor, observations of \HtwoO\ in the ISM are extremely difficult. As a result, the small number of detections of water in molecular clouds have typically been taken towards single lines of sight mostly containing luminous protostars. The very convincing detections of water by ISO provide the greatest evidence for high water abundances in hot gas, with $x(\rm{H_2O}) \sim 1 - 6 \times 10^{-5}$ inferred towards hot stars (\cite{Helmich_etal96}; \cite{vDH96}) and HH54 (\cite{Liseau_etal96}). Towards Sgr B2 Zmuidzinas et al. (1994) observed \HtwoeiO\ in absorption with the Kuiper Airbourne Observatory (KAO), while Neufeld et al. (1997) observed \HtwoO\ in emission using ISO. Combining these observations with previous ground based detections Neufeld et al. (1997) estimate an abundance of $x(\rm{H_2O}) = 3.3 \times 10^{-7}$ for the cooler outer parts of Sgr B2 and $x(\rm{H_2O}) \sim 5 \times 10^{-6}$ in the hot core. These observations are important not only because \HtwoO\ was detected in dense gas, but also because there appears to be a range of water abundances. However, it is difficult to discern whether the enhanced water abundances are the result of high-temperature chemistry that occurs behind shocks, high-temperature chemistry appropriate to gas near young stars (c.f. Doty \& Neufeld 1997), or evaporation of grain mantle species (see discussion in Section 5.2) and there is little information on the spatial distribution of the water emission. There have been a few attempts to map the extended emission of water. Gensheimer, Mauersberger, \& Wilson (1996) mapped emission of the quasi-thermal $3_{13} \rightarrow 2_{20}$ transition of \HtwoeiO\ in both the Orion Hot Core and Sgr B2, and found that the emission originated in a compact region ($< 10''$). Cernicharo et al. (1994) and Gonzalez-Alfonso et al. (1995) find evidence for widespread water emission in Orion and W49N using the $3_{13} \rightarrow 2_{20}$ masing transition of \HtwoO . In Orion they argue that the water abundance is quite high, $x(\rm{H_2O}) > 10^{-5}$, over an extended $50'' \times 50''$ region centered on BN-KL and the Orion Nebular Cluster. From these results, the cloud-averaged water abundance for the central regions of the Orion core near the Orion Nebular Cluster is $^{>}_{\sim} 10^{-5}$ which, using Figures 11 and 12 (\nhtwo\ $= 10^6$ \cc ; \cite{BSG96}), is consistent with $\tau_s \sim 10^6$ years, as suggested in Section 3. However, the determination of abundances from masing transitions is a difficult task and these results must be viewed as suggestive until they are confirmed by mapping data obtained in other \HtwoO\ lines. For molecular oxygen, which also suffers from strong absorption due to atmospheric \Otwo , the situation is even more uncertain. There exists only one tentative detection of $^{16}$O$^{18}$O towards L134N by \cite{PLC93}, which implies $x(\rm{O_2}) \sim 4 - 8 \times 10^{-5}$. However, a search for $^{16}$O$^{18}$O emission in different positions in L134N and other galactic sources by \cite{MPLC97} did not confirm this detection and provides only upper limits of \Otwo /CO $< 0.1$, which, assuming $x$(CO) $= 2.7 \times 10^{-4}$ (Lacy et al 1994), gives $x(\rm{O_2}) < 3 \times 10^{-5}$. Searches, with similar negative results, have also been performed for molecular oxygen in extra-galactic sources with favorable redshifts (Goldsmith \& Young 1989; Combes et al. 1991; Liszt 1992) finding $x(\rm{O_2}) < 10^{-6}$. A recent study by Combes, Wiklind, \& Nakai (1997) towards a z = 0.685 object provides the lowest limit to date of \Otwo /CO $< 2 \times 10^{-3}$. For our most realistic case, the Monte Carlo gas-grain model, we find the time-averaged \Otwo\ abundance is $\sim 2 \times 10^{-5}$. Thus, the combined effects of shocks and gas-grain chemistry could lower \Otwo\ abundances below the observed limits in Galactic sources, but another explanation is required to account for extra-galactic observations. The low time-averaged \Otwo\ abundance suggests that these results could have some bearing on the high abundances of neutral carbon relative to CO (C/CO $\sim$ 0.1) that are observed towards a variety of star forming regions (c.f. Plume 1995, Schilke et al 1996). However, because the number of dissociative shocks which destroy CO is not high enough, the time-averaged carbon abundance in the models is well below the observed value. The best opportunity to test these results will be offered by two spaceborne observatories, {\em The Submillimeter Wave Astronomy Satellite (SWAS)} (\cite{Melnick_etal95}) and {\em ODIN} (\cite{Hjalmarson95}), both of which should launch within the next year. {\em SWAS} and {\em ODIN} are capable of observing and mapping the fundamental transition $1_{10} \rightarrow 1_{01}$ of \HtwoO\ at 557 GHz and the $3,3 \rightarrow 1,2$ transition of \Otwo\ at 487 GHz. These transitions have low upper state energies ($\sim 26$ K) and should be readily excited in hot gas near star-forming sites and, more importantly, in the colder more extended material such as the ridge of dense gas in Orion (c.f. \cite{Ungerechts_etal97}). \begin{figure*} \figurenum{16} \plotfiddle{fig16.ps}{2.5in}{0}{50}{50}{-160}{-75} \caption{ Points defining the plane of O$_2$ and H$_2$O abundances for the gas-grain Monte-Carlo model with $\tau_s = 10^{6}$ years and $n_{H_2} = 10^5$ cm$^{-3}$. The spread of plotted points has been artificially increased by smearing them by 0.2 dex so as to give a better sense of the density of points in different regions of the plot. The cross represents the mean (time-averaged) abundances. } \label{shockspec} \end{figure*} Besides the computation of averaged abundances from mapping observations, combined observations of \HtwoO\ and \Otwo\ should be a powerful tool in examining the shock history of gas. This is demonstrated in Figure 16 where the water abundance is plotted versus the molecular oxygen abundance for the Monte-Carlo model, including grain depletion and desorption (Section 4.4.2). In this plot, the majority of points trace an area of roughly constant water abundance of $x(\rm{H_2O}) \sim 10^{-6}$, with the \Otwo\ abundance ranging between $x(\rm{O_2}) = 3 - 30 \times 10^{-6}$. This area has, by far, the largest number of points and defines the ``main-sequence'' of quiescent chemistry. From this main sequence a shock will trace a line extending almost horizontally to the right, with constant \Otwo\ abundance. Very strong shocks continue at the end of the horizontal line and extend down almost vertically, lowering the \Otwo\ abundance for constant \HtwoO\ abundance. This dependence of horizontal and vertical lines occurs because the \HtwoO\ is created more efficiently than \Otwo\ destruction (see Figures 3 and 4). This plot is a different way of examining the models, as opposed to average abundances, because it presents the evolution in a continuous fashion; the various possible solutions define a plane that traces evolutionary tracks including quiescent chemistry through a broad range of shock strengths. \subsection{Post-Shock Chemistry and Water Ice Mantles} For comparison with previous modeling efforts, our three-stage model is similar to the episodic models presented in Charnley et al. (1988a,b). However, our model is simpler in that we do not model the formation process of a dense clump nor do we account for any mixing between shocked and non-shocked layers. Like their models, we find that the chemistry converges to well defined abundance values even when numerous cycles are repeated. Charnley et al. (1988a) also find that the water abundance varies greatly between shock and quiescent cycles, but they do not examine in detail the chemistry in the post-shock layer. An interesting result in the gas-grain models is that the abundance of water on grain surfaces in post-shock gas is quite large $x(\rm{H_2O})_{gr} \sim 10^{-4}$. This abundance is quite close to that inferred for water ice along lines of sight towards background stars in Taurus, where $x(\rm{H_2O})_{gr} \sim 8 \times 10^{-5}$ (\cite{Whittet93}). Thus, these results offer an alternative explanation for the large abundance of water in grain mantles -- one that requires no grain surface chemistry! We stress that this mechanism may not be responsible for all \HtwoO\ observed on grains. Grain surface chemistry formation of \HtwoO\ must be considered -- especially during cloud formation stages when the abundance of atomic hydrogen and oxygen is higher. There are also other potential desorption mechanisms that could be active and are not included in these models, such as grain mantle explosions induced by UV radiation (\cite{SG91}) or water desorption via the infrared radiation field (\cite{WHW92}). However, since most molecular clouds are very active star-forming sites, we suggest that caution must be applied when interpreting observations of ice mantles in molecular clouds as the sole result of grain surface chemistry. The interpretation of high gas-phase water abundances towards hot stars as the result of either high-temperature chemistry or grain mantle evaporation of \HtwoO\ is further obscured because water on grains may have been produced in an earlier shock episode. It is possible that the HDO/H$_2$O ratio could discriminate between water mantles created in shocks or in grain mantles and we are in the process of investigating this question (Bergin, Neufeld, \& Melnick 1997). The large abundance of water on grains, with CO remaining in the gas phase, will also alter the ratio of total carbon-to-oxygen (C/O) in the gas phase. Comparison of chemical theory with observations of molecular abundances in GMC cores associated with massive star formation has found that the chemical abundances of many species are best reproduced with C/O ratios greater than the solar value ($>$ 0.4; \cite{BSMP87}; \cite{BGSL97}). For chemical modeling this is the result of depletion of the initial abundance of oxygen relative to carbon, which reduces the abundances of small oxygen-bearing species that are major carbon destroyers in the gas phase. In the three-stage gas-grain model shown in Figure 7, the C/O ratio in the post-shock gas at $t \sim 10^{5}$ years is C/O $\sim 0.7$. It is difficult to gauge the effect of this on other species, because of the lack of certainty with regard to high-temperature reactions. It is therefore possible that the C/O ratios inferred in Blake et al. (1987) and Bergin et al. (1997) are indicative of the core formation process, which could involve gas temperatures rising high enough to convert atomic oxygen to water. In this case, these results suggest a simple mechanism to place large amounts of oxygen on grain surfaces while still keeping most of the carbon in the gas-phase. Other molecular species, such as SO, SiO (\cite{MPBF92}), and \CHthreeOH\ (\cite{BLWC95}) have been observed with enhanced abundances in energetic outflows (see also \cite{Bachiller96}; \cite{Bachiller_Perez97}). These species are included in the three-stage chemical model presented in Section 4.3, but we do not present the results in detail here because of uncertainties in the high-temperature reaction rates. However, these results do have some bearing on the chemistry of these species in the post-shock layer. The formation pathway of \CHthreeOH\ in the gas phase is linked to a reaction of CH$_3^+$ with \HtwoO\ (\cite{MHC91}). Thus, it is possible that the enhanced abundance of water through high-temperature reactions could lead to larger \CHthreeOH\ abundances through this reaction. In the three-stage model, the \CHthreeOH\ abundance in the post-shock layer is $x(\rm{CH_3OH}) = 1 - 5 \times 10^{-8}$ when the shock temperature ranges from 1000 to 2000 K. These abundances are at least an order of magnitude below values inferred in molecular outflows, which can be as high as $x(\rm{CH_3OH}) \sim 10^{-6}$ (\cite{BLWC95}). The mechanism of methanol enhancements may therefore be the result of grain mantle evaporation or an unidentified high- or low-temperature pathway. The abundance of \CHthreeOH\ accreted onto grain surfaces provides another constraint. In our model the abundance of \CHthreeOH\ ice in the grain mantle is only 0.1 percent of the water-ice abundance. This ratio is below that observed towards NGC 7538 or W33A where $x(\rm{CH_3OH})$/$x(\rm{H_2O}) \sim 10-40$ percent (\cite{ASTH92}). Our models do not set constraints on the chemistry of SO and SiO because the large abundances of these species in outflows may be the result of sputtering of grain refractory and/or mantle material (c.f. Schilke et al. 1997; Caselli et al. 1997). When the abundances of these species are enhanced, through any mechanism, the high abundances will persist until the timescale for the individual molecule to deplete onto grain surfaces. If the dust temperature is high enough to keep a given species in the gas-phase, or for pure gas-phase chemistry, the lifetime will be $\tau_{ra} = 4 -7 \times 10^{5}$ years. In these models all SiO depletes onto grain surfaces at $\tau_{dep} \sim 10^4$ years (for \nhtwo\ = 10$^{5}$ \cc\ and assuming a sticking coefficient of unity). For \CHthreeOH\ abundance enhancements we find that the depletion timescale for gas-grain chemistry is equal to that of \HtwoO\ ($\tau_{dep}(H_2O) \sim 10^5$ years). The disparity in post-shock lifetimes between \CHthreeOH , \HtwoO , and SiO suggests that differences might exist between younger outflows, perhaps those associated with so-called Class~0 sources, and the more evolved sources (e.g. Class I). A survey of such sources in these tracers might prove to be useful in probing the links between evolutionary effects observed in outflows and the relationship to the driving source. Another possibility is that differences could also be found within different components inside a single outflow (see \cite{Bachiller_Perez97}). The abundance of some molecular species, notably \HCOp\ and \NtwoHp , are adversely affected by the high water abundance. In Figure 7, for the three-stage model, the abundance of \HCOp\ is decreased through reaction R6 when the water abundance is raised. Thus, the abundances of these two important molecular species should be anticorrelated. Because \NtwoHp\ also reacts with \HtwoO , a similar anticorrelated behavior would be found between the abundances of \NtwoHp\ and \HtwoO\ and may account for the low \HCOp\ and \NtwoHp\ abundances in hot regions such as the Orion Hot Core (c.f. Blake et al. 1987) and in the L1157 outflow (\cite{Bachiller_Perez97}). | 98 | 3 | astro-ph9803330_arXiv.txt |
9803 | astro-ph9803324_arXiv.txt | The IRAS 100 micron image of the GRB 970228 field shows that the amount of galactic dust in this direction is substantial and varies on arcminute angular scales. From an analysis of the observed surface density of galaxies in the $2.6' \times 2.6'$ HST WFPC image of the GRB 970228 field, we find $A_V = 1.1 \pm 0.10$. From an analysis of the observed spectra of three stars in the GRB 970228 field, we find $A_V = 1.71^{+0.20}_{-0.40}$. This value may represent the best estimate of the extinction in the direction of GRB 970228, since these three stars lie only $2.7''$, $16''$, and $42''$ away from the optical transient. If instead we combine the two results, we obtain a conservative value $A_V = 1.3 \pm 0.2$. This value is significantly larger than the values $A_V = 0.4 - 0.8$ used in papers to date. The value of $A_V$ that we find implies that, if the extended source in the burst error circle is extragalactic and therefore lies beyond the dust in our own galaxy, its optical spectrum is very blue: its observed color $(V-I_c)_{\rm obs} \approx 0.65^{+0.74}_{-0.94}$ is consistent only with a starburst galaxy, an irregular galaxy at $z > 1.5$, or a spiral galaxy at $z > 2$. On the other hand, its observed color and surface brightness $\mu_V \approx 24.5$ arcsec$^{-2}$ are similar to those expected for the reflected light from a dust cloud in our own galaxy, if the cloud lies in front of most of the dust in this direction. | 98 | 3 | astro-ph9803324_arXiv.txt |
||
9803 | astro-ph9803262_arXiv.txt | We present accurate measurements of the central wavelengths of 4947 atomic absorption lines in the solar optical spectrum. The wavelengths, precise to a level $\sim 50-150$ m s$^{-1}$, are given for both flux and disc-centre spectra, as measured in relatively recent FTS solar atlases. This catalogue modernizes existing sources based on photographic measurements and provides a benchmark to test and perform wavelength calibrations of astronomical spectra. It will also permit observers to improve the absolute wavelength calibration of solar optical spectra when lamps are not available at the telescope. | \label{sec1} Wavelength calibration is almost always needed in the process of producing useful astronomical spectra. To calibrate accurately is a non-trivial problem, in particular when working at high or very-high spectral resolution. Fourier transform spectrographs (FTS) are specially well-suited to this task, but they are not readily applied in conditions requiring high spatial or time resolution, so grating spectrometers are much more commonly used for astronomical observations. In this case, it becomes necessary to set reference positions corresponding to known wavelengths on the detector. This can be achieved by using very sharp observed telluric lines, but their location in the spectrum cannot be chosen by the astronomer. It is very usual to find spectral calibration lamps available for use with an astronomical spectrograph. The emission lines produced in the lamps have been previously measured at the laboratory, and this method usually provides a valid reference frame. However, it is often impractical to expose the calibration lamp simultaneously with the astronomical target and, unless the spectrograph is installed at a very stable focal station, the position of the spectrum on the detector varies depending on the telescope position. Accuracy is then limited by the instrument characteristics and observations of the calibration lamps are required between successive astronomical exposures. Nonetheless, calibrations via arc or hollow cathode spectra are normally accurate enough for most purposes. Ingenious techniques have been used to improve the accuracy of wavelength calibrations, such as placing gas cells at the entrance of the spectrograph (e.g., Deming \& Plymate 1994), but it is rare to find such systems available and convenient for regular observations. On occasion the available lamps are not very rich in lines in the spectral range of observation. In some circumstances, an external check of the final precision in the translation into wavelengths would be desirable. One method for tackling problems such as these is to use solar spectra as templates. Changes in the wavelengths of the lines in the integrated sunlight spectrum around the solar cycle have been proved to be very small, bellow some 15 m s$^{-1}$ (Jim\'enez et al. 1980; Wallace et al. 1988; McMillan et al. 1993; Deming \& Plymate 1994). At 5000 \AA, this translates into $\sim$ 0.3 m\AA, so the solar spectrum does offer a very stable source. In most practical cases, the accuracy will be imposed by the spectral resolution achieved. During night-time observations, the solar flux spectrum is observable after reflection from the Moon. Measurements of solar wavelengths in the integrated solar optical spectrum were published in 1929 by Burns and collaborators (Burns 1929; Burns \& Kiess 1929; Burns \& Meggers 1929), using photographic detectors and a grating spectrograph. The relatively recent solar flux FTS atlases offer a much higher quality spectrum of the Sun seen as a star. As the solar spectrum is so intense, on some solar telescopes no calibration lamps are deemed necessary, and the wavelength scale is set using the solar spectrum itself. Reasonable precision can be reached using the spectrum at the centre of the disc to compare with previously measured disc centre wavelengths, thus avoiding differential shifts due to the limb effect. In this case, small scale motions have to be averaged out, integrating in time and/or space, in order to minimize errors. The {\it Kitt Peak Table of Photographic Solar Spectrum Wavelengths} (Pierce \& Breckinridge 1973) has been extensively used by solar observers to set up the wavelength scale on their spectra. These observations, made on photographic plates, have been superseded in quality by the more recent FTS observations at the centre of the disc. To improve on the various sets of photographically based measurements (which date back to 1930 in the case of the solar flux spectrum), provide them in a homogeneous machine-readable format, use them to test spectral calibrations of very high resolution stellar spectra (e.g., Allende Prieto et al. 1995), and improve the accuracy of our own solar observations, we have determined the position of the central wavelengths of 4947 atomic lines in the optical solar spectrum. The employed source solar atlases, prepared from FTS data, and the fashion in which we performed the measurements is described in the succeeding sections. | 98 | 3 | astro-ph9803262_arXiv.txt |
|
9803 | astro-ph9803218_arXiv.txt | The goal of modern cosmology is to find the large scale matter distribution and spacetime structure of the Universe from astronomical observations, and it dates back from the early days of observational cosmology the realization that in order to achieve this aim it is essential that an accurate empirical description of galaxy clustering be derived from the systematic observations of distant galaxies. As time has passed, this realization has become a program, which in the last decade or so took a great impulse forward due to improvements in astronomical data acquisition techniques and data analysis. As a result of that an enormous amount of data about the observable universe was accumulated in the form of the now well-known {\it redshift surveys}, and some widely accepted conclusions drawn from these data created a certain confidence in many researchers that such an accurate description of the distribution of galaxies was just about to being achieved. However, those conclusions are mainly based in a standard statistical analysis derived from a scenario provided by the standard Friedmannian cosmological models, which assume homogeneity and isotropy of the matter distribution, scenario which is still thought by many to be the best theoretical framework capable of explaining the large scale matter distribution and spacetime structure of the Universe. The view outlined above, which now has become the orthodox homogeneous universe view, has, however, never been able to fully overcome some of its objections. In particular, many researchers felt in the past, and others still feel today, that the relativistic derived idea of an eventual homogenization of the {\it observed} matter distribution of the Universe is flawed, since, in their view, the empirical evidence collected from the systematic observation of distant cosmological sources also supports the claim that the universal distribution of matter will not eventually homogenize. Therefore, the critical voice claims that the large-scale distribution of matter in the Universe is intrinsically inhomogeneously distributed, from the smallest to the largest observed scales and, perhaps, indefinitively. Despite this, it is a historical fact that the inhomogeneous view has never been as developed as the orthodox view, and perhaps the major cause for this situation was the lack of workable models supporting this inhomogeneous claim. There has been, however, one major exception, in the form of a hierarchical cosmological model advanced by Wertz (1970, 1971), although, for reasons that will be explained below, it has unfortunately remained largely ignored so far. Nevertheless, by the mid 1980's those objections took a new vigour with the arrival of a new method for describing galaxy clustering based on ideas of a radically new geometrical perspective for the description of irregular patterns in nature: {\it the fractal geometry}. In this review we intend to show the basic ideas behind this new approach for the galaxy clustering problem. We will not present the orthodox traditional view since it can be easily found, for instance, in Peebles (1980, 1993) and Davis (1997). Therefore, we shall concentrate ourselves in the challenging voice based on a new viewpoint about the statistical characterization of galaxy clustering, whose results go against many traditional expectations, and which keep open the possibility that the universe never becomes observationally homogeneous. The basic papers where this fractal view for the distribution of galaxies can be found are relatively recent. Most of what will be presented here is based on Pietronero (1987), Pietronero, Montuori and Sylos Labini (1997), and on the comprehensive reviews by Coleman and Pietronero (1992), and Sylos Labini, Montuori and Pietronero (1998). The plan of the paper is as follows. In section 2 we present a brief, but general, introduction to fractals, which emphasizes their empirical side and applications, but without neglecting their basic mathematical concepts. Section \ref{Distr Galaxies} briefly presents the basic current analysis of the large scale distribution of galaxies, its difficulties and, finally, Pietronero-Wertz's single fractal (hierarchical) model that proposes an alternative point of view for describing and analysing this distribution, as well as some of the consequences of such an approach. The paper finishes with a discussion on some aspects of the current controversy about the fractal approach for describing the distribution of galaxies. | 98 | 3 | astro-ph9803218_arXiv.txt |
||
9803 | astro-ph9803133_arXiv.txt | \input abstract | 98 | 3 | astro-ph9803133_arXiv.txt |
||
9803 | astro-ph9803305_arXiv.txt | We present an $R$-band and $J$-band photometry of an optical transient which is likely to be associated with the gamma-ray burst event GRB\,971214. Our first measurement took place 13 hours after the gamma-ray event. The brightness decayed with a power-law exponent $\alpha = -1.20 \pm 0.02$, which is similar to those of GRB\,970228 and GRB\,970508 which had exponents of $\alpha = -1.10 \pm 0.04$ and $\alpha = -1.141\pm 0.014$ respectively. The transient decayed monotonically during the first four days following the gamma-ray event in contrast with the optical transient associated with GRB\,970508 which increased in brightness, peaking two days after the burst, before settling to a power-law decay. | The launch of the BeppoSAX satellite \cite{boel97} in 1996 has led to a recent breakthrough in the study of gamma-ray bursts by detecting fading X-ray counterparts and localizing them to a few arc-minutes on the sky. This has allowed subsequent identification of optical counterparts in three cases, GRB\,970228 \cite{groo97}, GRB\,970508 \cite{bond97}, and GRB\,971214 \cite{halp97}. GRB\,971214 triggered the BeppoSAX Gamma Ray Burst Monitor on December 14.97272 UT with a peak flux of 650 counts s$^{-1}$. In addition, the burst was localized to an error radius of \decmin{3}{9} (99\% confidence) with the Wide Field Camera \cite{heis97}. The gamma-ray event was also detected by BATSE which measured a total fluence above 20 keV of $1.09 \pm 0.07 \times 10^{-5}$ erg cm$^{-2}$ and by one RXTE-ASM camera yielding a peak intensity of $470 \pm 140$ mCrab \cite{kipp97}. Shortly thereafter, a fading optical source was detected within the BeppoSAX error circle at $\alpha(J2000)={\rm11^h56^m}$\decsectim{26}{4}, $\delta(J2000)={\rm65^\circ12'}$\decsec{00}{5} with I-band magnitudes of $21.2 \pm 0.3$ on Dec 15.47 UT and $\sim 22.6$ on Dec 16.47 UT \cite{halp97}. Further observations by BeppoSAX detected a previously unknown fading X-ray source, later designated 1SAX J1156.4+6513, within the initial error circle at $\alpha(J2000)={\rm11^h56^m}25^{\rm s}$, $\delta(J2000)={\rm65^\circ13'}11''$ with an error radius of about $1'$ \cite{anto97}. Since this second X-ray detection is consistent with the position of the fading optical source identified by Halpern \etal\ (1997) it is quite likely that these objects are the X-ray and optical afterglow of GRB\,971214. We report here $R$-band and $J$-band observations of this optical transient (OT) made with the Apache Point Observatory(APO) 3.5 m telescope. This work extends the preliminary photometry reported in Diercks \etal\ (1997), Castander \etal\ (1997), and Tanvir \etal\ (1997). | The well-observed light curve of the GRB\,970508 OT, the brightest observed thus far, shows a dramatic rise, peaking nearly two days after the initial burst, before beginning a power-law decay. GRB\,970228 was not observed nearly as often through a consistent filter, but there is also evidence (depending on spectral assumptions) \cite{guar97} that the transient increased in brightness until $\sim20$ hr after the burst, after which it also began fading with a power-law slope essentially identical to that of GRB\,970508. Despite the same decay slope, the GRB\,970228 OT was $\sim1.5$ mag fainter than the GRB\,970508 OT. A detailed analysis of the difference between these two light curves is presented in Pedersen et al. (1998). A power-law of the form $F = F_{0}t^{\alpha}$ was fit to the four $R$-band data points yielding $\alpha = -1.20 \pm 0.02$. This rate of decline is similar to the two previously identified bursts although there is no evidence of the OT brightening over the course of the observations. \placefigure{fig-2} The observations from all four nights are combined into one deep image totaling 3.25 hrs of integration with a limiting $R$-band magnitude $\sim25.4$ (Figure~\ref{fig-3}). The two brightest objects within 20$''$ of the OT are an extended source (A) \decsec{4}{6} southwest of the transient with $R = 22.7 \pm 0.1$, and an unresolved source (B) \decsec{4}{9} north of the transient with $R = 23.43 \pm 0.08$. The resolution of these images is insufficient to identify any structure in the extended object. \footnote{To obtain the images discussed in this work, contact A. Diercks or E. Deutsch} | 98 | 3 | astro-ph9803305_arXiv.txt |
9803 | astro-ph9803075_arXiv.txt | We report on two observations of the Seyfert galaxy IRAS18325-5926 made in 1997 December and 1998 February with the Rossi X-ray Timing Explorer (RXTE). We find evidence for periodicities in the resulting X-ray lightcurves which are shorter than the 58~ks period found from data of the source taken in 1997 March with the imaging satellite ASCA. It is therefore likely that IRAS18325-5926 has a quasi-periodic oscillation (QPO) similar to, but at a much longer period than, the QPO seen in some Galactic Black Hole Candidates. The power spectrum of the February data has several peaks, the second highest of which is consistent with a monotonic decrease in the X-ray period. The period change is then consistent with that expected from two massive black holes spiralling together due to the emission of gravitational radiation. This possibility is very unlikely but mentioned because of its potential importance. | We recently discovered a 16~hr X-ray periodicity in the Seyfert galaxy IRAS~18325-5926 using data from ASCA (Iwasawa et al 1998). Nearly 9 cycles of the oscillation were observed, with a total amplitude of about 15 per cent. The active nucleus has similar properties to that of Seyfert 1 galaxies, except that the power-law continuum is slightly steeper than most (photon index $\Gamma\sim 2.1$) and there is moderate absorption by a column density of $\sim 10^{22}\pcmsq$. There is also a broad iron line in the X-ray spectrum (Iwasawa et al 1996) which indicates the presence of an accretion disk in the X-ray emission region, viewed at moderate inclination (40--50 deg). The periodicity is plausible for Keplerian motion at 10--20 gravitational radii around a black hole of mass $2\times 10^8 - 2\times 10^7\Msun$. It also scales well with quasi-periodic oscillations (QPO) seen from some Galactic Black Hole Candidates (BHC; Belloni et al 1997). The cause of the oscillation in IRAS~18325-5926, or indeed in any BHC, is unknown. In order to determine whether the variation is exactly periodic or a QPO, we observed IRAS~18325-5926 with the Rossi X-ray Timing Explorer in 1997 December 25--27 and 1998 February 21--23. We report here the light curves and power spectra of those observations, which also show oscillations, but of different periods. We conclude that the AGN has a clear QPO signal and note that our results are consistent with the exciting possibility that 2 black holes are rapidly spiralling together. | \begin{table} \begin{center} \caption{Mean flux and period for X-ray observations of IRAS~18325-5926. The Ginga data are reported by Iwasawa et al (1995). The error bars indicate the period change over which the power drops by a factor of two from the peak. 1998a and b refer to the highest and next highest peaks in the February power spectrum, respectively.} \begin{tabular}{ccr} Detector & Flux (4--10~keV) & Period \\ & $\ergpcmsqps$ & ks \\[5pt] Ginga 1989 & $1.5\times 10^{-11}$ & $>30$ \\ ASCA\ 1994 & $5.6\times 10^{-12}$ & $>80$ \\ ASCA\ 1997 & $1.0\times 10^{-11}$ & $58.0^{+2.6}_{-1.8}$ \\ RXTE\ 1997 & $2.6\times10^{-11}$ & $38.7^{+4.2}_{-2.9}$ \\ RXTE\ 1998a & $2.1\times 10^{-11}$ & $40.3^{+4.5}_{-2.7}$ \\ RXTE\ 1998b & $2.1\times 10^{-11}$ & $28.2^{+1.9}_{-1.4}$ \\ \end{tabular} \end{center} \end{table} The X-ray emission from IRAS~18325-5926 oscillates in a manner similar to that seen in BHC QPO. Such a large clear oscillation from matter around a black hole may suggest that we are detecting the fundamental frequency of some space-filling corona above the disk. This may be possible if the corona has a proton temperature close to the virial value and a lower electron temperature (see e.g. Di Matteo, Blackman \& Fabian 1997). Of course this must happen over a very restricted range of radii in order that a single dominant oscillation is seen. Perhaps it corresponds to the radius where the surface emission from the disk peaks (i.e. $\sim$7 Schwarzschild radii for a non-spinning hole). The variation of period with flux (Fig.~5) is similar to that seen in some QPO. Otherwise, as mentioned by Iwasawa et al (1998), it could be due to the Bardeen-Petterson effect if the angular momentum vectors of the black hole and accreting matter are not aligned. A range of radii are then selected over which the accretion disk tilts over to match the equatorial plane of the black hole. (The disk is not actually precessing in this case; Markovic \& Lamb 1998.) Reflection/obscuration by blobs of gas in the tilt zone might then lead to observed flux variations, especially if the inclination is fairly high (as the broad iron line may indicate; Iwasawa et al 1996), but we consider it unlikely that 50 per cent variations can be obtained in this way. \begin{figure} \centerline{\psfig{figure=pflux.ps,width=0.48\textwidth,angle=270}} \caption{Variation of period with 4--10~keV flux. The two highest power spectrum peaks are shown for February.} \end{figure} The data are consistent with a continuous decrease of period with time from GINGA observations in 1989 to December 1997, and continuing if the second highest peak in the February power spectrum is used. This raises the possibility that the period is due to a massive object orbiting the black hole (see Cunningham \& Bardeen 1973). The separation of the objects requires that this second object is compact. The period decrease is then explained as due to orbital energy loss by the emission of gravitational waves (as for the binary pulsar). The rate of period change $$\dot P ={{-96}\over5}{{G^{5/3}}\over c^5}{{M^{2/3}\mu(4\pi^2)^{4/3}}\over P^{5/3}},$$ where $M=M_1+M_2$, the sum of the separate masses, and $\mu=M_1M_2/(M_1+M_2).$ This integrates to give $P\propto(t_0-t)^{3/8}$ where $t_0$ is the time when the masses finally merge. We have fitted this last relation to the data and find an acceptable fit (Fig.~6). It suggests that the merger may occur in late April 1998. The rate of spiral-in enables us to estimate that the product $M^{2/3}\mu\approx 1.5\times 10^{10}$ where the masses are in units of $\Msun$. This means that for $M_1\sim 2\times 10^6\Msun - 10^8\Msun$, $M_2>10^5\Msun$. There is no solution for $M_1<1.5\times 10^6\Msun$. Much of the energy from such a merger would emerge as gravitational waves, but some is likely throughout the electromagnetic spectrum. We note that it is {\it a priori} unlikely to find an object close in time to a spiral-in merger event. Possibilities that can enhance that probability are if a) black holes are built out of many merger events where the basic unit has a mass of only a few $10^5\Msun$ (say from dwarf galaxies) and b) the inspiralling black hole switches on, or significantly enhances, an otherwise quiescent active nucleus. If we make the extreme argument that this process happens in all galaxies of space density $3\times10^{-3}n_{-2.5}\Mpc^{-3}$, all of which have a central black hole of mass $5\times10^7M_{7.5}\Msun$ growing by the addition of smaller black holes of $10^5m_5\Msun$, then a merger will take place within $120\Mpc$ (the distance to IRAS18325-5926) every $1000m_5 M_{7.5}^{-1} n_{-2.5}^{-1}\yr$. We therefore see that such an event is not completely improbable but is unlikely. The probability that we should find the signal of a merger in its last year (it was the ASCA result of an observation in 1997 March which alerted us to make the RXTE observations) is $\sim 10^{-3}$. Such mergers would of course be more common at fainter flux levels ($\sim 1\yr^{-1}$ within 1 Gpc), and the prospects for space-based gravitational wave astronomy (which will be sensitive to events in massive black holes) could be very positive. A lower mass black hole captured by a central black hole is likely initially to have a highly eccentric orbit (Sigurdsson 1997). When the eccentricity $e$ is high, the above estimates for $\dot P$ are dramatically increased (Shapiro \& Teukolsky 1983), for example by a factor of 100--1000 when $e\sim 0.8-0.9$, respectively. This allows $M_2$ to be much lower than estimated above for a circular orbit (it can be as low as about $100\Msun$). The overall temporal behaviour of the system is then more complex ($e$ decreases with time). It may be difficult for a low mass black hole to modulate the observed X-ray emission greatly; this problem is offset by the enhanced probability of such an event occurring. We stress that this last, exciting, interpretation involving an in-spiralling black hole is unlikely and depends upon the identification of the second highest power peak in the February 1998 data with the orbital period of the second black hole. It does not explain the other peaks in the power spectrum nor why some peaks expected in the lightcurve do not occur. We discuss it here only because of its potential importance. Clearly further observations are urgently required. Even if such observations fail to show a consistently decreasing period, IRAS1832-5926 has a convincing, high-amplitude, QPO, which requires explanation. \begin{figure} \centerline{\psfig{figure=spiral_in_new.ps,width=0.48\textwidth,angle=270}} \caption{The peak period displayed against time in days, measured backwards from the February 1998 observation. Error bars show where the peak power has decreased by about a factor of 2. The line shows the best-fitting relation $P=6203.(t+63.5)^{3/8}\s$. For the February data we plot both the period with the peak power and the longer one which had slightly less power.} \end{figure} | 98 | 3 | astro-ph9803075_arXiv.txt |
9803 | astro-ph9803243_arXiv.txt | We report the discovery of two low redshift HI 21cm absorbers, one at $z = 0.2212$ towards the $z_{em} = 0.630$ quasar OI 363 (B0738+313), and the other at $z = 0.3127$ towards PKS B1127-145 ($z_{em} = 1.187$). Both were found during a survey of MgII selected systems at redshifts $0.2 < z < 1$ using the new UHF-high system at the Westerbork Synthesis Radio Telescope (WSRT). New HST/FOS observations also identify both systems as damped Ly$\alpha$ (DLa) absorbers. By comparing the column density from the DLa line with that from the HI 21cm line, we calculate the spin temperature, T$_s$ of the two systems. We find $T_s \approx 1000$ K for both of these low redshift absorbers. For the $z = 0.3127$ system towards PKS B1127-145, two galaxies have been previously identified with emission lines at the absorber redshift (Bergeron \& Boiss\'e, 1991), with the galaxy at a closer projected distance to the quasar assumed to be responsible for the absorption system. An ESO-NTT/EFOSC2 spectrum of a 3rd, fainter companion at 3.9 arcsec or 11 $h^{-1}_{100}kpc$ from the line of sight of PKS 1127-145 reveals [OIII]4958 and 5007 at $z = 0.3121 \pm 0.0003$. We consider this object the most likely to be responsible for the 21cm absorption, as it is much closer to the QSO sightline than the two galaxies identified by Bergeron \& Boiss\'e. \end {abstract} | The study of low redshift examples of the quasar absorption line systems responsible for the damped Ly$\alpha$ (DLa) and HI 21cm absorption lines is important to help bridge our understanding of neutral gas-rich systems between those at redshift $z = 0$ and those at high $z$. Our knowledge of the neutral gas in nearby spiral galaxies is mainly based on observations of the HI 21cm line in emission. At high redshift, however, we observe the HI 21cm line in absorption, which can only be seen along a limited number of lines of sight through the intervening absorber, making detailed knowledge of the gas characteristics difficult. The low redshift ($z < 1$) neutral absorbers are still close enough that both optical and radio data of reasonable quality can be obtained in order to understand their kinematics and physical gas characteristics. Such information is necessary to build a framework for a correct interpretation of the higher z counterparts to these systems. Searches for redshifted HI 21cm absorption can be time-consuming since radio spectrometers typically observe only relatively narrow instantaneous bandwidths and only the highest column density QSO absorption line systems have measurable optical depths in the 21cm line. Since DLa systems have high column densities of neutral HI, they are the most likely objects to have HI 21cm absorption. Unfortunately for low redshift work, the Ly$\alpha$ line is not shifted into the optical window until $z \simeq 1.65$, so finding these lines requires UV spectra to be taken with space telescopes. This, combined with the small cross section for DLa absorption, means that only a small number of DLa systems have been identified at low redshift. Therefore, alternative selection criteria which are reasonably effective at finding an HI 21cm absorber must be used. All known DLa and HI 21cm absorbers have associated low-ionization metal lines (cf. Lu and Wolfe, 1994) such as the MgII $\lambda\lambda$2796, 2803 doublet, which can be observed easily in ground-based spectra down to about $z = 0.1$. A study of MgII selected systems using previously existing UV data yielded about 1 DLa system per 10 MgII systems observed (Rao, et al., 1995). A similar statistic exists for HI 21cm absorption in MgII systems (Briggs \& Wolfe, 1983). This suggests that MgII can be used to optically select systems likely to have high column densities of neutral gas observed either as DLa or HI 21cm absorption. We are conducting a survey for HI 21cm absorption in low redshift MgII selected systems using the Westerbork Synthesis Radio Telescope (WSRT). In this paper we present two new HI 21cm absorbers from our survey, one towards PKS B1127-145 at $z = 0.3127$ and the other towards OI 363 at $z = 0.2212$. Recent HST/FOS spectra have identified DLa absorption in both of these objects as well. | \subsection{OI 363} OI 363 (0738+313), is a core dominated quasar at $z_{em} = 0.630$. Observations at 1640 MHz (Murphy, et al., 1993) show that the lobes extending $\approx 30\arcsec$ from the core contain only $3\%$ of the total flux of the quasar. The quasar is slightly variable at low frequencies (Bondi, et al., 1996b). The metal line absorption system was originally reported by Boulade, et al. (1987) at a redshift of $z = 0.2213$, and subsequently by Boiss\'e, et al. (1992) at a redshift of $z = 0.2216 \pm 0.0003$. The only identified lines in the spectrum are the MgII $\lambda\lambda2796,2803$ doublet and a possible MgI line. Deep optical imaging was made by LeBrun et al. (1993), who identified what they considered a likely absorber at a projected separation from the quasar of 5.70'', or $1.24R_H$ if it lies at $z = 0.221$. They identified three additional galaxies, at smaller angular separations from the quasar, which are very faint and would be dwarf galaxies at the absorption redshift. Unfortunately, they do not report a confirmed redshift for any object in the field near the quasar. The HI 21cm absorption line, shown in fig. 1, has a narrow width of only two channels over most of its depth, and may be unresolved by this observation. This implies an upper limit to the line width of $\sim8$ km s$^{-1}$ at the 6 km s$^{-1}$ resolution of the data. The HI column density from the DLa profile is $N(HI) = 7.9\pm1.4 \times 10^{20}$ cm$^{-2}$, and the calculated mean harmonic spin temperature is T$_s = 1230 \pm 335$ K. The termal kinetic temperature of the gas for a line of width 8 km s$^{-1}$ is T$_k = 1400$ K. This means that within the errors, T$_s = $T$_k$. \subsection{PKS B1127-145} PKS B1127-145 is a compact, gigahertz peaked radio source at $z = 1.187$. VLBI observations at 1670 MHz show a slightly elongated structure with an extent of approximately 20 mas. Observations at 408 MHz give a flux variability of 1.2 Jy/year (Bondi, et al., 1996a), and indicate structural variations as well. Fig. 2 shows the neutral HI 21cm absorption for this system. The observed flux of the quasar is 5.25 Jy in this observation. This is somewhat lower than reported fluxes in NED\footnote{The NASA/IPAC Extragalactic Database (NED) is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration.}, however, as discussed above the quasar is a known variable at low frequency. If the absorption line is fit by one gaussian component, the optical depth of the absorption is $6.2\%$, and the FWHM is approximately 42 km s$^{-1}$. However, there is some evidence for structure in the line. In particular, the split in the middle appears to be real, suggesting at least two components, and the asymmetric low frequency side of the profile may result from a third component. There is some low level interference in the spectrum adjacent to the low frequency side of the absorption line, at $\sim$1081.75 MHz, which makes interpretation of the real shape difficult. The HI column density from the DLa profile is $N(HI) = 5.1\pm0.9 \times 10^{21}$ cm$^{-2}$, and the calculated T$_s = 1000 \pm 200$ K. Bergeron and Boiss\'e (1991) studied this $z_{abs}=0.313$ metal-line system in some detail. They identify MgII, FeII and MgI absorption in the spectrum. The average redshift of the metal lines is $z = 0.3127 \pm 0.0002$. In addition to the absorption spectrum, and deep images of the field, they present emission spectra for two of the bigger bright galaxies near the quasar (different from the two faint companions discussed in Sect. 4), both of which have redshifts similar to that of the metal line system. They identify the galaxy at the smaller projected distance from the quasar sightline as the absorber. The ESO/EFOSC2 spectral observations (discussed in Sect. 3) identify a new candidate absorber galaxy for this system. The new emission object is a close companion to the west of the quasar and lies at a projected distance of $3.9\arcsec$ or 11 $h^{-1}_{100}$kpc from the quasar sightline. Based on comparison with other objects in our image for which apparent magnitudes are given in Bergeron and Boiss\'e (1991), it has an apparent magnitude of $m_r = 22.3$. The galaxy which Bergeron and Boiss\'e identify as the absorber is at a projected separation of $9.6\arcsec$, or 37 $h^{-1}_{100}$kpc from the quasar at the redshift of the galaxy. This corresponds to a galactic radius of $2.7 R_H$. A column density of neutral gas of $10^{21}$ cm$^{-2}$ is unlikely at this galactic radius, and we consider the new emission object, although smaller and fainter, to be the more likely absorber given its proximity to the quasar sightline. It is also possible that the three galaxies at similar redshift have undergone strong interaction with each other, in which case the absorbing gas could be tidal debris. \subsection{Spin Temperature} Fig. 4 shows T$_s$ vs. redshift for all of the known HI 21cm and DLa absorber systems, calculated from the literature (as summarized by Carilli, et al., 1996), with the two new data points marked by open symbols. The shaded region shows the range of Galactic T$_s$ values at optical depths comparable to those found in the DLa systems, using numbers from Braun and Walterbos (1992). The large errorbars in any given measurement are dominated by uncertainties in the true optical continuum level, due to confusion from the Ly$\alpha$ forest lines, which make fitting the damped profile difficult. As noted in previous studies (cf. de Bruyn, et al. 1996), all of the redshifted absorbers except the highest optical depth (lowest T$_s$) system have T$_s$ values which are roughly two or more times greater than the Galactic values at similar optical depths. Most estimates of T$_s$ for the present epoch have been based on studies of the Milky Way or Andromeda (cf. Dickey and Brinks (1988); Braun and Walterbos (1992), and references therein). These estimates use column densities derived from HI 21cm emission lines rather than from DLa absorption lines, but studies have shown that N(HI)$_{Ly\alpha} \approx$ N(HI)$_{21cm}$ (Dickey and Lockman, 1990) . This suggests that T$_s$ values for the galaxy and the redshifted DLa systems, although calculated from different quantities, can be compared. On the other hand, values for T$_s$ have a strong correlation with the column density, N(HI), in each of the clouds along the line of sight in the Galaxy, and are thus sensitive to the location of individual gas clouds. A single cloud usually has a greater angular size on the sky than the beam with which it is observed, so it is possible to calculate the spin temperature for just one cloud. This is not the case in redshifted systems, where a line of sight includes more of the galaxy and possibly many clouds. Thus it is difficult to usefully compare present epoch spin temperature values to those calculated at higher redshifts. The new systems in this study were observed at low redshift with the same line of sight limitations as the higher redshift systems, allowing a meaningful comparison. There is no clear trend for a change in T$_s$ with increasing redshift among the eight systems shown in Fig. 4. If the gap between T$_s$ values in the galaxy and those in the DLa systems is an effect of evolution over time, then all of that evolution must have occured between $z = 0.2$ and the present epoch. Instead, we consider it more likely that the gap arises from the presence of many clouds in one line of sight at high redshift, or from differences in the radio and optical sightlines. The sizes of the radio emission regions of quasars are much larger than the optical regions, and usually larger than an average cloud as well. It is therefore likely that the optical and radio lines of sight actually sense different clouds in a redshifted galaxy, and hence have different column densities of neutral gas. This implies that the spin temperatures derived by assuming both column densities are equal may be meaningless. Unfortunately, the resolution obtained in most radio survey observations (with single dishes or synthesized beams from arrays like the WSRT) is at least as large on the sky as compact background radio sources, and gives no spatial information about the clouds in front of the quasar. Future use of VLBI techniques to pinpoint the radio sightlines more accurately may help to clarify this problem. | 98 | 3 | astro-ph9803243_arXiv.txt |
9803 | hep-ph9803295_arXiv.txt | The effects of a possible rotation of the galactic dark halo on the calculation of the direct detection rates for particle dark matter are analyzed, with special attention to the extraction of the upper limits on the WIMP--nucleon scalar cross section from the experimental data. We employ a model of dark halo rotation which describes the maximal possible effects. For WIMP masses above 50 GeV, the upper limit exclusion plot is modified by less than a factor of two when rotation is included. For lighter masses the effect can be stronger, suggesting the necessity to develop specific models of halo rotation in order to provide more accurate conclusions. | The possibility to detect Weakly Interacting Massive Particles (WIMPs) distributed in the halo of our Galaxy has been a major issue in the last years, since these particles could provide the amount of dark matter necessary to explain many observed dynamical properties of galaxies, clusters and of the Universe itself. Different kinds of possible signals have been identified and looked for, in order to outline the presence of WIMPs in our Galaxy. These signals are usually referred to as ``direct" and ``indirect" detection rates. Direct detection refers to the possibility to measure a WIMP--nucleus interaction in a low--background detector, while indirect detection relies on the measurement of WIMP annihilation products: photons, antiprotons and positrons produced by the annihilation in the galactic halo, or neutrinos coming out of the Earth or the Sun where WIMPs may have been accumulated as a consequence of gravitational capture. It is remarkable that the present sensitivity of the different experiments is already at the level of the predicted rates for specific WIMP candidates, like the neutralino, which represents one of the most interesting and studied cold relic particles \cite{noi}. The calculation of the different detection rates depends not only on the particle physics properties of the WIMPs interactions, but also on the characteristics of the galactic halo where the WIMPs are distributed. Direct detection rates and upgoing-muon fluxes at neutrino telescopes, which both rely on the WIMP elastic scattering off nuclei, depend on the WIMP matter density $\rho_\odot$ and velocity distribution $f_\odot(v)$ at the Earth position $r_\odot$ in the Galaxy. In particular, the dependence of the signals on $\rho_\odot$ is a linear one. The other indirect signals (photon, antiproton and positron fluxes) have a stronger dependence on the matter distribution, since they are proportional to the square of the matter distribution function (DF) $\rho(\vec r)$ integrated over the effective region of production and propagation of the annihilation products. On the contrary, this kind of signals are essentially independent on the details of the velocity DF, since the annihilating WIMPs are almost at rest and corrections due to their velocity dispersion are negligible. Detailed estimates of the detection rates would require specific and accurate models of the galactic halo able to provide a reliable WIMPs DF $g(\vec x, \vec v)$ (not necessarily separable in phase space, i.e. $g(\vec r, \vec v) = \rho(\vec r) f(\vec v)$). Unfortunately, detailed halo models are not available at present, mainly because the constraints obtained from astrophysical observations are not stringent enough to restrict different possibilities. The most important observational constraint is provided by the flatness of the rotation curves at large radii. Although the available data on our Galaxy do not provide a compelling evidence of a flat rotation curve, this feature is observed in a large number of spiral galaxies and therefore it looks reasonable to assume its validity also for our Galaxy. The standard and simplest model of the dark galactic halo, which is compatible with a flat rotation curve, is the so--called isothermal sphere. This model relies on the two basic assumptions of spherical symmetry and thermal equilibrium, which find a strong support in the argument of ``violent relaxation" introduced by Lynden--Bell 30 years ago\cite{lynden-bell}. In this model, the DF is separable into a matter density distribution $\rho(r)$, which has a $r^{-2}$ behaviour at large radii, and into a Maxwell--Boltzmann (MB) velocity DF $f(v)$ \cite{binney}. Although such a model gives a divergent total mass and therefore an appropriate cut-off has to be introduced at large radii, its range of validity has been tested at least in the inner parts of many galactic systems. Moreover, since it represents a simple and reasonable approximation, in the absence of a more detailed model it is widely adopted to describe the dark halo of our Galaxy. However, many different models are known to be consistent with flat rotational curves. For instance, models which describe non--spherically symmetric or flattened halo distributions have been discussed \cite{binney}. In these models, the specific form of $\rho(\vec r)$ differs from the standard isothermal sphere matter DF, especially at small radii, entailing quite large uncertainties on the local value $\rho_\odot$. A comprehensive numerical study which takes into account a large number of models indicates that the local value of the non--baryonic dark matter density falls in the (rather conservative) range 0.1 $\lsim \rho_\odot \lsim$ 0.7 GeV cm$^{-3}$\cite{turner}. Contrary to the matter DF, the specific form of the velocity DF $f(v)$ has been much less investigated. Modifications to the standard MB velocity DF are known \cite{binney,evans}, but the problem of determining the correct form of the distribution of the WIMP velocities in the halo has no clear and simple solution at present, both theoretically and observationally. The velocity DF is required to be consistent with a given $\rho(\vec r)$ but this, in general, does not determine $f(v)$ in a unique way. The calculation of the WIMP detection rates is usually performed by using the standard isothermal sphere model. However, modifications in the isothermal model can affect the detection rates, introducing uncertainties in the theoretical predictions and in the extraction of the experimental limits on the WIMPs parameters. The effects induced on the detection rates by a modification in the matter DF are simple to take into account, since the dependence of the detection rates on $\rho(\vec r)$ can be factorized. Specifically, the physical range of $\rho_\odot$ quoted above implies an uncertainty of about a factor of 7 in the evaluation of the direct detection rates and in the neutrino fluxes \cite{noi} (it has to be remarked that this large factor reflects a rather conservative attitude). Even larger uncertainties affect the indirect rates from WIMP annihilation in the halo, since in this case a modification in the matter density profile can strongly affect the integral of $\rho^2(\vec r)$ over the effective production region of the signal \cite{pbar_to,pbar_japan,gamma}. Contrary to the case of the matter DF, a modification of the standard MB velocity DF would affect the direct detection rates and the indirect rates at neutrino telescopes in a much more involved way. This is because the dependence of these rates on $f(v)$ is through a convolution of $f(v)$ with the differential WIMP--nucleus cross section. Since the WIMP--nucleus scattering depends on the relative velocity of the WIMPs with respect to the detector nuclei, a potentially significant effect could be due to a bulk rotation of the halo. This would necessarily modify the WIMP phase--space DF with respect to the standard MB form. In this paper we wish to discuss the possible effects induced by a halo rotation on the direct detection rates, with special attention to the ensuing consequences on the determination of the upper limits on the WIMP--nucleus cross section from the experimental data. A calculation of the direct detection rates in the case of a rotating halo has been addressed in Ref. \cite{kamion}, where it has been concluded that the maximal effect of rotation leads to a 30\% effect on the total detection rates for a Ge nucleus, in the case of an ideal detector with no threshold. However, when considering a real detector the behaviour of the differential rates at threshold and the detector characteristics are crucial in determining the experimental limits on the WIMP--nucleus cross section \cite{noi}. Therefore, we explicitly take into account the features of running detectors, such as thresholds, quenching factors and energy resolution, in order to estimate the largest uncertainties induced by a possible halo rotation in a confident way. To this aim, following Ref.\cite{kamion} we model the galactic rotation as described by Lynden--Bell in Ref.\cite{lynden-bell2}, where the maximally rotating velocity DF compatible with a given mass distribution has been derived, on the ground of purely kinematical arguments. Even if Lynden--Bell's model of halo rotation may not represent a situation which is realized in a physical halo, we consider it useful to bracket the size of the effect of halo rotation on the direct detection rates. The plan of our paper is the following. In Sect.II we briefly describe the calculation of the direct detection rates in the presence of halo rotation. In Sect.III we discuss our results for Ge, NaI and Xe detectors, taking into account the most recent experimental data of the different Collaborations. Finally, in Sect.IV we draw our conclusions. An Appendix is added, where we report the analytical expressions of the relevant part of the direct detection rates which contain the details of the velocity DF in the case of the standard non--rotating, maximally co--rotating and maximally counter--rotating haloes. | In this paper we have investigated the effect induced by a possible rotation of the galactic halo on the rates of WIMP direct detection. In particular, we have discussed the implication of halo rotation on the determination of the exclusion plots on the WIMP--nucleon cross section for different detectors, namely Ge, NaI and Xe ones. The rotation of the halo has been described by using a model \cite{lynden-bell2} which corresponds to a situation where the halo possesses the maximal rotation compatible with a given mass DF, which for simplicity we have chosen to be that of the isothermal sphere. We found that the exclusion plots obtained from the data are affected by less than a factor of 2 in the case of counter--rotating models. The same size of uncertainty occurs also for the co--rotating models, when the WIMP mass is larger than about 50 GeV. For lighter WIMPs and co--rotation, the exclusion plots are modified by a larger amount. We have to remind that, due to the particular model of halo rotation which we have employed here, these are expected to be maximal effects. For specific physical rotation models, the effect of halo rotation will be plausibly smaller. We can therefore conclude that, at least for WIMP masses greater than about 50 GeV, the determination of the exclusion plots from the experimental data are affected by an uncertainty smaller than a factor of two due to the possibility that the galactic halo rotates, independently on the specific model of halo rotation. We notice that recent preliminary data from accelerators indicate that the lower limit on the mass of the most plausible WIMP candidate, the neutralino, is $m_\chi\simeq 30$ GeV for low value of the susy parameter $\tan\beta$, and $m_\chi\simeq 45$ GeV for $\tan\beta\gsim 3$ \cite{lep183}. Therefore, for this dark matter candidate, the uncertainty on the exclusion plot due to a possible rotation of the halo is expected to be relatively small. The situation is different for lighter WIMPs. In this case, it would be required to develop specific models of halo rotation in order to obtain more accurate conclusions. | 98 | 3 | hep-ph9803295_arXiv.txt |
9803 | astro-ph9803239_arXiv.txt | We report the discovery of a 2.1hr optical modulation in the transient source GS1826-24, based on two independent high time-resolution photometric observing runs. There is additional irregular variability on shorter timescales. The source also exhibited an optical burst during each observation, with peak fluxes consistent with those of the three X-ray bursts so far detected by {\it Beppo}SAX. We compare the low-amplitude variation ($\sim 0.06^m$) to that seen on the orbital periods of the short period X-ray bursters, X1636-536 and X1735-444, as well as the similarity in their non-periodic fluctuations. Other transient neutron star LMXBs possess short periods in the range 3.8-7.1 hrs. However, if confirmed as the orbital, a 2.1 hr modulation would make GS1826-24 unique and therefore of great interest within the context of their formation and evolution. | GS 1826-24 was discovered serendipitously in 1988 by the {\it Ginga} LAC during a satellite manoeuvre (Makino {\it et al.} 1988). The source had an average flux level of 26 mCrab (1-40 keV), and a power law spectrum with $\alpha = 1.7$. Observations both a month before and after by the {\it Ginga} ASM, by TTM in 1989 \cite{intZ92} and by ROSAT in 1990 and 1992 \cite{barr95} found comparable flux levels. Temporal analyses of both the {\it Ginga} detection and ROSAT data yielded a featureless $f^{-1}$ power spectrum extending from $10^{-4} - 500$ Hz \cite{tan95,barr95}, with neither QPO nor pulses being detected. Despite its detection by {\it Ginga}, the source had not been previously catalogued. Neither were X-ray bursts detected by {\it Ginga}. Together with its similarities to Cyg X-1 (hard X-ray spectrum, strong flickering), this led to an early suggestion by \scite{tana89} that it was a soft X-ray transient with a possible black-hole primary. Later, \scite{stri96} called this suggestion into doubt, following examination of data from CGRO/OSSE observations. They found that fitting both the Ginga and OSSE spectra produced a model with an exponentially cutoff power law plus reflection term. The observed cut-off energy of $\sim$58 keV is typical of the cooler neutron star hard X-ray spectra. The recent report of three X-ray bursts detected by {\it Beppo}SAX \cite{uber97} and our detection of optical bursts here confirms the presence of a neutron star accretor. Following the first ROSAT/PSPC all-sky survey observations in September 1990, and the determination of a preliminary X-ray position, a search for the counterpart yielded a time variable, UV-excess, emission line star \cite{motc94,barr95}. This source had $B= 19.7 \pm 0.1$, and an uncertain V magnitude of $V \simeq 19.3$, due to contamination by a nearby star. There was also evidence for $\simeq 0.3^m$ variations on a one hour timescale, but the time sampling was fairly poor. For this reason, we included this object in our target list for a high-speed photometry run at the South African Astronomical Observatory (SAAO). Further time-series photometry was also obtained on the William Herschel Telescope, La Palma (WHT) to confirm the variability that was seen. | \begin{table*} \begin{minipage}{150mm} \caption{Properties of GS1826-24 and the short-period transient bursters$^a$\label{tab:sptb}} \footnotetext[1]{Data taken from \scite{vP95} and references therein, unless cited elsewhere.} \begin{tabular}{c c c l c l l l } \hline Source & $P_orb$ & V & B-V, U-B & E${\rm _{B-V}}$ & F$_X$ & Active &Quiescent \\ & (hr) & & & & ($\mu$Jy)& & \\ \hline GS1826-24& 2\footnote[2]{This work.} &19.3 \footnote[3]{\scite{motc94}.} & 0.4$^c$, -0.5$^c$ & 0.4$^c$ & 30$^c$& 1988-present\footnote[4]{\scite{mak88}.} & before 1988\\ & & & & & &&\\ X0748-678& 3.82 &$>$23\footnote[5]{\scite{wad85}.}-16.9&0.1, -0.9 &0.42 & 0.1-60 & 1985-present\footnote[6]{\scite{whi95}.} &before 1985\\ & & & & & & \\ X2129+470 & 5.24 &16.4-17.5& 0.65,-0.3 &0.5 & 9 & $<$1979 - 1983$^f$ & 1983-present\\ & & & & & & &\\ X1658-298 & 7.11& 18.3 &0.45,-0.4 & 0.3 & $<$5-80& pre-1980s $^f$ & 1980-present \\ & & & & & &&\\ \end{tabular} \end{minipage} \end{table*} The period distribution of low-mass X-ray binaries (LMXB) has shown a scarcity of systems below 3 hr presumably in part due to their faintness, but there is a notable absence of any systems with periods between \til 1 and 3 hr \cite{whi85a,whi85b}. Recently, \scite{king97} have investigated the formation of neutron star LMXBs. They found that in order to produce the relatively large fraction of soft X-ray transients in the $\simlt$ 1-2 day range, the secondaries must have 1.3\msun $\simlt M_2\simlt$ 1.5\msun\ at the onset of mass transfer and be significantly nuclear evolved (provided that the SN kick-velocity is small compared to the pre-SN orbital velocity). This mass range ensures that the mass transfer rates driven by angular momentum loss are below the critical rate needed for SXT behaviour in the required fraction of neutron star LMXBs. The large initial masses then account for the rarity of any short $P \simlt 3$ hr systems. If indeed the 2 hr modulation of GS1826-24 is confirmed as orbital in origin, this would make it of great interest within the context of transient neutron star LMXB formation and evolution. However, in many respects GS1826-24 does show similarities to other neutron star binaries with comparable periods. The X-ray bursters X1636-536 and X1735-444 exhibit optical modulations of a similar amplitude on their orbital periods (3.80 and 4.65 hr respectively), plus irregular variability on somewhat shorter timescales (\pcite{whi95} and references therein), although unlike GS1826-24 these systems are not transient on a timescale of decades. The optical emission of LMXBs is dominated by the reprocessed X-rays from the accretion disc and companion. The origin of the underlying sinusoidal modulation is interpreted as the varying contribution from the X-ray heated face of the companion star \cite{vP95a}. As for the irregular variability, this is probably caused by changes in the spatial distribution of the reprocessing material in the accretion disc and/or fluctuations in the central X-ray luminosity, as suggested in the case of X1636-536 \cite{vP90b}. Moreover, GS1826-24 has a high $L_X/L_{opt}(\sim500)$, very similar to that of the compact 41 min binary X1627-673, but lower than the $L_X/L_{opt}\sim700$ of the 50 min binary X1916-053, which is a higher inclination dipping source (once again these are persistent sources). Since this ratio is related in part to the physical size of the system and the disc reprocessing area available, the similarity supports the hypothesis that GS1826-24 is also a relatively compact system. Furthermore, the non-detection of GS1826-24 prior to 1988 is characteristic of the observed variability of the transient bursting systems X0748-678, X1658-298 and X2129+470 \cite{whi95}, which have either been detected over many years and then gone into quiescence for a similar period of time or vice-versa. These transients all have known periods in the range 3.82-7.1hrs (see Table \ref{tab:sptb}). Clearly, further high-speed optical photometric monitoring is required in order to confirm the stability of the 2hr modulation and hence its orbital origin. With only a short {\it ROSAT} lightcurve published to date \cite{barr95}, confirmation might be possible from a longer X-ray observation. However, the low-amplitude of the observed modulation ($0.06^m$) implies a low-inclination system ($< 70 \degree$), and hence X-ray dipping behaviour is unlikely to be seen. | 98 | 3 | astro-ph9803239_arXiv.txt |
9803 | astro-ph9803149_arXiv.txt | The local luminosity function at 25 $\mu$m provides the basis for interpreting the results of deep mid-infrared surveys planned or in progress with space astrophysics missions including ISO, WIRE and SIRTF. We have selected a sample of 1458 galaxies from the IRAS Faint Source Survey with a flux density limit of 250 mJy at 25 $\mu$m. The local luminosity function is derived using both parametric and non-parametric maximum-likelihood techniques, and the classical $1/V_{max}$ estimator. Comparison of these results shows that the $1/V_{max}$ estimate of the luminosity function is significantly affected by the Local Supercluster. A maximum-likelihood fit to the radial density shows no systematic increase that would be caused by density evolution of the galaxy population. The density fit is used to correct the $1/V_{max}$ estimate. We also demonstrate the high quality and completeness of our sample by a variety of methods. The luminosity function derived from this sample is compared to previously published estimates, showing the prior estimates to have been strongly affected by the Local Supercluster. Our new luminosity function leads to lower estimates of mid-infrared backgrounds and number counts. | \label{sec:intro} Much of the effort to study infrared-luminous galaxies has centered on wavelengths greater than 50 $\mu$m. Modeling work is focused on the near-IR (e.g. \markcite{chok94}Chokshi et al. 1994) and far-IR (e.g. \markcite{hac87}Hacking et al. 1987; \markcite{rrbinson96}Rowan-Robinson et al. 1996) portions of the galaxian spectrum. However, the mid-infrared is well-suited for studying starburst and ultraluminous galaxies. About 40\% of the luminosity from starburst galaxies is radiated from 8-40 $\mu$m (\markcite{soi87}Soifer et al. 1987). Extinction effects are small, and problems due to infrared cirrus are minimized. Most importantly for space astrophysics, for a fixed telescope aperture, the spatial resolution is higher at shorter wavelengths, and the confusion limit lies at higher redshifts. Recent work using the {\it Infrared Space Observatory (ISO)} (e.g. \markcite{knapp96}Knapp et al. 1996; \markcite{boul96}Boulade et al. 1996; \markcite{rrobinson96}Rowan-Robinson et al. 1996) shows the relative importance of the 7 $\mu$m and 15 $\mu$m bands for galaxy studies. The {\it Wide-Field Infrared Explorer (WIRE)}, a Small Explorer mission due to launch in late 1998 (\markcite{hac96}Hacking et al.\ 1996; \markcite{schemb96}Schember et al. 1996), will conduct a very deep survey at 24 $\mu$m to study starburst galaxy evolution. The {\it Space Infrared Telescope Facility (SIRTF)} is also expected to conduct surveys in mid-infrared bands. To interpret the results of these surveys now in progress or soon to commence, it is necessary to better understand the mid-infrared properties of galaxies in the local Universe. The 25 $\mu$m luminosity function provides the basis for predicting the faint source counts in the mid-infrared. The empirical model of \markcite{hac91}Hacking \& Soifer (1991) uses an analytic fit to the luminosity function derived by Soifer \& Neugebauer \markcite{soi91} (1991). This function was estimated from a complete subsample of the Bright Galaxy Sample (\markcite{soi87}Soifer et al.\ 1987) containing 135 galaxies to a flux density limit of 1.26 Jy. The availability of many more redshifts of IRAS galaxies (principally from the 1.2 Jy Survey (\markcite{str90}Strauss et al.\ 1990; \markcite{fis95} Fisher et al.\ 1995)) enables a much larger sample to be studied, reducing uncertainties at high and low luminosities. In this paper we present the selection of a large galaxy sample that is flux-limited at 25 $\mu$m, and derive the local luminosity function based on this sample. The sample selection is described in the next section. In Section \ref{sec:lfresults} we describe the $1/V_{max}$ and the maximum-likelihood estimators for deriving the local luminosity function, and present the results. The completeness of the sample is discussed in Section \ref{sec:compdisc}. In Section \ref{sec:correct} we calculate the radial density distribution of the sample using a maximum-likelihood method. The radial density fit is used to correct the $1/V_{max}$ estimate of the local luminosity function, as well as the redshift distribution with which the luminosity function can be compared. Section 6 includes discussions of the different luminosity function estimators, a comparison of our newly derived luminosity function with previous estimates, the implications for mid-infrared backgrounds and number counts, and the effects of evolution on the derivation of the luminosity function. The color properties of the sample are treated in another paper (\markcite{fang98}Fang et al.\ 1998). | \label{sec:conclusions} The following are the results of this paper: 1. We have selected a sample of 1458 galaxies with redshifts from the IRAS Faint Source Survey with a flux density limit of 250 mJy at 25 $\mu$m. An additional 17 galaxies do not have redshifts available. 2. The local luminosity function is derived using the $1/V_{max}$ estimator and both parametric and non-parametric maximum likelihood methods. The $1/V_{max}$ estimate is significantly affected by the Local Supercluster. The maximum likelihood methods are independent of density variations, and we consider the parametric fit with parameters in Table \ref{tab:fitpars} to be the best estimate of the local luminosity function at 25 $\mu$m. 3. A maximum likelihood fit to the radial density in this sample is used to correct the $1/V_{max}$ estimate. The fit shows no sign of a systematic increase with redshift of the density, as would result from density evolution of the galaxy population. 4. The $1/V_{max}$ luminosity function derived from a smaller sample by Soifer \& Neugebauer (1991) is significantly contaminated by the Local Supercluster. Predictions of number counts and local luminosity density based on that function are 15-20\% higher than those indicated by our improved luminosity function. The new function also leads to lower predictions of the mid-infrared background due to galaxies. | 98 | 3 | astro-ph9803149_arXiv.txt |
9803 | astro-ph9803180_arXiv.txt | We report radial velocities for 99 galaxies with projected positions within $30\arcmin$ of the center of the cluster A3733 obtained with the MEFOS multifiber spectrograph at the 3.6-m ESO telescope. These measurements are combined with 39 redshifts previously published by Stein (1996) to built a collection of 112 galaxy redshifts in the field of A3733, which is used to examine the kinematics and structure of this cluster. We assign cluster membership to 74 galaxies with heliocentric velocities in the interval $10\,500$--$13\,000$ \kms. From this sample of cluster members, we infer a heliocentric systemic velocity for A3733 of $11\,653^{+74}_{-76}$ \kms, which implies a mean cosmological redshift of 0.0380, and a velocity dispersion of $614^{+42}_{-30}$ \kms. The application of statistical substructure tests to a magnitude-limited subset of the latter sample reveals evidence of non-Gaussianity in the distribution of ordered velocities in the form of lighter tails and possible multimodality. Spatial substructure tests do not find, however, any significant clumpiness in the plane of the sky, although the existence of subclustering along the line-of-sight cannot be excluded. | The rapid development of multifiber spectroscopy in recent years has made possible the simultaneous acquisition of large numbers of galaxy spectra. The obtention of extensive and complete redshift data bases for clusters of galaxies has hastened the investigation of the physical properties of their visual component which, in turn, is allowing for a better understanding of the characteristics of the dark matter distribution on Mpc scales. Here, we report a total of 104 redshift measurements for 99 galaxies in the field of A3733 and use these data, in combination with a previously published sample of 39 redshifts, to perform a kinematic and spatial analysis of the central regions of this cluster. A3733 is a southern galaxy cluster listed in the ACO catalog (Abell, Corwin, \& Olowin 1989)\cite{ACO89} as of intermediate Abell's morphological type and richness class $R=1$. This cluster hosts a central cD galaxy, included in the Wall \& Peacock (1985)\cite{WP85} all-sky catalog of brightest extragalactic radio sources at 2.7 GHz, which has led to its classification as of Bautz-Morgan type I--II (Bautz \& Morgan 1970)\cite{BM70}. A3733 is also one of the 107 nearby rich ACO clusters ($R\ge 1$, $z\le 0.1$) included by Katgert et al. (1996)\cite{Ka96} in the ESO Nearby Cluster Survey (ENACS), as well as a one of the X-ray-brightest Abell clusters detected in the ROSAT All-Sky Survey by Ebeling et al. (1996)\cite{Eb96}. The only major kinematical study of A3733 done so far is that of Stein (1997)\cite{St97}. From a sample of 27 cluster members located within $r\la 16\arcmin$ from the cluster center, this author has found no evidence of significant substructure in the cluster core. This study of A3733, which is part of a more general investigation of the frequency of substructure in the cluster cores from an optical spectroscopic survey conducted on a sample of 15 nearby ($0.01\la z\la 0.05$) galaxy clusters (Stein 1996)\cite{St96}, is based on a dataset that has many characteristics in common with the ENACS data gathered for the same field. Indeed, the two datasets have been obtained with the OPTOPUS multifiber spectrograph at the ESO 3.6-m telescope and cover essentially the same area on the sky. Besides, they have also a very similar number of galaxies: 39 and 44, respectively (28 of which are shared). The MEFOS redshift dataset for A3733 reported in this paper contains two and a half times the number of galaxy radial velocities reported by Stein (1996)\cite{St96}, including 26 reobservations, while it covers a circular region around the center of A3733 four times larger. Furthermore, its high degree of completeness offers the possibility of extracting a complete magnitude-limited subset with a number of galaxies large enough for its use on statistical analysis. The plan of the paper is as follows. In Sect.~2 we discuss the MEFOS spectroscopic observations and data reduction, and present a final sample with 112 entries built by the combination of the MEFOS and Stein's (1996)\cite{St96} data. Section~3 begins with a brief description of the tools which will be used for the analysis of the data. Next, we identify the galaxies in our sample that belong to A3733, and use this dataset and a nearly complete magnitude-limited subset of it to examine the kinematical properties and structure of the central regions of the cluster. Section~4 summarizes the results of our study. | We have reported 104 radial velocity measurements performed with the MEFOS multifiber spectrograph at the 3.6-m ESO telescope for 99 galaxies in a region of $30\arcmin$ around the center of the cluster A3733. To augment this data, we have combined the MEFOS measurements with 39 redshifts measured by Stein (1996)\cite{St96} with the OPTOPUS instrument at the same telescope. This has given a final dataset with a total of 112 entries in the field of A3733. Radial velocities have been then supplemented by COSMOS \bj\ magnitudes and accurate sky positions in order to investigate the kinematics and structure of the central regions of the cluster. From a sample containing 74 strict cluster members, we have derived a heliocentric systemic velocity for A3733 of $11\,653^{+74}_{-76}$ \kms, resulting in a $\overline{z}_{\mathrm CMB}$ of 0.0380, and a velocity dispersion of $614^{+42}_{-30}$ \kms, in good agreement with the estimates by Stein (1997)\cite{St97} from the OPTOPUS data alone. Statistical tests relying exclusively on the distribution of observed velocities have yield suggestive indication of the possible kinematical complexity of A3733, especially when applied to a nearly complete magnitude-limited (\bj\ $\leq 18$) sample of cluster members. Despite this result, two powerful substructure tests that incorporate spatial information have failed to detect in this latter sample any statistically significant evidence of clumpiness in the galaxy component, in agreement with the findings of a previous study based on a spatially less extended and less complete dataset. Given that the sensitivity of the spatial substructure tests we have used is reduced when the subunits are seen with small projected separations, the results of the present study cannot exclude, however, the possibility that the signs of kinematical complexity detected in the velocity histogram of A3733 might be due to the existence of galaxy subcondensations superposed along the line-of-sight. | 98 | 3 | astro-ph9803180_arXiv.txt |
9803 | astro-ph9803194_arXiv.txt | An analysis of the observed characteristics of the Galactic Cepheid variables is carried out in the framework of their \plr\ being used as a standard candle for the distance measurement. The variation of the observed number density of Galactic Cepheids as function of their period and amplitude along with stellar pulsation characteristics is used to divide the population into two groups: one with low periods, probably multi-mode or higher mode oscillators, and another of high period variables which should be dominantly fundamental mode radial pulsators. Methods to obtain extinction-corrected colors from multi-wavelength observations of the second group of variables are described and templates of the $\vi $ \lig s are obtained from the $V$ \lig s. Colors computed from the model atmospheres are compared with the extinction-corrected colors to determine the Cepheid instability strip in the {\em mean surface gravity--effective temperature diagram}, and relations are derived between mean colors {\em $\BV$ vs period of pulsation}, {\em $\vi$ vs period}, and {\em $\vi$ at the brightest phase vs amplitude of pulsation}. The strength of the $\kappa$-mechanism in the envelope models is used to estimate the metal dependency of the instability strip from which an idea of the sensitivity of the \plr\ to the helium and metal abundance is given. Some estimate of the mass of Cepheids along the instability strip is provided. | \label{sec:intro} The classical Cepheid variables provide an important standard candle to measure distances to galaxies up to $\sim 30$ Mpc. The Cepheid Distance Scale is considered to be among the most reliable methods because the physics of Cepheid pulsation is well-understood and the relation between the pulsation period and luminosity of the star is observationally well-established. The Cepheids are luminous, have a narrow range of surface temperatures; their pulsation is very stable and exhibit large amplitude. The intrinsic scatter in their \plr\ is believed to be only around $0.3$ mag. However, the Cepheid distance scale cannot be directly calibrated from the observation of nearby stars and consequently, several systematic effects still undermine its effectiveness as a standard primary candle to determine extragalactic distances beyond a few Mpc. Some of the questions which have direct bearing on the problem of distance calibration, but whose answers remain inconclusive in spite of extensive research, are listed below. \bei \item Are the preferential pulsation modes of Cepheids period-dependent? \item Is a single \plr\ applicable to the entire instability strip? \item Is the \plr\ modified appreciably due to metallicity dependency of the stellar structure? \item Is the \plc\ relation a better indicator of distance than the \plr\ only? \eni \noindent Iben and Tuggle (1975) numerically computed the period and luminosity of Cepheids for a range of masses and obtained a relation between metallicity, surface temperature, period and luminosity. The \plc\ relation is found to be dependent on metallicity due to extreme sensitivity of the color--temperature relation on chemical composition. However, according to Becker, Iben and Tuggle (1977), within the uncertainties, the relation between period and luminosity for the first and second crossings of the Cepheid instability strip by a particular star does not crucially depend on the chemical composition. Since the time spent in traversing the strip is largest for the second crossing, most of the observed Cepheids are in this stage of evolution. So, although conversion of the period into a $V$-magnitude will introduce small effects due to surface temperature and metallicity, a \plr\ derived from observations should not be affected by small variations in the chemical composition. Indeed, the robustness of the \plr\ against changes in the chemical composition is borne out by theoretical as well as observational studies. The theoretical models of Bressan \ea (1993) produce the same period (within an error of $2 \% $) for a given luminosity, irrespective of the chemical composition. From observations of Cepheids in M31, it appears that there is no significant dependence of the \pl\ zero point on metallicity gradients (\cite{fm:90}). The recent review on the metallicity dependence of the Cepheid Distance Scale in the context of the HST Key Project on Extragalactic Distance Scale (\cite{kenn:98}) also leads to the same conclusion. Similarly, even though color-color diagram of Cepheids can be used to determine their metallicity, it is not an improvement in terms of its application to the estimation of distances, given that after extinction correction the color has larger error than the $V$-magnitude (e.g., \cite{fg:93}). Clearly, in order to address the above question, systematic work is required on the pulsation properties, evolution as well as stellar atmospheric structure, taking into account the onset of convection in the atmosphere during the pulsation cycle. In the present study we shall adopt the working hypothesis that the luminosity of the star {\it as a function of period} is not directly altered by metallicity (subject to the star being a classical Cepheid), and that the color of the star provides a better diagnostic for the estimation of extinction than for the determination of the distance. We devise methods to determine the extinction by the interstellar medium, and particularly emphasize the importance of observations in multi-wavelength bands, and also address the question of pulsation modes of Cepheids. The Cepheids in our own Milky Way galaxy have been observed by several astronomers over the years, and it is possible to obtain multi-wavelength data as well as accurate periods for a large number of them. A careful analysis of these Galactic Cepheids will naturally provide a useful template for identifying and estimating the various errors in the calibration of the Cepheid Distance Scale. A robust calibration of this distance scale is particularly important for extending it to extragalactic domains, as Cepheids are being observed in several far away galaxies, including those in the Virgo Cluster, by the Hubble Space Telescope. The hope is that distances based on these observations will ultimately lead to an accurate determination of the Hubble Constant. In this context, we attempt to provide a new calibration of the Cepheid Distance Scale, which is free from many of the systematic errors. In an accompanying communication (which we will refer to as Paper II), we apply these results for estimating the distance to the Virgo Cluster, based on the HST data for the Cepheids in the spiral M100. This paper is organized as follows. In Section~\ref{sec:number} we discuss the number distribution of Cepheids against their periods. In Section~\ref{sec:ligcur}, we demonstrate the feasibility of obtaining accurate $\vi $ \lig s from the $V$ \lig\ of a Cepheid and limited number of observations in the I band. This method is particularly useful for analyzing Cepheid data with few observations in one band (as in HST observations). In Section~\ref{sec:extcor}, we devise a formalism for extinction correction for each individual Cepheid, based on model atmospheres and an $\rv$-dependent extinction law. Several useful period--color and amplitude--color relationships are also derived. Section~\ref{sec:modes} is concerned about the different modes of pulsation of Cepheid variables and their manifestations in the observed properties like period and amplitude of pulsation. We argue that it is necessary to choose the correct lower cutoff period in the \plr\ in order to prevent contamination from multi-mode pulsators. In Section~\ref{sec:mass}, we give an estimate for Cepheid masses at different periods, based on our results about the instability strip in the surface gravity versus effective temperature plane. Some discussion on the metallicity effects are presented in Section~\ref{sec:metal} and the major limitations of the present work are listed in Section~\ref{sec:limit}. The main conclusions from this work are summarized in Section~\ref{sec:concl}. | 98 | 3 | astro-ph9803194_arXiv.txt |
|
9803 | hep-ph9803412_arXiv.txt | We show that the decaying magnetohydrodynamic turbulence leads to a more rapid growth of the correlation length of a primordial magnetic field than that caused by the expansion of the Universe. As an example, we consider the magnetic fields created during the electroweak phase transition. The expansion of the universe alone would yield a correlation length at the present epoch of 1 AU, whereas we find that the correlation length is likely of order 100 AU, and cannot possibly be longer than $10^4$ AU for non-helical fields. If the primordial field is strongly helical, the correlation length can be much larger, but we show that even in this case it cannot exceed 100 pc. All these estimates make it hard to believe that the observed galactic magnetic fields can result from the amplification of seed fields generated at the electroweak phase transition by the standard galactic dynamo. | Recently, considerable interest has been focused on the possibility that a primordial magnetic field may have been created at some early stage of the evolution of the Universe. While the existence of a weak widespread extragalactic chaotic magnetic field cannot be ruled out, almost the only place where such a primordial field may have left an observable imprint is in galaxies, many of which possess microgauss, kiloparsec scale fields that are thought to be the result of dynamo amplification of a weak seed field \cite{Kronberg}. The idea that the seed for the galactic dynamo may be the field created in the very early Universe has inspired a number of works. For example, such fields may have appeared at the electroweak phase transition \cite{BaymMcLerran}. There are numerous other proposals involving physics at various scales (for a brief overview and further references, see \cite{Olesen-rev}). One feature shared by most particle-physics scenarios is the smallness of the correlation length of the magnetic field which results. Indeed, at the moment of creation, the correlation length is limited by the horizon radius. By pushing this moment to a very early stage in the history of the Universe, one makes the correlation length much smaller than it would be if the magnetic fields were created more recently, say, during proto-galactic contraction \cite{Ostriker}. For example, the fields generated at the electroweak phase transition, when the horizon radius is about 1 cm, would have a correlation length of at most about $10^{15}$ cm in the present Universe. Contraction of proto-galaxies is likely to reduce this length scale by 2 orders of magnitude, which gives $10^{13}$ cm, or 1 AU, as the characteristic length scale of the seed field which the galactic dynamo is supposed to amplify. Realistically, this length scale is likely much smaller, since the correlation length at creation is typically much less than the horizon size\footnote{A magnetic field created during or before inflation \cite{TurnerWidrow} may have a large correlation length, however the proposed inflationary models that could possibly generate a large-scale magnetic field may seem rather contrived \cite{Ratra}. We will not consider inflation-induced magnetic fields in this paper, but one should keep in mind this alternative.}. Because any primordial magnetic fields generated at early times (via particle physics occurring at, say, the electroweak scale) have such small correlation lengths, these fields are not attractive candidates to play the role of seed fields for the galactic dynamo, unless the correlation length is somehow increased. The best developed theory of magnetic amplification in galaxies, the mean-field dynamo \cite{Ruzmaikin}, operates under the assumption that the magnetic field is smooth on the scale of turbulent motion of the interstellar gas, which is of order 100 pc. One may attempt to apply the mean-field dynamo theory for large-scale Fourier components of the chaotic magnetic field, neglecting the small-scale ones, however it is not obviously the correct procedure. That the magnetic field at very small scales exponentiates rapidly is a well known fact that poses a serious problem for the mean-field dynamo theory \cite{KurlsrudAnderson}. In the situation when the seed field itself resides at small scales, this problem is likely to become more severe. In this paper, we will not address the question of how a seed field is created, whether at the electroweak phase transition or during some other early epoch. Our goal is to investigate the possibility that magnetohydrodynamic (MHD) effects can lead to a substantial increase in the length scale of the magnetic field at the present epoch. We show that decaying MHD turbulence typically leads to a faster growth of the magnetic correlation length than one would expect from the expansion of the Universe alone. We estimate that in the case of magnetic fields generated at the electroweak phase transition, the enhancement factor is $10^2$, and the correlation length may reach 100 AU. If the primordial field has a large Chern-Simons number, the enhancement factor may be much larger, but the correlation length cannot possibly exceed 100 pc today if the magnetic field comes from the electroweak epoch. Thus, although the magnetohydrodynamic effects we consider are of some help, they cannot increase the correlation length enough to make the electroweak generation of a primordial seed field a viable option. We also consider generating the seed field at the QCD phase transition. This may be a possibility, but is only viable if the bubble separation at the phase transition is very large. This paper is organized as follows. In Sec.\ \ref{sec:MHDeq} we write the basic equation governing the MHD of the Universe. Sec.\ \ref{sec:non-hel} is devoted to the decay of non-helical MHD turbulence, whereas Sec.\ \ref{sec:hel} describes the scaling laws of the decay of helical turbulence. Sec.\ \ref{sec:concl} contains concluding remarks. The Appendix contains details about the EDQNM approximation used in our numerical simulation. | \label{sec:concl} In this paper we have considered the behavior of the correlation length of the primordial magnetic field, if such a field is generated in the early Universe. We found the decay law of the MHD turbulence, which is responsible for a substantial increase of the correlation length. We also observed a qualitative difference between the cases of non-helical and helical initial magnetic field. Taking the case in which the magnetic field is generated during the electroweak phase transition as an example, we find that if the magnetic field is not helical the factor gained from MHD is about $10^2$ and today the field may be correlated at scales as large as 100 AU. If by some chance the field is helical, the enhancement factor is much larger, but the final correlation length cannot exceed 100 pc, corresponding to 1 pc after proto-galactic collapse. Let us note some uncertainties remaining in our estimation. (In general, addressing these issues would lead to less optimistic values of the enhancement factor.) First, it is not at all clear whether in MHD turbulence the magnetic energy reaches equipartition with the kinetic energy of bulk fluid motion. While semi-analytical calculations (including our simulation) favor equipartition \cite{Pouquet}, some numerical results indicate that the mean magnetic energy density remains small and concentrated at scales shorter than the largest turbulence scale \cite{dns}. If the latter remains true at very high Reynolds number, the magnetic field will be smaller after the electroweak phase transition and the correlation length at present will be smaller than we have estimated. The second factor is neutrino diffusion. After the electroweak phase transition, neutrinos are the particles with the longest mean free path. There is a certain time interval when neutrinos are still not decoupled from the physics at the turbulence scale, but their diffusion length is so large that the neutrino contribution to viscosity makes the Reynolds number to drop below 1. During this time, there is no turbulence and the magnetic field is frozen. After the neutrinos decouple from the fluid motion at the turbulence scale, magnetic stress leads to restoration of turbulence, and the magnetic length may continue to grow. The overall effect is some reduction in the estimate of the final magnetic correlation length. It would be nice to end this paper on a more positive note, and to this end we turn now to investigating the QCD phase transition, which occurs later than the electroweak transition and so may yield a longer correlation length. If the QCD phase transition is first order, it will introduce fresh turbulence stronger than the decaying electroweak turbulence. The amplitude and correlation length of the magnetic field will be determined by the QCD turbulence. If fields were regenerated at the QCD transition on length scales of order of the horizon, the length scale it would have today is very large, since the horizon size at the QCD phase transition is much larger than at the electroweak epoch. Even with no enhancement from MHD effects, the expansion of the universe yields a magnetic correlation length of order 1 pc. It can be estimated that the turbulence will survive till at least matter-radiation decoupling, at which the correlation length is is of order 1 kpc in the non-helical case and 100 kpc if the field is helical. However, a magnetic field correlated on the horizon size is not expected to be produced at the QCD phase transition, and these estimates are too optimistic. Instead, the initial length scale must be of the order of the bubble spacing at the end of the phase transition (which is the natural scale of turbulence). Bubble spacings larger than 100 cm are unlikely \cite{Ignatius} and may have problem with standard big bang nucleosynthesis \cite{bbn}. A bubble spacing smaller than 100 cm again implies a very small correlation length at the present epoch. If this constraint is respected, our analysis of the QCD case must end pessimistically, as in the electroweak case. However, one cannot completely rule out the possibility of a very large bubble spacing, which leads to a magnetic field correlated on a long enough length scale to be of interest. A nonstandard evolution in which mixing after the phase transition occurs rapidly by hydrodynamic flows instead of slowly by diffusion or nonstandard nucleosynthesis may be required in this case. Finally, although in this paper we presented the smallness of the magnetic correlation length as undesirable and tried to overcome it by invoking MHD turbulence, there may exist a non-standard dynamo mechanism where a small-scale seed gives rise to a large-scale magnetic field (see, e.g., \cite{Chandran}). This possibility is, however, outside the scope of the present paper. | 98 | 3 | hep-ph9803412_arXiv.txt |
9803 | astro-ph9803257_arXiv.txt | We present a systematic search for OVI(1032\AA,1037\AA) absorption in a Keck HIRES spectrum of the $z=3.62$ quasar Q1422+231, with the goal of constraining the metallicity and ionization state of the low density intergalactic medium (IGM). Comparison of CIV absorption measurements to models of the \lya forest based on cosmological simulations shows that absorbers with $\nh \ga 10^{14.5}\cdunits$ have a mean carbon abundance [C/H]~$\approx -2.5$, assuming a metagalactic photoionizing background with the spectral shape predicted by Haardt \& Madau (1996, HM). In these models, lower column density absorption arises in lower density gas where most CIV is photoionized to CV. Therefore, OVI should be the most sensitive tracer of metallicity in \lya absorbers with $\nh \la 10^{14.5}\cdunits$. OVI lines lie at wavelengths heavily contaminated by Lyman series absorption, so we interpret the search results by comparing to carefully constructed, mock Q1422 spectra drawn from a hydrodynamic simulation of a $\Lambda$-dominated cold dark matter model. A search for deep, narrow absorption features yields only a few candidate OVI lines in the spectrum of Q1422. HI absorption blankets the position of the doublet companion line in each case, and the total number of narrow lines is statistically consistent with that in zero-metallicity artificial spectra. Artificial spectra generated with the HM background and [O/H]~$\ga -2.5$ predict too many narrow lines and are statistically inconsistent with the data. We also search for OVI associated with CIV systems, using the optical depth ratio technique of Songaila (1998). With this method we {\it do} find significant OVI absorption; matching the data requires [O/C]~$\approx +0.5$ and corresponding [O/H]~$\approx -2.0$. Taken together, the narrow line and optical depth ratio results imply that (a) the metallicity in the low density regions of the IGM is at least a factor of three below that in the overdense regions where CIV absorption is detectable, and (b) oxygen is overabundant in the CIV regions, consistent with the predictions of Type II supernova enrichment models and the observed abundance pattern in old halo stars. The photoionizing background spectrum would be truncated above 4~Ry in regions that have not undergone helium reionization (HeII$\longrightarrow$HeIII), and in this case matching the Q1422 data requires lower [C/H] but higher [O/H]. Taking [O/C]$\approx +1$ as the maximum plausible overabundance of oxygen, we conclude that helium must have been reionized through at least 50\% of the volume from $z \sim 3 - 3.6$. | The Lyman alpha (Ly$\alpha$) ``forest" (\cite{lyn71}; \cite{sar80}) of spectral features caused by HI absorption along the line of sight to a quasar probes the state of the intergalactic medium over a wide range of physical conditions. During the past few years, high-precision observations made using the HIRES spectrograph (\cite{vog94}) on the 10m Keck telescope have quantified the statistics of these low column density absorbers to unprecedented accuracy (e.g., \cite{hu95}; \cite{lu96b}; \cite{kim97}; \cite{kir97}). During the same time span, cosmological simulations that incorporate gas dynamics, radiative cooling, and photoionization have been able to reproduce many of the observed properties of quasar absorption spectra (\cite{cen94}; \cite{zha95}; \cite{her96}; \cite{mir96}; \cite{dav97a}). The rapid progress on theoretical and observational fronts has led to the emergence of a new paradigm for the origin of the high-redshift ($z\ga 2$) \lya forest, in which most \lya forest lines are produced by regions of low to moderate overdensity in hierarchically collapsing structures that are not in dynamical or thermal equilibrium. \lya lines of lower column density generally arise in gas of lower physical density, which has a lower neutral hydrogen fraction because of the reduced recombination rate. In this paper, we use a matched comparison between a cosmological hydrodynamic simulation and a Keck HIRES spectrum of the quasar Q1422+231 to constrain the metal abundance of this low density gas. The recent detection of metal lines associated with \lya forest absorbers having column densities $\nh \la 10^{15} \cdunits$ (\cite{cow95}; \cite{tyt95}; Songaila \& Cowie 1996, hereafter \cite{son96}) has provided a new avenue for investigating the ionization state and enrichment history of the high-redshift intergalactic medium (IGM). \cite{son96} showed that 75\% of \lya absorbers with $\nh > 10^{14.5} \cdunits$ have associated CIV absorption. Using simple photoionization models they estimate that the mean metallicity of these absorbers is between\footnote{We use the standard notation of brackets to denote the relative abundance, in logarithm, versus solar.} [C/H]$\sim -2$ and $-3$. Studies that use cosmological simulations to model the density and temperature of the absorbing gas obtain a better match to the data for a mean metallicity of [C/H]$\sim -2.5$ with around one dex of scatter, assuming either a power-law ionizing background (\cite{hae96}) or a reprocessed quasar ionizing background (Hellsten \etal 1997, hereafter \cite{hel97}). The question of how metals came to reside in these intermediate-density \lya absorbers remains unanswered. Since \lya absorbers with $\nh \la 10^{15}\cdunits$ are optically thin and associated with density peaks of overdensity $\la 20$, it is unlikely that they contain star-forming regions that produce {\it in situ} enrichment. Thus some transport mechanism must be invoked to explain the presence of metals in these regions. One possibility is that the metals are ejected from nearby, forming galaxies, by some combination of supernova blowout (\cite{mir97}) and tidal stripping (\cite{gne97}; \cite{gne98}). Since the enriching proto-galaxies are likely to form more efficiently in high density environments, these scenarios predict a significant correlation between metallicity and density (see, e.g., figure~6 of Gnedin \& Ostriker 1997). Alternatively, the IGM may have been enriched at a very early epoch by more ubiquitous Population III objects (e.g., \cite{hai97}), in which case all \lya absorbers might be expected to have roughly the same metallicity. Because CIV lines probe only a small range of HI column densities (and hence physical densities) with adequate statistics, it is difficult to distinguish between these enrichment models using only CIV data. Metal abundances can be studied at higher densities in Lyman limit systems ($\nh \ga 10^{17}\cdunits$) and damped \lya systems ($\nh \ga 10^{20}\cdunits$), but these are objects where {\it in situ} enrichment is likely and where (especially for Lyman limit systems) uncertain radiative transfer effects complicate the inference of metal abundances from line strengths. In this paper, we attempt to extend metallicity constraints to the low column density ($\nh \la 10^{14.5}\cdunits$), and hence low physical density, \lya forest. Carbon is difficult to detect in this regime because the low density reduces $N_{C}$ and ionizes more CIV to CV. However, \cite{lu98} have used composite spectra to attempt to detect CIV in such systems, and we briefly discuss their results in \S\ref{sec: disc}. In this paper, we focus instead on OVI, which is expected to be the one detectable metal absorption feature tracing \lya absorbers with $\nh \la 10^{14.5}\cdunits$ because of its high ionization state and large oscillator strength (Hellsten \etal 1998, hereafter \cite{hel98}). The difficulty with this approach, and the reason that it has not been previously attempted, is that the OVI absorption features lie embedded within the \lya forest, which is quite crowded at these redshifts. We overcome this problem by using a line identification scheme specifically designed to select candidate OVI features and by using artificial spectra extracted from a realistic cosmological simulation to calibrate the efficiency of OVI detection and the contamination from narrow \lya lines. With these procedures, we can test whether the low density regions of the \lya forest are consistent with a uniform metallicity extrapolated from the CIV data at higher densities. The low density IGM is highly photoionized by the metagalactic ultraviolet (UV) background. For our standard spectral shape, we assume that the UV background is produced by quasar emission reprocessed by \lya forest absorption (Haardt \& Madau 1996, hereafter \cite{haa96}). We find that if the IGM metallicity is uniform at [C/H]$\sim -2.5$, the UV background has the spectral shape given by \cite{haa96}, and oxygen has the factor of three overabundance (relative to solar) predicted by Type II supernova enrichment models, then OVI should be readily detectable in the spectrum of Q1422+231. However, our detection algorithm finds very few candidate OVI lines in the spectrum. The absence of detectable OVI features has several possible interpretations: (1) oxygen is not overabundant relative to carbon in the low density IGM, (2) the metallicity of low density regions as traced by $\nh \la 10^{14.5}\cdunits$ \lya absorption systems is lower than the metallicity of intermediate-density regions traced by higher column density absorption, or (3) there are many fewer high-energy photons capable of photoionizing OV to OVI than are predicted by the \cite{haa96} ionizing background. To distinguish between these interpretations, we apply a second algorithm, the optical depth ratio technique of Songaila (1998, hereafter \cite{son98}) designed to detect OVI in regions where significant CIV absorption is found. Since we have independently determined [C/H] in these regions, we can use this technique to discriminate between different ionization conditions and abundance patterns. By applying this technique to the spectrum of Q1422+231 and calibrating the results using artificial spectra, we find that: (1) a significant metallicity gradient (declining metallicity with declining density) must exist regardless of whether helium has mostly reionized or not, (2) if helium has reionized by $z\sim 3.6$, our results are consistent with [O/C]~$\approx +0.5$, and (3) if the epoch of helium reionization does not begin until $z\sim 3$, a highly implausible oxygen overabundance of [O/C]~$\ga +2$ is required. Combining the results from these two analysis techniques, our favored scenario for the low density IGM at $z\ga 3$ is one in which more than half of the volume of the universe has helium predominantly reionized by $z\sim 3.6$, oxygen is overabundant relative to carbon by a factor $\ga 3$, and spatial regions with overdensities $\sim 10$ have an average metallicity {\it at least} a factor of 3 higher than regions near the mean baryonic density. Section~\ref{sec: modeling} describes our cosmological simulation, the Q1422+231 data, and our procedure for constructing artificial absorption spectra. Section~\ref{sec: OVIsearch} discusses previous OVI searches and reviews \cite{hel98}'s argument that OVI should be the most effective tracer of metallicity in \lya absorbers with $\nh \la 10^{14}\cdunits$. Section~\ref{sec: civ} examines CIV absorption at intermediate column densities, repeating the general arguments of \cite{hel97} and Rauch et al.\ (1997a) but with a much closer match between the theoretical and observational analysis procedures. Section~\ref{sec: search} is the heart of the paper. It describes our algorithm for identifying candidate OVI lines, presents the results of the OVI searches in the real and artificial spectra, and discusses the properties of candidate OVI absorbers in Q1422+231 and the simulations. Figures~\ref{fig: auto5}--\ref{fig: OVIsel} demonstrate the paper's central results. Section~\ref{sec: assumptions} discusses the impact of varying the assumptions of our standard theoretical model. Section~\ref{sec: pixmet} describes the optical depth ratio technique for quantifying OVI absorption, shows the results from this technique applied to Q1422+231 and artificial spectra, and discusses these results in conjunction with the results from Section~\ref{sec: search}. Section~\ref{sec: disc} summarizes our conclusions and discusses them in light of other recent observational and theoretical developments. | \label{sec: disc} We present a systematic search for OVI absorption in the spectrum of Q1422+231 ($z=3.62$), using a narrow line detection algorithm proven effective at identifying OVI absorption in artificial spectra, and an optical depth ratio technique introduced by \cite{son98}. The first technique traces OVI predominantly in systems with $10^{13.5}\la \nh\la 10^{15}\cdunits$, whereas the second technique traces only OVI associated with CIV absorption, \ie in $10^{14.5}\la \nh\la 10^{16}\cdunits$ systems. By comparing Q1422 and artificial spectra having varying metallicities, we determine that \begin{enumerate} \item{[O/H] must be lower in lower density regions, for either an \cite{haa96} ionizing background or an ionizing background significantly truncated above 4~Ry. If [O/C] is constant in systems up to $\nh\sim 10^{16}\cdunits$, our results imply that regions traced by $\nh\la 10^{14}\cdunits$ systems (corresponding to gas at roughly the mean baryonic density) have a mean metallicity lower by at least a factor of 3 compared to regions traced by $\nh\sim 10^{15}\cdunits$ systems (corresponding to a baryonic overdensity of $\sim 10$).} \item{More than half the universe must have helium reionized by $z\sim 3$. If helium has completely reionized by $z\sim 3.6$ (the highest redshift probed by the Q1422 data), then our analysis implies [O/C]~$\approx +0.5$, in good agreement with overabundance measurements of Type II supernovae enriched systems. If a significant portion of the universe has not reionized helium by $z\sim 3$ and therefore has a softer ionizing background spectrum, then the required oxygen overabundance is higher. For example, if half of the volume has not reionized helium, then [O/C]~$\approx +1.2$, already greater than the observed overabundance of any class of Type II supernovae enriched systems. If the spectrum were soft throughout the universe at $z\ga 3$ then an implausibly high overabundance, [O/C]~$\approx +2.3$, would be required. } \end{enumerate} These conclusions are in good agreement with the recent study of \cite{lu98}, who used composite spectra to investigate CIV absorption in systems with $10^{13.5}< \nh < 10^{14} \cdunits$ and found that the metallicity of these absorbers must be [C/H]~$\la -3.5$. Cosmological simulations show that $\nh \sim 10^{14} \cdunits$ roughly corresponds to the dividing line between overdense and underdense regions of the universe (though the value of $\nh$ that marks this division depends on redshift and, to a lesser extent, on cosmological parameters). Our results therefore imply that mildly overdense regions such as filaments and sheets have been enriched, while underdense regions are virtually chemically pristine. Simulations that self-consistently enrich the IGM by tracking metal production and transport find that a strong metallicity gradient is predicted between the mildly overdense and underdense regions (see figure~3 in \cite{gne98}); this predicted gradient is in good agreement with the \cite{lu98} data and with the scenario we present above. Recent measurements of a jump in the SiIV/CIV ratio around $z\sim 3$ (\cite{son98}) may be difficult to reconcile with conclusion~(2) above. While we have yet to conduct a systematic comparison of SiIV in observed and artificial spectra, primarily because of the small numbers of SiIV systems detectable in our one available quasar spectrum, we expect our results will be in agreement with \cite{son98}, who argues that such a jump requires a much softer ionizing background at $z\ga 3$. One way to reconcile these results may be to invoke patchy helium reionization at that epoch, as suggested by \cite{rei97}; such a model may be tested in greater detail by searching for an anti-correlation between OVI and SiIV detections. If helium reionization occurs around $z\sim 3$, our narrow line algorithm should yield many more OVI detections at redshifts $z\la 3$. Such searches are difficult because of the poor blue sensitivity of the HIRES spectrograph (and complete loss of sensitivity at $\lambda\la 3800$\AA), but quasar spectra do exist that could provide constraints down to $z\sim 2.7$. The presence of a substantial number of OVI lines in this regime would strongly favor the late helium reionization scenario; there is already some weak evidence that OVI is more abundant at $z\la 3$ (\cite{son98}). If OVI features continue to be virtually undetectable down to $z\sim 2.7$, this would be compelling evidence against the late helium reionization scenario, since the HeII absorption measurements of Davidsen, Kriss, \& Zheng (1996) imply that helium has been reionized by this redshift. We hope to work with observers to attempt this search in the near future. In a broader context, our work illustrates the power of combining cosmological hydrodynamic simulations of structure formation with high-quality quasar spectra to infer the ionization state and the enrichment history of the high-redshift IGM. Future observations and simulations promise a wealth of information, which, when combined, will help us to better understand the evolution of the IGM and its connection to early star formation and the epoch of primeval galaxies. | 98 | 3 | astro-ph9803257_arXiv.txt |
9803 | gr-qc9803026_arXiv.txt | Clues as to the geometry of the universe are encoded in the cosmic background radiation. Hot and cold spots in the primordial radiation may be randomly distributed in an infinite universe while in a universe with compact topology distinctive patterns can be generated. With improved vision, we could actually see if the universe is wrapped into a hexagonal prism or a hyperbolic horn. We discuss the search for such geometric patterns in predictive maps of the microwave sky. | \label{compf} We want to predict a map of the temperature fluctuations. In a homogeneous and isotropic space, an angular average over the fluctuations contains all of the essential information. Of the six compact, orientable flat spaces, all destroy global isotropy and all except for the hypertorus destroy global homogeneity. As a result, there is more information in a map of temperature fluctuations than just the angular power spectrum. Although we argue that the angular average overlooks conspicuous features in general, for the equal sided flat cases angular spectra do provide a reasonable bound. Four of the six orientable, compact topologies of $\e$ are constructed from a parallelepiped as the fundamental domain. The other two are built from a hexagonal prism. The hypertorus is the simplest and has been studied by many authors \cite{{flat},{moreflat},{sss},{add}}. Stevens, Scott, and Silk \cite{sss} pointed out that in a flat $3$-torus, the spectrum of temperature fluctuations was truncated at long wavelengths in order to fit within the finite box. Contrary to standard lore, we find all of the equal sided compact flat manifolds show a truncation in the power of fluctuations on wavelengths comparable to the size of the fundamental domain \cite{{lss},{tarun}}. The longest wavelength fluctuation observed, namely the quadrupole, is in fact low. Some might even take this as evidence for topology \cite{workshop}. Cosmic variance is also large on large scales. Consequently, a fundamental domain the size of the observable universe is actually consistent with the COBE data \cite{lss}. A very small universe however is incompatible with the data. The cutoff in long wavelength perturbations is accompanied by gaps in power at wavelengths that do not correspond to integer windings through the fundamental domain. {\it All} compact spaces show discrete harmonics and as such the sharp harmonics may be a more generic sign of compact topology. The jaggy spectra of such small compact flat spaces are tens of times less likely than the smooth spectrum of infinite $\e$. We conclude, quite conservatively, that the universe, if finite and flat and equal-sided, must be at least $80$\% the radius of the surface of last scatter and so $40$\% of the diameter of the observable universe. There could still be as many as eight copies of our universe within the observable horizon. \begin{figure} \centerline{{\psfig{file=hexglue.eps,width=3in}}} \centerline{{\psfig{file=hex0_1_.1.ps,width=3in}}} \caption{A hexagonal prism with a $2\pi/3$ twist. The observer is at the center of the universe in the map of $\delta T(\hat n)/T$. The fundamental domain is half the diameter of the observable universe in two directions and one-tenth that in the twisted direction. \label{fighex}} \end{figure} If instead of an equal-sided space we consider a fundamental domain with disparate length scales, the angular power spectrum is in general a poor discriminant. The averaging over the sky fails to recognize the strong features in the cosmos. Fig.\ \ref{fighex} shows a predictive map of the hot and cold fluctuations in a $2\pi/3$-twisted hexagonal prism. We have set the length of the fundamental domain to be ten times smaller in the twisted direction than along the face of the hexagon. The average large angle power in fluctuations is actually consistent with the data, although clearly this anisotropic space does not look like the sky we observe. A better statistic to discern patterns and correlations is badly needed \cite{{lss},{lssb}}. The promising suggestions of \cite{{bps},{css},{lssb}} may be the key and are discussed more in \S \ref{comph}. \begin{figure} \centerline{{\psfig{file=tile2.eps,width=1.5in}}} \centerline{{\psfig{file=hexmodes.eps,width=1.75in}}} \caption{Top: A guess at one mode. Bottom: A contour plot of the temperature fluctuation for a similar mode. \label{onemode}} \end{figure} We could have predicted certain features of the map of Fig.\ \ref{fighex}, even if we had not known the eigenmodes explicitly. A 2D slice through the 3D tiling of space is represented in Fig.\ \ref{onemode}. If we draw bands connecting opposite sides of the hexagons and highlight any overlaps, we can predict the imprint of one mode as shown on the top of Fig. \ref{onemode}. Given that we do know the eigenmodes, we can show the actual contour plot of the hot and cold fluctuations for a similar mode on the bottom of Fig.\ \ref{onemode}. Comparing the guess with the actual contours shows our guess did quite well. The hexagonal shape of the universe is clearly seen. In actuality, there are many modes competing to imprint a pattern on the sky which blurs the signature hexagons. In Fig.\ \ref{mode2}, is another contour plot of hot and cold spots for a different mode which exhibits the $2\pi/3$ twist through the prism. \begin{figure} \centerline{{\psfig{file=hexmodes2.eps,width=1.75in}}} \caption{A contour plot of the temperature fluctuation for a mode that winds through the twisted prism. \label{mode2}} \end{figure} The competition between fluctuations obscurs some features while enhancing others. The surface of last scatter cuts a sphere out of the full 3D space, an elliptic projection of which is given in the map of Fig. \ref{fighex}. Can you see hexagons in the map? Almost. Is the $2\pi/3$ twist in the space visible? We are currently developing ways of looking at the sky that pull the underlying patterns out of the noise \cite{lssb}. | 98 | 3 | gr-qc9803026_arXiv.txt |
|
9803 | astro-ph9803311_arXiv.txt | We present the results of deep spectropolarimetry of two powerful radio galaxies at $z\sim2.5$ (4C 00.54 and 4C 23.56) obtained with the W.M. Keck II 10m telescope, aimed at studying the relative contribution of the stellar and non-stellar components to the ultraviolet continuum. Both galaxies show strong linear polarization of the continuum between rest-frame $\sim$1300-2000~\AA, and the orientation of the electric vector is perpendicular to the main axis of the UV continuum. In this sense, our objects are like most 3C radio galaxies at $z\sim1$. The total flux spectra of 4C 00.54 and 4C 23.56 do not show the strong P-Cygni absorption features or the photospheric absorption lines expected when the UV continuum is dominated by young and massive stars. The only features detected can be ascribed to interstellar absorptions by SiII, CII and OI. Our results are similar to those for 3C radio galaxies at lower $z$, suggesting that the UV continuum of powerful radio galaxies at $z\sim2.5$ is still dominated by non-stellar radiation, and that young massive stars do not contribute more than $\approx$50\% to the total continuum flux at 1500~\AA. | High-$z$ radio galaxies (H$z$RGs) are observable to very high redshifts and can be used to study the formation and evolution of massive elliptical galaxies (see McCarthy 1993 for a review). One of the most controversial issues is the physical cause of the alignment between the radio source and UV continuum axes of the H$z$RGs (the so called `alignment effect', Chambers, Miley \& van Breugel 1987, McCarthy et al. 1987). Two main competing scenarios have been proposed. The first is star formation induced by the propagation of the radio source through the ambient gas (see McCarthy 1993 and references therein); the second explains the alignment effect as the result of a hidden quasar whose radiation is emitted anisotropically and scattered towards the observer, producing strong linear polarization perpendicular to the radio-UV axis (Tadhunter et al. 1988; di Serego Alighieri et al. 1989). The latter scenario is closely related to the unification of powerful radio-loud AGN, and provides a way of testing it directly (see Antonucci 1993 and references therein). After the first detections of strong UV polarization in H$z$RGs obtained with 4m-class telescopes (di Serego Alighieri et al. 1989; Jannuzi \& Elston 1991; Tadhunter et al. 1992; Cimatti et al. 1993), recent observations made with the Keck I 10m telescope have demonstrated the presence of spatially extended UV continuum polarization and of hidden quasar nuclei in some of the 3C radio galaxies at $0.7<z<1.8$, favoring the beaming and scattering scenario (Cohen et al. 1996; Cimatti et al. 1996,1997; Dey et al. 1996; Tran et al. 1998). On the other hand, Dey et al. (1997) have recently shown that the UV continuum of 4C 41.17 ($z=3.8$) is unpolarized and consistent with that of a typical starburst galaxy. The most stringent comparison between the starburst and the scattering scenarios can be performed at $\lambda_{rest} \sim 1000-2000$~\AA, where most of the strongest spectral features of O and B stars are located. This spectral window can be covered from the ground by observing radio galaxies at $z>2$. We have started a program of observations of these galaxies using spectropolarimetry at the Keck II 10m telescope, and in this Letter we report on the first two objects we have studied, concentrating on their continuum and absorption line properties. Throughout this paper we assume $H_0=50$ kms$^{-1}$ Mpc$^{-1}$ and $q_0=0$. | Our observations suggest that the UV spectra of 4C 23.56 and 4C 00.54 are not dominated by young massive stars, whereas the strong perpendicular polarization indicates the presence of a relevant scattered continuum, making 4C 23.56 and 4C 00.54 similar to the polarized 3C radio galaxies at 0.7$<z<$2. Adopting the prescriptions of Dickson et al. (1995) and Manzini \& di Serego Alighieri (1996) and assuming the average HeII$\lambda$1640/H$\beta$ ratio (3.18) observed in radio galaxies (McCarthy 1993), we estimate that the nebular continuum contributes only $\sim$8\% and $\sim$13\% to the total flux at 1500~\AA~ for 4C 23.56$a$ and 4C 00.54 respectively. If we assume that all H$z$RGs have an obscured quasar nucleus which feeds the powerful radio source and whose light is scattered by dust and/or electrons, we can interpret the low (or null) polarization of 4C 41.17 (Dey et al. 1997) as due to dilution of the scattered radiation by the unpolarized light of young stars. A limit on the amount of stellar light in the UV continuum of 4C 23.56$a$ and 4C 00.54 can be derived by assuming that the observed polarization is diluted by the unpolarized stellar and nebular continua. The ratio between the stellar light and the total flux at 1500~\AA~ can be roughly estimated as $F_{stars}/F_{total}=\{[1-(P_{obs}/P_0)]-\kappa\}$, where $P_{obs}$ and $P_0$ are the observed and intrinsic degree of polarization and $\kappa$ is the ratio between the nebular and the total continuum. Adopting a half-cone opening angle of 45$^{\circ}$ and an angle of 90$^{\circ}$ between the cone axis and the line of sight, we derive $P_0\sim$30\% and $P_0\sim$50\% for dust (Manzini \& di Serego Alighieri 1996) and electron scattering (Miller, Goodrich \& Mathews 1991) respectively. Thus, adopting $P_{obs}$(1500~\AA)=13.1\% and 14.4\% for 4C 00.54 and 4c 23.56$a$ respectively, for dust scattering we obtain that $F_{stars}/F_{total} \leq$43\% and $\leq$44\% for 4C 00.54 and 4C 23.56$a$ respectively, whereas $F_{stars}/ F_{total}$ increases to $\leq$61\% and $\leq$63\% for electron scattering. These ratios can be considered upper limits because we do not know if the observed scattered light is really diluted by a stellar continuum, and they imply that stellar light cannot contribute more than about half of the UV continuum at 1500~\AA. The properties of 4C 23.56, 4C 00.54 and 4C 41.17 can be interpreted in a evolutionary scenario where H$z$RGs at $z>3$ have a major episode of star formation, and their AGN scattered component is diluted by the stellar light, but it becomes observable at lower $z$ when the starburst ceases. However, given the rapid evolution of the UV light from a starburst, it is also possible that 4C 41.17 simply represents a case dominated by the starburst rather than an evolutionary sequence. Future observation of a complete sample of H$z$RGs will help us to understand the nature of the alignment effect and the evolution of the host galaxies of powerful radio sources. | 98 | 3 | astro-ph9803311_arXiv.txt |
9803 | astro-ph9803127_arXiv.txt | We present high-quality HST/GHRS spectra in the Hydrogen L$\alpha$ spectral region of Vega and Sirius-A. Thanks to the signal-to-noise ratio achieved in these observations and to the similarity of the two spectra, we found clear evidence of emission features in the low flux region, $\lambda\lambda$1190-1222\,\AA. These emission lines can be attributed unambiguously to \ion{Fe}{ii} and \ion{Cr}{ii} transitions. In this spectral range, silicon lines are observed in absorption. We built a series of non-LTE model atmospheres with different, prescribed temperature stratification in the upper atmosphere and treating \ion{Fe}{ii} with various degrees of sophistication in non-LTE. Emission lines are produced by the combined effect of the Schuster mechanism and radiative interlocking, and can be explained without the presence of a chromosphere. Silicon absorption lines and the L$\alpha$ profile set constraints on the presence of a chromosphere, excluding a strong temperature rise in layers deeper than $\tau_{\rm R} \approx 10^{-4}$. | On the main sequence, A-type stars are at a juncture point between hot and cool stars. While hot, massive stars undergo strong mass loss in fast winds ($\dot{M}\ge 10^{-9}$\,M$_\odot$/yr), cool stars show chromospheric activity connected to their subsurface convective layers. Both phenomena apparently disappear or become much weaker at spectral type A. Many studies have thus been devoted to the outer layers of A-type stars to search for indications of a wind or of stellar activity. Several attempts to detect signatures of weak winds in main-sequence A stars have been unsuccessful (e.g. Lanz~\& Catala \cite{lanz92}). Recently, however, a quite weak, blue-shifted absorption was detected in the \ion{Mg}{ii} resonance lines of Sirius, and interpreted as a wind signature (Bertin et~al. \cite{sirius3}). A mass loss rate of $\dot{M}\approx 10^{-12}$\,M$_\odot$/yr was derived, consistent with the idea that A-type star winds are radiatively-driven like the winds of hotter stars. On the cool side, a limit to chromospheric activity has been set at A7 (B\"ohm-Vitense~\& Dettmann \cite{bohmvitense80}, Marilli et~al. \cite{marilli97}, Simon~\& Landsman \cite{simon97}). The most common diagnostics of chromospheres and winds are emission features. Therefore, we are not expecting emission lines in A stars, except cases where such lines arise from the circumstellar environment. High-quality ultraviolet spectra have become available with the {\em Goddard High Resolution Spectrograph} (GHRS) aboard the {\em Hubble Space Telescope} (HST). Even the core of strong resonance lines, including \ion{H}{i} L$\alpha$, can be observed with a reasonably good signal-to-noise ratio. This makes it possible to investigate in greater detail the line profile of strong resonance lines. They are the best tool to probe the outer layers of stars, being indeed formed very high in the atmosphere. In this respect, L$\alpha$ is most interesting because it spans the largest range of depth of formation, from the far wing to the line core. This large variation in opacity also affects the formation of lines of other elements, especially close to L$\alpha$ core. Such lines see a much lower local pseudo-continuum than lines outside L$\alpha$, and will be formed much higher in the atmosphere than weak lines in other regions. \begin{table*} \caption[]{Observation log.} \label{TabObs} \begin{tabular}{lccccr} \hline &&&&& \\ [-3mm] Target & Spectral Range & Date of Observation & GHRS Grating & Aperture & Exposure time \\ \hline &&&&& \\ [-3mm] Sirius-A & 1188 \AA\ -- 1218 \AA & 1996 Nov 20 & G140M & SSA & 1632.0 s \\ Sirius-A & 1278 \AA\ -- 1307 \AA & 1996 Nov 20 & G140M & SSA & 217.6 s \\ Sirius-A & 1308 \AA\ -- 1337 \AA & 1996 Nov 20 & G140M & SSA & 217.6 s \\ Vega & 1185 \AA\ -- 1222 \AA & 1996 Dec 23 & G160M & SSA & 435.2 s \\ Vega & 1274 \AA\ -- 1311 \AA & 1996 Dec 23 & G160M & SSA & 108.8 s \\ Vega & 1303 \AA\ -- 1341 \AA & 1996 Dec 23 & G160M & SSA & 108.8 s \\ \hline \end{tabular} \end{table*} In Sect. 2 and 3, we will describe our GHRS observations around L$\alpha$ of two bright A stars, Vega and Sirius-A. We will in particular point out the presence of \ion{Fe}{ii} and \ion{Cr}{ii} emission lines between 1190 and 1222\,\AA. Bertin et~al. (\cite{sirius2}) noticed the presence of emission features around L$\alpha$ in a Cycle~1 GHRS spectrum of Sirius-A, originally recorded to derive the D/H abundance ratio in the local interstellar medium. This prompted us to repeat and extend these observations to investigate their origin. An explanation of these emission features is given in the second half of the paper. In Sect.~4, we describe our new non-LTE model atmospheres. We explore and set limits on a chromosphere (Sect.~5), and investigate non-LTE effects in \ion{Fe}{ii} line formation (Sect.~6). | We have reported emission features in the L$\alpha$ profile of Vega and Sirius-A. These emission lines have been attributed to \ion{Fe}{ii} and \ion{Cr}{ii} transitions. The identification appears quite secure because all the lines of several multiplets appear in emission. We have built non-LTE model atmospheres with different assumed temperature structures in the outer layers and incorporating \ion{Fe}{ii} with different degrees of sophistication. We found that the emission features cannot be explained by a chromospheric temperature rise. To produce the observed \ion{Fe}{ii} emissions, the temperature would have to increase in relatively deep layers, turning other lines into emission (e.g. \ion{Si}{ii}, \ion{Si}{iii} lines). However, we cannot exclude a chromospheric rise in shallower layers ($\tau_{\rm R} \le 10^{-4}$) based on our present observations, in agreement with earlier results (Freire-Ferrero et~al. \cite{freire83}). Non-LTE \ion{Fe}{ii} line formation calculations with different model atoms have demonstrated that some \ion{Fe}{ii} lines can turn into emission in the wavelength range between 1190 and 1240\,\AA. We stress that emission lines are predicted {\em only} in this very low flux, central region of L$\alpha$. This results from the combined effect of the Schuster mechanism and radiative interlocking. Some highly-excited levels are overpopulated by transitions occurring in a high-flux region, and preferentially de-excite in this region near L$\alpha$. This mechanism explains the similarity of Vega and Sirius spectra. Differences between the two stars can also be understood with this mechanism. The higher heavy-element content in Sirius' photosphere results in depressing the flux, in particular near the flux maximum. The efficiency of the pumping is thus reduced, yielding generally weaker emissions in Sirius than in Vega. The details depend on the exact wavelength of the pumping transitions. The flatter L$\alpha$ profile in Vega is also a consequence of the different metallicity. Lyman continuum heating must be more efficient in Vega's case (less heavy-element line opacity) yielding a somewhat higher temperature in the outer layers and a higher flux in the central region of L$\alpha$. We believe that the origin of the \ion{Cr}{ii} emission lines may be explained by similar mechanisms. While we cannot rule out that radiative interlocking is also effective in \ion{Cr}{ii}, the difference between Vega and Sirius points to the Schuster mechanism being the major cause of the emission in this case. The lines are thus stronger in Sirius due to the larger chromium abundance, and are simply too weak in Vega to stand out of the noise. Although our model atmosphere calculations provide an explanation to an unexpected observation of emission lines in the spectrum of early A-type stars, we did not achieve a good fit to the L$\alpha$ profile at this stage. It seems however likely that the flux observed in the central region of L$\alpha$ may be explained by increasing somewhat the fraction of non-coherent scattering in the PRD approximation that we have used. Matching these observations would require (at least) non-LTE line-blanketed model atmospheres, treatment of L$\alpha$ in partial redistribution tuning the ratio between coherent and non-coherent scattering, and improved, non-hydrogenic \ion{Fe}{ii} photoionization cross-sections. Such an approach is necessary to gain a deeper insight into the outer layers of Vega and Sirius, and this certainly deserves further study. Finally, we did not find emission lines very close to the L$\alpha$ core, especially in the 0.5\,\AA\ blueward of the central wavelength. This implies fortunately that we have so far no reason to question the previous results on the local interstellar cloud (Bertin et~al. \cite{sirius2}), and on the wind absorption feature (Bertin et~al. \cite{sirius3}). | 98 | 3 | astro-ph9803127_arXiv.txt |
9803 | astro-ph9803037_arXiv.txt | We have preliminary results on the parallelization of a Tree-Code for evaluating gravitational forces in N-body astrophysical systems. For our T3D CRAFT implementation, we have obtained an encouraging speed-up behavior, which reaches a value of 37 with 64 processor elements (PEs). According to the Amdahl'law, this means that about 99\% of the code is actually parallelized. The speed-up tests regarded the evaluation of the forces among $N = 130,369$ particles distributed scaling the actual distribution of a sample of galaxies seen in the Northern sky hemisphere. Parallelization of the time integration of the trajectories, which has not yet been taken into account, is both easier to implement and not as fundamental. | Super computers are allowing a rapid development of numerical simulations of large N--body systems in Astrophysics. These systems are generally composed by both collisionless matter (such as: stars, galaxies, ...) and collisional matter (i.e. gas). Both phases are usually characterized by being self--gravitating, that is the dynamics of the bodies (stars or fluid elements) is strongly influenced by the gravitational field produced by the bodies themselves. This {\it self-influence} is what makes the evaluation of the long--range gravitational force the heaviest computational task to perform in a dynamical simulation. In fact, the number of terms which has to be considered in a direct and trivial evaluation of all the interactions between bodies grows like $N^2$, and since many astrophysically realistic simulations require very large $N$ (greater than $10^5$), such a direct numerical evaluation seems hard to face with presently available computers. To overcome this problem various approximate techniques to compute gravitational interactions have been proposed. Among them, the Tree--code algorithm proposed by Barnes \& Hut\footnote{ Barnes J., Hut P. ``A hierachical $O(N\log N)$ force calculation algorithm''. {\it Nature}, vol. 324, p. 446 (1986).} is now widely used in Astrophysics because it does not require any spatial fixed grid (like, for example, methods based on the solution of Poisson's equation). This makes it particularly suitable to follow very inhomogeneous and variable (in time) situations, typical of self-gravitating systems out of equilibrium. In fact its intrinsic capability to give a rapid evaluation of forces allows spending more CPU-time to follow fast dynamical evolution, in contrast to other higher accuracy methods that are more suitable for other physical situations, e.g. dynamics of polar fluids, where the Coulomb term is present. With the help of the parallelization of our codes, we intend to increase by one or two order of magnitude the number of particles we can use to represent physical systems, in respect to that generally adopted on serial computers ($\sim 10^4$). In particular our first scientifical aim is the study of close encounters between massive black holes and globular clusters. These latter are systems formed by more than $10^5$ stars gravitationally bounded in a spherical peaked distribution. Such a problem is important in the effort to understand better the nature and formation mechanisms of the {\it Active Galactic Nuclei}\ \footnote{ Capuzzo-Dolcetta R., Miocchi P., ``Galactic Nuclei Activity Sustained by Globular Cluster Mass Accretion'', {\it PaSS} (1998) in press.}. We hope parallelization makes possible to represent each star with a single particle, in a one--to--one correspondence. This fact clearly will make simulations much more physically meaningful. | 98 | 3 | astro-ph9803037_arXiv.txt |
|
9803 | astro-ph9803171_arXiv.txt | The isothermal gravitational collapse and fragmentation of a molecular cloud region and the subsequent formation of a protostellar cluster is investigated numerically. The clump mass spectrum which forms during the fragmentation phase can be well approximated by a power law distribution $dN/dM \propto M^{-1.5}$. In contrast, the mass spectrum of protostellar cores that form in the centers of Jeans unstable clumps and evolve through accretion and $N$-body interaction is best described by a log-normal distribution. Assuming a star formation efficiency of $\sim\!10\;\!$\%, it is in excellent agreement with the IMF of multiple stellar systems. | \label{sec:intro} Understanding the processes leading to the formation of stars is one of the fundamental challenges in astronomy and astrophysics. However, theoretical models considerably lag behind the recent observational progress. The analytical description of the star formation process is restricted to the collapse of isolated, idealized objects (Whitworth \& Summers 1985). Much the same applies to numerical studies (e.g.~Boss 1997, Burkert et al.~1997 and reference therein). Previous numerical models that treated cloud fragmentation on scales larger than single, isolated clumps were strongly constrained by numerical resolution. Larson (1978), for example, used just 150 particles in an SPH-like simulation. Whitworth et al.~(1995) were the first who addressed star formation in an entire cloud region using high-resolution numerical models. However, they studied a different problem: fragmentation and star formation in the shocked interface of colliding molecular clumps. While clump-clump interactions are expected to be abundant in molecular clouds, the rapid formation of a whole star cluster requires gravitational collapse on a size scale which contains many clumps and dense filaments. Here, we present a high-resolution numerical model describing the dynamical evolution of an entire {\em region} embedded in the interior of a molecular cloud. We follow the fragmentation into dense protostellar cores which form a hierarchically structured cluster. | \label{sec:summary} Large-scale collapse and fragmentation in molecular clouds leads to a hierarchical cluster of condensed objects whose further dynamical evolution is extremely complex. The agreement between the numerically-calculated mass function and the observations strongly suggests that gravitational fragmentation and accretion processes dominate the origin of stellar masses. The final mass distribution of protostellar cores in isothermal models is a consequence of the chaotic kinematical evolution during the accretion phase. Our simulations give evidence, that the star formation process can best be understood in the frame work of a probabilistic theory. A sequence of statistical events may naturally lead to a log-normal IMF (see e.g.~Zinnecker 1984, Adams \& Fatuzzo 1996; also Price \& Podsiadlowski 1995, Murray \& Lin 1996, Elmegreen 1997). | 98 | 3 | astro-ph9803171_arXiv.txt |
9803 | chao-dyn9803019_arXiv.txt | We study the dynamical and statistical behavior of the Hamiltonian Mean Field (HMF) model in order to investigate the relation between microscopic chaos and phase transitions. HMF is a simple toy model of $N$ fully-coupled rotators which shows a second order phase transition. The canonical thermodynamical solution is briefly recalled and its predictions are tested numerically at finite $N$. The Vlasov stationary solution is shown to give the same consistency equation of the canonical solution and its predictions for rotator angle and momenta distribution functions agree very well with numerical simulations. A link is established between the behavior of the maximal Lyapunov exponent and that of thermodynamical fluctuations, expressed by kinetic energy fluctuations or specific heat. The extensivity of chaos in the $N \to \infty$ limit is tested through the scaling properties of Lyapunov spectra and of the Kolmogorov-Sinai entropy. Chaotic dynamics provides the mixing property in phase space necessary for obtaining equilibration; however, the relaxation time to equilibrium grows with $N$, at least near the critical point. Our results constitute an interesting bridge between Hamiltonian chaos in many degrees of freedom systems and equilibrium thermodynamics. | Many-particle systems can show collective behavior when the average kinetic energy is small enough. This collective macroscopic behavior can coexist with chaos at the microscopic level. Such a behavior is particularly evident for systems that have a phase transition, for which a nonvanishing order parameter measures the degree of macroscopic organization, while at the microscopic level chaotic motion is a source of disorder. The latter can induce non trivial time dependence in the macroscopic quantities, and it would be desirable to relate the time behavior of such quantities and their fluctuations to the chaotic properties of microscopic motion, measured through the Lyapunov spectrum. A naive idea is that an increase of chaos as the energy (temperature) is increased should be accompanied with a growth of fluctuations of some macroscopic quantity. These should be maximal at the critical point and then drop again at high energy. In this paper we study a model of $N$ fully-coupled Hamiltonian rotators which realizes such a behavior, it has been called Hamiltonian Mean Field (HMF) model~\cite{antoni,latora}. It can also be considered as a system of interacting particles moving on a circle. This system has a second order phase transition and in the ordered phase the rotators are clustered; the high temperature phase is a gaseous one, with the particles uniformly distributed on the circle. It has been shown in ref.~\cite{latora} that the maximal Lyapunov exponent grows up to the critical energy density $U_c$ and then drop to zero in the whole high temperature phase in the $N \to \infty$ limit. Correspondingly one observes a growth of kinetic energy fluctuations up to the critical point and then a phase of vanishing fluctuations. Finite $N$ effects complicate this simple picture. In the high temperature phase the maximal Lyapunov exponent vanishes quite slowly (with $N^{-1/3}$) and finite size effects influence the first region below the critical point. In this region the system displays metastability: starting far from equilibrium, this is reached in a time $\tau_r$ which grows with $N$. On the contrary, the extremely low energy phase is characterized by a weak $N$ dependence, with the maximal Lyapunov exponent $\lambda_1$ which behaves as $\lambda_1 \sim \sqrt{U}$. Although the model is extremely simplified, it shares many features with more complex models, for which the relation between chaotic motion at the microscopic level and collective macroscopic properties has been studied. Let us mention studies in solid state physics and lattice field theory~\cite{solid,nayak,dellago,yama,lapo}. However, it has been actually in nuclear physics~\cite{ata,cmd}, where there is presently a lively debate on multifragmentation phase transition~\cite{ata,cmd,gsi,eos,bondgro,bond,perco,cmd1,mastinu}, that the interest in the connection between chaos and phase transitions has been revived. In this case in fact, an energy/temperature relation quite close to the HMF model has been observed~\cite{gsi} and critical exponents have been measured experimentally~\cite{eos}. Statistical thermodynamical models~\cite{bondgro} and percolation approaches ~\cite{perco} have been proved to give a good description of the experimental data, though the dynamics is missing. On the other hand classical molecular dynamics models \cite{cmd,bond,cmd1} seem to contain all the main ingredients, but have the disadvantage that a detailed understanding of the dynamics can be too complicated. In this respect, the HMF model can be very useful in clarifying some general dynamical features which could be eventually compared with real experimental data. In fact, when studying nuclear multifragmentation, one deals with excited clusters of 100-200 particles interacting via long-range (nuclear and Coulomb) forces. Quantum effects are relevant only at very low energy. In fact in the nuclear case, at very low energy, $T$ is not linear in $U$, but grows as $\sqrt{U}$ because nucleons are fermions~\cite{gsi,bondgro}. However, a classical picture should be quite realistic in the critical region where the excitation energy is substantial ~\cite{mastinu}. In this paper we present new numerical data concerning both statistical quantities, like specific heat and distribution functions, and chaotic probes, like Lyapunov spectra and Kolmogorov-Sinai entropy. Moreover, we add to the theoretical analysis of the model a thorough treatment of differences in the fluctuating quantities between the canonical and microcanonical ensembles. We also investigate in detail the relaxation to equilibrium and compare numerical results with a complete self-consistent Vlasov calculation of distribution functions. Finally a comparison of numerically obtained maximal Lyapunov exponents with theoretical formulas is attempted. The paper is organized as follows. In Sec. 2 we briefly discuss the details of the HMF model. The equilibrium statistical mechanics and the continuum Vlasov solution are described in Sects. 3 and 4 respectively. In Sec. 5 we discuss the relaxation to equilibrium and in Sec. 6 we present the numerical calculations of the Lyapunov spectra and Kolmogorov-Sinai entropy as a function of the energy and $N$. Analytical estimates are discussed in Sec. 7 and conclusions are drawn in Sec. 8. | We have investigated the dynamical and statistical behavior of a system with long-range forces showing a second order phase transition. Both the maximal Lyapunov exponent $\lambda_1$ and the Kolmogorov-Sinai entropy density $S_{KS}/N$ are peaked at the phase transition point, where kinetic energy fluctuations and specific heat are maximal. There is actually a small shift to lower energies due to finite size effects. The latter are present also in the Lyapunov spectra and in the Kolmogorov-Sinai entropy. Above the phase transition point, both $\lambda_1$ and $S_{KS}$ vanish as $N \to \infty$. We think that this toy model contains some important ingredients to understand the behavior of macroscopic order parameters when dynamical chaos is present at the microscopic level. Most of our findings are probably common to other Hamiltonian systems showing second order phase transitions. In particular our results could be very important in order to understand the relaxation to the equilibrium solution and the success of statistical approaches in describing the nuclear multifragmentation phase transition. \begin{ack} We thank M.C. Firpo for communicating us her results before publication and P. Holdsworth for interesting suggestions. We thank A. Torcini for many useful discussions and a careful reading of the text. A.R. thanks the Centre for Theoretical Physics of MIT for the kind hospitality and M. Robnik for stimulating discussions during his visits at CAMTP in Maribor, Slovenia. V.L. and S.R. thank INFN for financial support. S.R. thanks CIC, Cuernavaca, Mexico for financial support. This work is also part of the European contract No. ERBCHRXCT940460 on ``Stability and universality in classical mechanics". {\it In the 60's, Boris Chirikov was also an explorer of the (no man's land at that time) relation between chaotic motion and statistical behavior in classical systems with many degrees of freedom. We hope that he will be interested by this work.} \end{ack} | 98 | 3 | chao-dyn9803019_arXiv.txt |
9803 | astro-ph9803084_arXiv.txt | Using direct N-body simulations which include both the evolution of single stars and the tidal field of the parent galaxy, we study the dynamical evolution of globular clusters and rich open clusters. We compare our results with other N-body simulations and Fokker-Planck calculations. Our simulations, performed on the GRAPE-4, employ up to 32,768 stars. The results are not in agreement with Fokker-Planck models, in the sense that the lifetimes of stellar systems derived using the latter are an order of magnitude smaller than those obtained in our simulations. For our standard run, Fokker-Plank calculations obtained a lifetime of 0.28 Gyr, while our equivalent $N$-body calculations find $\sim4$ Gyr. The principal reason for the discrepancy is that a basic assumption of the Fokker-Plank approach is not valid for typical cluster parameters. The stellar evolution timescale is comparable to the dynamical timescale, and therefore the assumption of dynamical equilibrium leads to an overestimate of the dynamical effects of mass loss. Our results suggest that the region in parameter space for which Fokker-Planck studies of globular cluster evolution, including the effects of both stellar evolution and the galactic tidal field, are valid is limited. The discrepancy is largest for clusters with short lifetimes. | Theoretical models for the evolution of star clusters are generally too idealized for comparison with observations. However, detailed model calculations with direct $N$-body methods are not feasible for real globular clusters, even with fast special-purpose computers such as GRAPE-4 (Makino et al.\ 1997)\nocite{1997ApJ...480..432M} or advanced parallel computers (Spurzem \& Aarseth 1996).\nocite{1996MNRAS.282...19S} If we could scale the results of $N$-body simulations with relatively small numbers of particles (such as $\sim30,000$) to real globular clusters, then it would become feasible to perform computations with relatively small numbers of particles and still derive useful qualitative conclusions about larger, more massive systems. However, to determine the proper scaling is difficult because the ratio between two fundamental time scales, the relaxation times and the dynamical time, is proportional to $N$. In typical globular clusters, this ratio exceeds $10^3$ and the two time scales are well separated. In $N$-body simulations, the ratio is generally much smaller. The inclusion of realistic effects such as mass loss due to stellar evolution and the effect of galactic tidal fields (with the galaxy approximated as a point mass, but also with the inclusion of disc shocking) further complicate the scaling problem (see, e.g., Heggie 1996).\nocite{heg96} A proper treatment of stellar evolution is particularly problematic, since its characteristic timescale changes as stars evolve. Chernoff \& Weinberg (1990, CW90)\nocite{cw90} performed an extensive study of the survival of star clusters using Fokker-Planck calculations which included 2-body relaxation and some rudimentary form of mass loss from the evolving stellar population. In their simulations the number of particles is not specified. Their models are defined by the initial half-mass relaxation time and by the initial mass function of the cluster. Since their models do not specify the number of stars per cluster, each of their model calculations corresponds to a one-dimensional series of models, when plotted in a plane of observational values, such as total mass versus distance to the galactic center (Fig.\ 1). All points of the solid line in that figure correspond to a single calculations by CW90, since they have an identical relaxation time. As we will see later, it is useful to consider other series of models, for which the crossing time is held constant while varying the mass. An example of such a series is indicated by the dashed line in Fig.\ 1. The shapes of these lines are derived under the assumption of a flat rotation curve for the parent galaxy. The main conclusion of CW90 was that the majority of the simulated star clusters dissolve in the tidal field of the galaxy within a few hundred million years. Fukushige \& Heggie (1995, FH95)\nocite{fh95} studied the evolution of globular clusters using direct $N$-body simulation, using the same stellar evolution model as used by CW90. They used a maximum of 16k particles and a scaling in which the dynamical timescale of the simulated cluster was the same as that of a typical globular cluster, corresponding to one of vertical lines in Fig. 1. FH95 found lifetimes much longer than those in CW90's Fokker-Planck calculations, for the majority of the models used in CW90. However, the reason for the discrepancy is rather unclear, because the calculations of FH95 and those of CW90 differ in several important respects. The relaxation times differ because FH95 held the cluster crossing time fixed in scaling from the model to the real system. However, the crossing times themselves are also different, since the crossing time is by definition zero in a Fokker-Planck calculation. Finally, the implementation of the galactic tidal field is also quite different. CW90 used a simple boundary condition in energy space (spherically symmetric in physical space), in which stars were removed once they acquired positive energy, but the underlying equations of motion included no tidal term. FH95 adopted a much more physically correct treatment, including tidal acceleration terms in the stellar equations of motion and a proper treatment of centrifugal and coriolis forces in the cluster's rotating frame of reference (see FH95). \begin{figure} \centerline{ \psfig{file=fig_isomodels.ps,bbllx=570pt,bblly=40pt,bburx=110pt,bbury=690pt,height=5cm,angle=-90}} \caption {Cluster mass versus the distance to the galactic center. The solid line indicates the model parameters for which the relaxation time is constant (iso relaxation time); the dashed line indicates the initial conditions for which the crossing time of the star cluster is constant (iso crossing time) } \label{isomodels}\end{figure} In order to study the behavior of star clusters with limited numbers of stars, and to compare with the results of the Fokker-Planck simulations of CW90, we selected one of their models and perform a series of collisional N-body simulations in which the evolution of the individual stars is taken into account. According to CW90 the results should not depend on the number of stars in the simulation as long as the relaxation time is taken to be the same for all models. It is, among others, this statement which we intend to study. We find that for this set of initial conditions Fokker-Planck models do not provide a qualitatively correct picture of the evolution of star clusters. The effects of the finite dynamical time scale are large, even for models whose lifetime is several hundred times longer than the dynamical time. The main purpose of this paper is to study the survival probabilities of star clusters containing up to a few tens of thousands of single stars, in order to gain a deeper understanding of the influence of the galactic tidal field and the fundamental scaling of small $N$ clusters to larger systems. Only single stars are followed; primordial binaries are not included. The computation of gravitational forces is performed using the GRAPE-4 (GRAvity PipE, see \cite{emf+93}, a special-purpose computer for the integration of large collisional $N$-body systems). Hardware limitations (speed as well as storage) restrict our studies to $\aplt 32$k particles. The dynamical model, stellar evolution, initial conditions and scaling are discussed in Sect. 2. Section 3 reviews the software environment and the GRAPE-4 hardware, and discusses the numerical methods used. The results are presented in Sect. 4 and discussed in Sect. 5; Sect. 6 sums up. | We have followed the evolution of a star cluster, to the point of dissolution in the tidal field of the parent galaxy, taking into account both the effects of stellar dynamics and of stellar evolution. Our calculations are based on direct $N$-body integration, coupled to approximate treatments of stellar evolution. Our results differ greatly from those obtained with Fokker-Planck calculations, as presented by CW90: their model clusters dissolve after a few times $10^8$ years, whereas our equivalent model clusters live at least ten times longer. As we discussed in the previous section, a number of different reason conspire to produce such a drastic difference. Our hope was that we would be able to find a way to bridge our $N$-body results and previous results based on Fokker-Planck approximations. The fact that the GRAPE-4 special-purpose hardware allowed us to model much larger numbers of particles, reaching to within an order of magnitude of that of real globular clusters, seemed to indicate that it would finally be possible to make a firm connection between the two types of simulations. However, our results indicate that no clear process of extrapolation has emerged yet. Even within the different runs we have studied, extrapolation from the smaller to the larger number of stars would have resulted in rather large errors. This suggests that further extrapolation will suffer from the same fate. In the present paper, we have studied in detail a single model. However, the way the result depends on the time scaling might be different for other models. In a subsequent paper, we plan to carry out a systematic study, similar to the one we have presented here, for a much wider range of initial models. | 98 | 3 | astro-ph9803084_arXiv.txt |
9803 | hep-ph9803270_arXiv.txt | The idea that the universe might be open is an old one, and the possibility of having an open universe arise form inflation is not new either. However, a concrete realization of a consistent single-bubble open inflation model is known only recently. There has been great progress in the last two years in the development of models of inflation consistent with observations in such an open universe. In this overview I will describe the basic features and the phenomenological consequences of such models, making emphasis in the predictions of the CMB temperature anisotropies that differ from ordinary inflation. | The idea that the universe might be open is an old one, see e.g. \cite{Peebles}. Early attempts to accomodate standard inflation in an open universe~\cite{ratra} failed to realize that in usual inflation homogeneity implies flatness~\cite{turner}, due to the Grishchuck-Zel'dovich effect~\cite{GZE}. The possibility of having a truly open universe arise form inflation is not new either, see \cite{Gott}, via the nucleation of a single bubble in de Sitter space. However, a concrete realization of a consistent model is known only recently, the single-bubble open inflation model~\cite{singlebubble,LM}. Soon afterwards there was great progress in determining the precise primordial spectra of perturbations~[8-19], most of it based on quantum field theory in spatially open spaces. Simultaneously there has been a large effort in model building~\cite{LM,Green,induced,GBL} and constraining the existing models from observations of the temperature power spectrum of cosmic microwave background (CMB) anisotropies~[21-24]. In this review talk I will concentrate in model building and constraints from CMB anisotropies. We will describe the nature of the various primordial perturbations and give the corresponding spectra, without deriving them from quantum field theoretical arguments. The interested reader should find this in the literature. We will then compute the corresponding angular power spectra of temperature anisotropies in the CMB. Furthermore, we will give a review of the different single- and mutiple-field open inflation models and constrain their parameters from present observations of the CMB anisotropies. Sometimes this is enough to rule out some of the models. Finally, we will describe how future observations of the CMB temperature and polarization anisotropies might be able to decide among different inflationary models, both flat and open inflation ones. | Single-bubble open inflation is an ingenious way of reconciling an infinite open universe with the inflationary paradigm. In this scenario, a symmetric bubble nucleates in de Sitter space and its interior undergoes a second stage of slow-roll inflation to almost flatness. In the near future, observations of the microwave background with the new generation of satellites, MAP and Planck, will determine with better than 1\% accuracy whether we live in an open universe or not. It is therefore crucial to know whether inflation can be made compatible with such a universe. Single-bubble open inflation models provide a natural scenario for understanding the large scale homogeneity and isotropy. Furthermore, these inflationary models generically predict a nearly scale invariant spectrum of density and gravitational wave perturbations, which could be responsible for the observed CMB temperature anisotropies. Future observations could then determine whether these models are compatible with the observed features of the CMB power spectrum. For that purpose it is necessary to know the predicted power spectrum with great accuracy. Open models have a more complicated primordial spectrum of perturbations, with extra discrete modes and possibly large tensor anisotropies. In order to constrain those models we have to compute the full spectrum for a large range of parameters. In this review we have shown that the simplest single-field models of open inflation are not only fine tuned, but actually ruled out because they induce too large tensor anisotropies in the CMB, which is incompatible with present observations. On the other hand, two-field models generically do not lead to infinite open universes, as previously thought, but to an ensemble of very large but finite inflating `islands'. Each one of these islands will be a quasi-open universe. We may happen to live in one of those patches, where the universe {\em appears} to be open. This new effect, semiclassical in origin, was recently discussed in Ref.~\cite{quasi} where it was found that many of the present models are in fact quasi-open. This does not mean that they are not good cosmological models. If the co-moving size of the inflating islands is sufficiently large, then the resulting semiclassical anisotropy may be unobservable. We have shown however that such a component imposes very stringent constraints on the models. Most of them have a narrow range of parameters for which they are compatible with observations. It is perhaps worth mentioning here some alternative proposals (not single-bubble) for the generation of an open universe in the context of inflation. First of all, the group of Roma~\cite{Roma} proposed a model based on higher order gravity that induces bubble nucleation and later percolation, resulting in a distribution peaked at $\Omega_0\simeq0.2$. Perhaps the most striking recent results are those of Hawking-Turok~\cite{HawTur} and Linde~\cite{creation}, who claim that an open inflationary universe could have been created directly from the vacuum, without the intermediate de Sitter phase. | 98 | 3 | hep-ph9803270_arXiv.txt |
9803 | astro-ph9803059_arXiv.txt | Numerical simulations of galaxy clusters including two species -- baryonic gas and dark matter particles -- are presented. Cold Dark Matter spectrum, Gaussian statistics and flat universe are assumed. The dark matter component is evolved numerically by means of a standard {\it particle mesh} method. The evolution of the baryonic component has been studied numerically by using a multidimensional (3D) hydrodynamical code based on {\it modern high resolution shock capturing} techniques. These techniques are specially designed for treating accurately complex flows in which shocks appear and interact. With this picture, the role of shock waves in the formation and evolution of rich galaxy clusters is analyzed. Our results display two well differenced morphologies of the shocked baryonic matter: filamentary at early epochs and quasi-spherical at low redshifts. | Galaxy clusters are the largest systems gravitationally bounded in the Universe. Their study has been a fashion topic in Cosmology since last years. Work on topics related with galaxy clusters is worthly to: i) understand the formation, evolution, dynamics and morphology of these systems, ii) learn on the physical processes involved in them, and iii) find out some information concerning with fundamental parameters in Cosmology as density parameter ($\Omega$) , Hubble's constant ($H$), and the spectrum of the primordial density field. During last years technical improvements have produced huge quantity of data about galaxy clusters. Let us mention the new galaxy surveys (Guzzo 1996, and references therein) and the extensive observation in X-rays using satellites as ROSAT or ASCA. These huge volume of data strongly motivates a lot of theoretical work trying to explain the observational results. From the theoretical point of view, numerical simulations are the best tools to understand physics involved in galaxy clusters. At the beginning, numerical simulations of galaxy clusters where performed using N-body techniques. Since then, they have been extensively used and have produced important results ( see, e.g, Efstathiou et al. 1985, Bertschinger \& Gelb 1991, Xu 1995). Next step in the full description of galaxy clusters was to introduce in the picture a baryonic component. The numerical methods developed in order to deal with baryonic matter were more sofisticated and expensive in computational resources. As a consequence, it was not possible to carry out numerical simulations with two species (dark matter and baryonic gas) until late eighties. Cosmological hydrodynamic codes have been usually classified in two main categories: a) the so-called Lagrangian methods, like the {\it Smoothed Particles Hydrodynamics} (SPH) or ulterior extensions based on them, and b) Eulerian codes. SPH methods were first proposed by Gingold \& Monaghan (1977), and Lucy (1977). Among the best features of this technique, it should be pointed out its high resolution in dense regions. This property is directly derived from its Lagrangian character. The first implementations of SPH techniques had some weak points: i) The low density regions were badly described due to the Lagrangian character of the method. ii) Discontinuities and strong gradients were poorly solved and an important diffusion was introduced. iii) They were not conservative. Nevertheless, these previous problems were overcome in the modern implementations of these techniques. Improved SPH techniques have been widely developed for cosmological applications (see, e.g., Evrard 1988, Hernquist \& Katz 1989, Navarro \& White 1993, Gnedin 1995). Numerical cosmological codes using an Eulerian approach to study baryonic gas inside galaxy clusters have been also developed. Some of these hydro-codes use {\it artificial viscosity} in order to deal with shock waves (Cen 1992, Anninos et al. 1994). These techniques require a good calibration of the free parameters which are introduced by hand and state some numerical problems. Recently, a new family of finite difference methods, which use Eulerian approaches and avoid artificial viscosity, has been developed in numerical Cosmology. They are the so-called {\it high resolution shock capturing methods} (HRSC), the modern extensions of the original Godunov's idea (1959). According to the Riemann solver and the procedure in order to achieve spatial accuracy, we can distinguish three groups: 1) the ones following Harten's scheme (1983), like Ryu et al. (1993), 2) those using the analytical solution of the Riemann problem for the Newtonian dynamics of ideal gases and the PPM scheme described by Collela \& Woodward (1984) , like Bryan et al. (1994), and 3) the codes using Roe's Riemann solver (Roe 1981) plus the MUSCL or PPM cell reconstruction, like in Quilis et al. (1996). In this last reference, the code used in present paper is described and tested appropriately. An exhaustive comparison among all these kinds of cosmological hydrodynamic codes can be found in Kang et al. (1994). Due to the Eulerian character of our code, it does not show --in dense regions-- a resolution as good as the Lagrangian ones, and it requires more computational resources. However, HRSC schemes --by construction-- have excellent properties in order to deal with shocks, discontinuities, and strong gradients. HRSC techniques typically solve shocks in two cells. Due to their intrinsic properties, the detection of shocks is independent on the number of cells used in the simulations. It should be pointed out that this last property is really important when three-dimensional simulations are carried out. In these simulations the size of the grid is a stringent constraint due to its high cost in computational resources. Moreover, these methods are conservative by construction, that is , quantities which should be physically conserved are numerically conserved up to the order of the method. It should be also noticed that these methods show good results in extreme low density regions (Einfeldt et al. 1991). As it has been pointed out by several authors, the role of shock waves can be extremely important in order to understand the heating processes in the intracluster medium (ICM). In this paper we are interested in understanding and quantifying the role of shocks. In order to do that it is crucial to use numerical codes able to manage with complex flows. Yes, indeed, one of the important features of HRSC techniques is just to treat numerically shocks and strong discontinuities giving sharp profiles (in a few numerical cells, as we have mentioned above) independently of the size of the grid. Hence, formation, evolution, and interaction of shocks in 3D flows can be analyzed accurately with HRSC schemes, and, consequently, their use is absolutely justified in order to study shocks and their consequences on the ICM's dynamics. Hereafter, $t$ stands for the cosmological time, $t_0$ is the age of the Universe, $a(t)$ is the scale factor of a flat background. Function $\dot{a}/a$ is denoted by $H$, where the dot stands for the derivative of $a$ with respect to the cosmological time. Hubble constant is the present value of $H$; its value in units of $100 \ Km \ s^{-1} \ Mpc^{-1}$ is the reduced Hubble constant $h$. In our computations we have assumed $h=0.5\, $. Velocities are given in units of the speed of light. Baryonic, dark matter, and background mass density are denoted by $\rho_{_b}$, $\rho_{_{DM}}$, and $\rho_{_{B}}$, respectively. The density contrast is $\delta_b=(\rho_b-\rho_{_{B}})/\rho_{_{B}}$ for baryonic matter, analogously is defined $\delta_{_{DM}}$ for dark matter. The plan of this paper is as follows: In Section 2, our numerical cluster model is described. In Section 3, the results of the simulations are analyzed. Finally, a general discussion is presented in Section 4. | In this paper we have used some numerical techniques recently applied to Cosmology. These techniques , HRSC, seems to be the most suitable in order to study the role of shocks in galaxy cluster evolution. The choice is justified by their properties to handle shocks. The capability of these techniques to capture shocks with very small diffusion is independent of the resolution used in the numerical simulations. Hence, by construction, shocks are captured even using coarse grids. This property is crucial in 3D applications. Previous sections illustrate the fact that non adiabatic processes, due to shocks, take an important role in the description of the ICM. In the model presented in this paper, that is, a baryonic fluid plus dark matter component coupled gravitationally, shocks are able to heat the ICM until values compatible with observational data. The calculations have been carried out in two cases: with and without cooling processes. This procedure allows us to distinguish between non-adiabatic effects coming from shocks and the ones from cooling. The role of the cooling, even when it could be important in other scenarios, is irrelevant for the simulations considered in this paper, while shocks play the most important role. In the picture describing the dynamics of the baryonic component, there are some clues showing the presence of shocks. Examining the quantities sensitive to shocks, all of them evidence the formation of a quasi-spherical shock. This shock seems to arise around $z\sim 2$ at the cluster center and moves outwards. Nevertheless, some irregular shocks could form at $z \geq 2$. This conclusion is supported by the behaviour of the entropy profiles (see Fig. 5), and the existence of shocked cells at these times (see Fig. 8). The quasi-spherical shock would form from the collapse of the quasi-spherical global structure , while other smaller shocks -- with a filamentary morphology -- would arise from some collapsing substructure and merging processes. In short, previous discussion manifest two different regimes in the shock formation. It should be noticed that the structures simulated in this paper correspond, due to the initial conditions, to a large Abell cluster. For this kind of clusters, gravitational collapse is fast and the dynamics is violent. Shocks form earlier and are stronger than in others smaller cluster-like objects ($ < 3\sigma$). Some discussion on the numerical resolution of the simulations is needed. The one used in present paper ($\sim 0.3 Mpc$) is not enough to simulate the very center of the clusters and galaxy formation, but it suffices to study the role of the shocks in ICM. It should be kept in mind that HRSC techniques are able to resolve shocks even with coarse grids. Nevertheless, higher resolution would be desiderable to perform more complete simulations. Improvements in numerical resolution will introduce smaller scales in the problem, as a consequence, the physics of the model should be enriched in order to describe this new scenario. Chemical reactions and radiative transfer should be considered. | 98 | 3 | astro-ph9803059_arXiv.txt |
9803 | astro-ph9803090_arXiv.txt | We present a scenario for the formation and evolution of disk galaxies within the framework of an inflationary cold dark matter universe, and we compare the results with observations ranking from the present-day up to $z\sim 1$. The main idea in this scenario is that galactic disks are built-up inside-out by gas infall with an accretion rate driven by the cosmological mass aggregation history (MAH). In Avila-Reese et al. (1997) the methods to generate the MAHs of spherical density fluctuations from a Gaussian random field, and to calculate the gravitational collapse and virialization of these fluctuations, were presented. Assuming detailed angular momentum conservation during the gas (5\% of the total mass) contraction, a disk in centrifugal equilibrium is built-up within the forming dark matter halo. The primordial angular momentum is estimated through the Zel'dovich approximation and normalized to the spin parameter $% \lambda $ given by analytical and numerical studies. The disk galactic evolution is followed through a physically self-consistent approach which considers (1) the gravitational interactions among the dark halo, the stellar and gas disks, and a bulge; (2) the turbulence and energy balance of the interstellar medium; (3) the star formation process due to gas disk gravitational instabilities; and (4) the secular formation of a bulge due to the gravitational instabilities of the stellar disk. We find that the main disk galaxy properties and their correlations are basically established by the combination of three fundamental physical factors: the mass, the MAH, and the spin parameter $\lambda $. Models calculated for a statistically significant range of values for these factors predict nearly exponential disk surface brightness profiles with realistic central surface brightnesses $\mu _{B_0},$ and scale lengths (including low surface brightness galaxies), nearly flat rotation curves, and negative gradients in the B-V color index radial distribution. The main trends across the Hubble sequence of the global intensive properties such as B-V, $\mu _{B_0},$ the gas fraction $f_g$, and the bulge-to-disk ratio b/d, are reproduced. For a given mass (luminosity) B-V correlates with the maximum circular velocity, and this correlation is in agreement with the scatter of the Tully-Fisher relation. We interpret the observed color-magnitude, and ``color '' Tully-Fisher relations as a result of the empirical dependence of extinction on luminosity (mass). The model properties tend to form a biparametrical sequence, where B-V and $\mu _{B_0}$ could be the two parameters. The star formation history depends on the MAH and on the $\lambda$ parameter. A maximum in the star formation rate for most of the models is attained at $z\sim 1.5-2.5$, where this rate is approximately 2.5-4.0 times larger than the present one. The scale radii and the bulge-to-total ratio decrease with $z$, while $\mu _{B_0}$ increase. The B-band TF relation remains almost the same at different redshifts. Our scenario of disk galaxy formation and evolution reveals that the cosmological initial conditions are able to determine the main properties of disk galaxies across the Hubble sequence and predict evolutionary features for the present-day dominant galaxy population that are in agreement with very recent deep field observational studies | The understanding of the formation and evolution of galaxies is one of the clearest challenges of contemporary astrophysics and cosmology. Since galaxies are both cosmological and astronomical objects, two general approaches can be used in order to study their formation and evolution (e.g., Renzini 1994): ({\it i}) the deductive approach, through which, starting from some initial conditions given by a theory of cosmic structure formation, one tries to follow the evolutionary processes until the reconstruction of the observable properties of the galaxies and ({\it ii}) the inductive approach, in which, starting from the present-day properties of galaxies, and through galactic evolutionary models, one tries to reconstruct the initial conditions of galaxy formation; the increasing observational data on galaxies at intermediate and high redshifts will enrich this approach with crucial constraints. Most of current theories about cosmic structure formation are based on the gravitational paradigm and on the inflationary cold dark matter (CDM) cosmological models. Since these models predict more power for the small density fluctuation scales than for the larger ones, cosmic structures build up hierarchically, through a continuous aggregation of mass. From the point of view of the galaxy cosmogony, a crucial question is whether this aggregation occurs through violent mergers of collapsed substructures and/or through a gentle process of mass aggregation. This question depends on the statistical distribution of the density fluctuation field and on its power spectrum. Nevertheless, even if the dark matter structures assemble through chaotic and violent mergers of subunits, the baryon gas, because of the reheating due to the shocks implied in the collapse, virialization and star formation (SF) feedback processes, will tend to aggregate around the density peaks in a more (spatially) uniform fashion than dark matter do it. Within the framework of the hierarchical clustering theory, from the most general point of view, two could be the galaxy formation scenarios. In one case, the main properties of galaxies, including those which define their morphological types, are supposed to be basically the result of a given sequence of mergers. This picture, that we shall call the{\it \ merger scenario,} has been widely applied in semianalytical models of galaxy formation where galaxies are constructed from the cosmological initial conditions through preconceived recipes (e.g., Lacey et al. 1993; Kauffmann, White, \& Guiderdoni 1993; Cole et al. 1994; Kauffmann 1995, 1996, Baugh, Cole, \& Frenk 1996). In the other case, the formation and evolution of galaxies is related to a gentle and coherent process of mass aggregation dictated by the forms of the density profiles of the primordial fluctuations: galaxies continuously grow inside-out. We shall call this picture, firstly developed by Gunn (1981, 1987), and by Ryden \& Gunn (1987), the e{\it xtended collapse scenario. }% Since disk galaxies ($\sim 80\%$ of present-day normal galaxies) could not have suffered major mergers due to the dynamical fragility of the disks (T\'{o}th \& Ostriker 1992), the extended collapse scenario results more appropriate to study their evolution. According to the merger scenario, the bulges of spiral galaxies and the elliptical galaxies arise from the mergers of galactic disks. A natural prediction of this scenario is that spirals with small bulge-to-disk ratios should have bulges older than those of spirals with large bulge-to-disk ratios (e.g., Kauffmann 1996). As Wyse, Gilmore, \& Franx (1997) have pointed out this does not appear compatible with recent observational data (de Jong 1996a; Peletier \& Balcells 1996; Courteau, de Jong, \& Broeils 1997). On the other hand, if elliptical galaxies are the product of relatively recent mergers, then a big dispersion is expected in their color-magnitude relationship (but see Kauffmann 1996). Bower, Lucey, \& Ellis (1992) showed that for the ellipticals in the Coma Cluster, this relationship is extremely tight. Ellis et al. (1997) confirmed this result for ellipticals in intermediate redshift clusters, up to $z\sim 0.6.$ The merger scenario could also have serious difficulties from the dynamical point of view: it is not conclusive if mergers of disks are able to reproduce the high central phase-space densities of elliptical galaxies (e.g. Hernquist 1993). The inductive approach yields the possibility to establish several constraints to the galaxy formation and evolution processes. Galactic evolutionary models{\it } have shown that due to the rapid disk gas consumption in SF, closed models are not able to explain several properties of disk galaxies, as well as the wide range of colors, gas fractions, etc. that galaxies present across the Hubble sequence (e.g., Larson \& Tinsley 1978; Tinsley 1980; Larson, Tinsley, \& Caldwell 1980; Kennicutt 1983; Gallagher, Hunter, \& Tutukov 1984; Firmani \& Tutukov 1992, 1994). On the other hand it was shown that the SF time scale in disk galaxies is not controlled by the initial gas surface density (Kennicutt 1983; Kennicutt, Tamblyn, \& Congdon 1994). Hence, models where gas accretion is introduced are more realistic. Gas accretion could also be necessary to maintain spiral structure. In the case of open models, galaxy formation and galactic evolution might be two related processes where the SF time scale is driven by the gas accretion rate at which the disk is being built up. Infall models of disk galaxy formation have been recently favored by studies of our own Galaxy and nearby galaxies (see for references Cay\'{o}n, Silk, \& Charlot 1996). Moreover these inside-out disk formation models seem also to be in agreement with constrictions provided by deep field observations (e.g., Bouwens, Cay\'{o}n, \& Silk 1997; Cayon et al. 1996; see also Section 4). The gas infall rate in luminous galaxies may be controlled by the global process of galaxy formation (cosmological accretion) and/or by a self-regulated process of SF formation. This latter process proposed by White \& Rees (1978) and White \& Frenk (1991) is commonly applied in the merger scenario models. According to this mechanism, the gas accretion rate is driven by the cooling of the hot gas corona sustained by the supernova-injected energy. In the extreme situation of instantaneous galaxy formation, supernova gas reheating, halo self-regulated SF, and cooling flows (if the reheated gas was not completely expelled out of the system) become the dominant processes in regulating luminous galaxy evolution. However, the self-regulated halo SF model suffers from some inconsistencies. As Nulsen \& Fabian (1996) pointed out, supernova feedback over large scales occurs on roughly the same time scales as the SF, not fast enough to tightly regulate the SF rate. In the same way, if a disk forms, then the self-regulating mechanism of SF will apply to the disk where other dynamical conditions prevail (see Firmani, Hern\'{a}ndez, \& Gallagher 1996), and not to the halo system. Unless the disk-halo connection is very effective, the SF in the disk will not be regulated by a balance of energy between the supernova input and the halo gas cooling. On the other hand, the X-ray gas corona predicted by the self-regulated SF mechanism lacks observational support, at least for the most massive galaxies for which the X-ray emission would have been above the minimum detection limits of the Rosat and ASCA experiments. The galactic infall models suggested by the inductive approach, are consistent with the cosmological (deductive) extended collapse scenario of galaxy formation and evolution. In Avila-Reese, Firmani, \& Hern\'{a}ndez (1997, hereafter AFH), within the framework of a standard CDM model, the MAHs corresponding to fluctuations of galactic scales were generated from the statistical properties of a Gaussian random field. After calculating the virialization of the fluctuations, a range of realistic dark halo structures were obtained. Now, with the aim to explore whether these cosmological initial conditions are able to predict the evolutionary and observational disk galaxy properties and their correlations, particularly those which go across the Hubble sequence (HS), we shall construct a self-consistent and unified model of disk galaxy formation and evolution in the cosmological context. Within the framework of the extended collapse scenario and using the galactic evolutionary models of Firmani et al. (1996), we shall study the formation and evolution of disks in centrifugal equilibrium into the evolving dark halos. In section 2, the methods we use are described. The model results at $z=0,$ the main predictions of the models, and the comparisons with observations as regards the local (\S 3.1) and global galactic properties and their main correlations (\S 3.2) are presented in section 3. In section 4 we compare our evolutionary models with observations at intermediate redshifts ($z\lesssim 1)$. Finally, the concluding remarks are given in section 5. | We have modeled the formation and evolution of disk galaxies within the framework of the extended collapse scenario, which is based on the inflationary CDM models. The gas disks in centrifugal equilibrium were built-up under the assumption of detailed angular momentum conservation into spherical virializing dark matter halos whose MAHs were calculated from the initial cosmological conditions. The disk SF is produced by global gravitational instabilities and is self-regulated by an energetic balance of the turbulent gas. The bulges are formed by secular evolution of the stellar disk based on gravitational instabilities. The main predictions of the models are: 1). The disks present exponential surface brightness profiles and negative radial B-V gradients. The scale lengths and central surface brightnesses are in agreement with the observations, including the LSB galaxies. 2). The rotation curves are nearly flat up to the Holmberg radius. Contrary to observational estimations, the rotation curve decompositions show dominion of dark matter down to the galaxy central regions. A constant density core in the dark halo solves this problem. 3). The intensive properties and their correlations (particularly those which go across the HS) of the models corresponding to the local ($z\approx 0)$ population of disk galaxies, including the LSB galaxies, are determined by the combination of three fundamental physical factors and their statistical distributions, related to the initial cosmological conditions. These three factors are the mass, the MAH, and the primordial angular momentum expressed through the spin parameter $\lambda .$ 4). The intensive properties of the models can be described in a biparametrical sequence, where the parameters may be the color index B-V and the central surface brightness $\mu_{B_o}$. Each one of these parameters is determined mainly by the MAH and $\lambda $, respectively. The third fundamental physical factor, the mass, exerts no practical influence the intensive properties. We have shown that the empirical luminosity (mass)-color relation (or equivalently the color TF relation) can be explained by the effects of the metallicity and the extinction.observed dependence of extinction on luminosity (mass). These effects also contribute to decrease the slope of the B-band TF relation. 5). The SF rates of models with the average MAHs and $\lambda =0.05$ grow by factors of 2.5-4.0 up to $z\sim 1.5-2.5$ with respect to the SF rates at $z=0 $. After this maximum, the SF rates slowly decrease with $z.$ The SFHs of systems with early, active MAH and/or low $\lambda ^{\prime }s$ show high SF rates at high redfshifts ($z>3),$ while the systems with extended MAHs and/or high $\lambda ^{\prime }s$ present small SF rates which slowly increase until the present epoch. 6). The structural properties of the models do not change abruptly. Between $z=0$ and $z\approx 1$ the disk scale radii in average decrease a factor $\sim 1.3$ and the central surface brightnesses increase $\sim 1$ mag/arcsec$^2.$ The bulge-to-total luminosity ratio also decreases with $z$ and decreases more severely for the low mass systems. The slopes of the ``structural'' and B-band TF relations do not change with $z.$ In the case of the ``structural'' TF relation, the zero-point decreases (0.75 mag for $z=0.7$ with respect to $z=0$), while for the B-band TF relation the zero-point slightly increases (0.1 mag at $z=0.7$ with respect to $z=0).$ The exploratory models presented in this work show that the main observational characteristics and correlations of disk galaxies can be well understood in the context of the extended collapse scenario, suggesting a direct connection between the conditions prevailing in the early universe and the properties of galaxies today. A serious shortcoming of galaxies emerging from Gaussian CDM cosmological models is the gravitational dominion of DM over baryon matter. The remedy to this problem is the introduction of a core in the DM halo. Fortunately, the intensive galaxy properties and their correlations are not significantly sensitive to the existence or non-existence of such a core, in such a way that all the results presented here are also true for galaxies with a core in their DM halos. The main limitations of our approach are connected to the facts that {\it (i)} the influence of the environment on galaxy formation and evolution was not taken into account, {\it (ii)} detailed angular momentum conservation for the baryon gas collapse was assumed, and {\it (iii)} the mass aggregation was treated only as gas accretion neglecting the possibility of mergers of stellar systems. In future we shall address ways of overcoming these limitations with the aim to improve the model predictions and to explain galactic properties and distributions related to environment. | 98 | 3 | astro-ph9803090_arXiv.txt |
9803 | astro-ph9803329_arXiv.txt | Magnetic field-aligned electric fields are characteristic features of magnetic reconnection processes operating in externally agitated magnetized plasmas. An especially interesting environment for such a process are the coronae of accretion disks in active galactic nuclei (AGN). There, Keplerian shear flows perturb the quite strong disk magnetic field leading to intense current sheets. It was previously shown that given field strengths of 200 G in a shear flow, reconnection driven magnetic field aligned electric fields can accelerate electrons up to Lorentz factors of about 2000 in those objects thus providing us with a possible solution of the injection (pre-acceleration) problem. However, whereas in the framework of magnetohydrodynamics the formation of the field-aligned electric fields can be described consistently, the question has to be addressed whether the charged particles can really be accelerated up to the maximum energy supplied by the field-aligned electric potentials, since the accelerated particles undergo energy losses either by synchrotron or inverse Compton mechanisms. We pre\-sent relativistic particle simulations starting from electric and magnetic fields obtained from magnetohydrodynamic simulations of magnetic reconnection in an idealized AGN configuration including nonthermal radiative losses. The numerical results prove that the relativistic electrons can be effectively accelerated even in the presence of an intense radiation bath. Energies from~$50 \Dim{MeV}$ up to~$40 \Dim{GeV}$ can be reached easily, depending on the energy density of the photon bath. The strong acceleration of the electrons mainly along the magnetic field lines leads to a very anisotropic velocity distribution in phase space. Not even an extremely high photon energy density is able to completely smooth the anisotropic pitch angle distribution which is characteristic for quasi monoenergetic particle beams. | \label{sec:intro} Active galactic nuclei (AGN) can be regarded as accreting supermassive black holes surrounded by accretion disks (Camenzind 1990; Miyoshi et al. 1995; Burke and Graham-Smith 1997 and references therein). Relativistic electrons in AGN reveal themselves by hard X-ray (probably due to pair production (cf. Svensson 1987; Done and Fabian 1989)) and $\gamma$-ray emissions as well as radio observations of superluminous motions (e.g. Abraham et al. 1994). The $\gamma$-radiation observed from quasars and bla\-zars may originate in a distance $R$ of $10^{2-3}$ gravitational radii from the central engine (Dermer and Schlickeiser 1993). At that distance no significant pair production happens but relativistic leptons scatter via the inverse Compton process the IR-UV radiation of the accretion disk within a relativistically moving jet. It is well known that ``standard'' mechanisms for the acceleration of high energy leptons, as diffusive shock wave acceleration and resonant acceleration by magnetohydrodynamical (MHD) turbulence can only work efficiently for Lorentz factors larger than $\gamma_{\rm crit}\simeq m_{\rm p}/{m_{\rm e}}$ (where $m_{\rm p}$ and $m_{\rm e}$ denote the proton and electron masses). Consequently, charged particles accelerated via shocks or MHD~turbulence have to be pre-accelerated which confronts us with the {\bf injection problem} (Blandford 1994; Melrose 1994) in the AGN context. In a differentially rotating magnetized accretion disk gas, the evolution of a magnetized corona is quite hard to suppress (Galeev et al. 1979; Stella \& Rosner 1984). Driven by the buoyancy force magnetic flux tubes ascend into the disk corona, thereby their footpoints are sheared by the differential rotation of the disk. Either by internal shear or by encountering already pre\-sent magnetic flux, magnetic reconnection and accompanied rapid dissipation of magnetic energy happens in the coronal plasma. Such a behavior can be studied with great detail for example in the solar corona (Parker 1994 and references therein). In a recent contribution we investigated the possible role of magnetic field-aligned electric fields ($E_\parallel $) in the context of magnetic reconnection operating in AGN coronae for the pre-ac\-cel\-er\-ation of leptons (Lesch and Birk 1997; hereafter LB). It could be shown that field-aligned electric potential structures in relatively thin current sheets form. For reasonable physical parameters such electric potentials are strong enough to accelerate electrons up to $\gamma \approx 2000$, in principle. However, in the framework of MHD the actual energies of the accelerated particles cannot be calculated. It is the aim of the present contribution to corroborate the model introduced in LB with the help of relativistic particle simulations by taking macroscopic electric and magnetic field configurations obtained by the MHD simulations as an input. In the next section we resume the MHD model in a nutshell and present the details of the resulting three-dimensional electric and magnetic fields. In Sec.~\ref{sec:simulation} we discuss our approach to the numerical study of high-energy particles and show the numerical results dwelling on particle spectra and energies. Eventually, we discuss our findings in Sec.~\ref{sec:Disc}. | \label{sec:Disc} We addressed the question of charged particle acceleration during magnetic reconnection processes by means of relativistic test particle simulations. Whereas this point is crucial for a great variety of cosmic plasmas in this contribution we dwelled on the pre-acceleration problem in the AGN context. Possible further applications include among others as different plasma systems as the terrestrial discrete auroral arcs, the solar coronal flares, radio activity in T-Tauri magnetospheres, non-thermal emission at the edges of high-velocity clouds that hit the galactic plane and the generation of electron beams in the magnetospheres of neutron stars. The starting point for our present investigations are results obtained by an MHD simulation study (LB). In contrast to previous work carried out by different other groups, we were able to study particle acceleration in large-scale non-linearly developed reconnection electromagnetic fields rather than being restricted to the prescription of somewhat idealized analytical field solutions. Moreover, since for the considered parameter regime we have to expect an intense radiation field the test particle simulations were performed including the relevant radiative losses. A very important question is whether charged particles can really be accelerated in the reconnection region up to the pretty high energy values one might deduce from the fluid treatment. Our findings indicate that, in fact, as expected from the fluid simulations (LB), leptons can be accelerated in reconnection zones located in AGN coronae up to high Lorentz factors; i.e. a significant portion of test particles gain about the maximum energy provided by the generalized field-aligned electric potential structures formed during the magnetic reconnection processes we have modeled within the framework of MHD. Thus, particle acceleration in reconnection zones may be considered as a way out of the injection problem we face in the AGN context. Since sheared magnetic fields can be expected as a very common phenomena in cosmic environments like accretion disks, stellar coronae and interstellar medium, we think that our relativistic particle studies can be regarded as realistic with respect to the used configuration and the dominant forces. In this contribution the magnetic reconnection process is driven by some resistive mechanism originating from plasma instabilities. The excited electromagnetic oscillations serve as resistance. We note that in plasmas which are collisionless (both no Coulomb collisions and no turbulent wave excitation), the particle inertia presents the ultimate source of resistivity and for the magnetic dissipation. The sheared magnetic fields in collisionless systems evolve into very thin filaments, in which the lifetime of the particle determines the electrical conductivity, thereby allowing for efficient dissipation via effective particle acceleration (Lesch and Birk 1998). Our simulations are test--particle simulations, thus, we plan for future studies to include, additionally, ponderomotive forces and the back reaction of the current carried by the high energy particles. Whether or not the latter aspect becomes important depends on the density of the run-away electrons limited by the Dreicer electric field (e.g. Benz 1993). | 98 | 3 | astro-ph9803329_arXiv.txt |
9803 | cond-mat9803258_arXiv.txt | The adsorption of large ions from solution to a charged surface is investigated theoretically. A generalized Poisson--Boltzmann equation, which takes into account the finite size of the ions is presented. We obtain analytical expressions for the electrostatic potential and ion concentrations at the surface, leading to a modified Grahame equation. At high surface charge densities the ionic concentration saturates to its maximum value. Our results are in agreement with recent experiments. | 98 | 3 | cond-mat9803258_arXiv.txt |
||
9803 | astro-ph9803165_arXiv.txt | A method offering an order of magnitude sensitivity gain is described for using quasar spectra to investigate possible time or space variation in the fine structure constant $\alpha$. Applying the technique to a sample of 30 absorption systems, spanning redshifts $0.5<z<1.6$, obtained with the Keck I telescope, we derive limits on variations in $\alpha$ over a wide range of epochs. For the whole sample $\Delta \alpha /\alpha =-1.1\pm 0.4\times 10^{-5}$. This deviation is dominated by measurements at $z>1$, where $\Delta \alpha /\alpha = -1.9\pm 0.5\times 10^{-5}$. For $z<1$, $\Delta \alpha /\alpha = -0.2\pm 0.4\times 10^{-5}$, consistent with other known constraints. Whilst these results are consistent with a time-varying $\alpha$, further work is required to explore possible systematic errors in the data, although careful searches have so far not revealed any. | 98 | 3 | astro-ph9803165_arXiv.txt |
||
9803 | astro-ph9803023_arXiv.txt | We undertake a quantitative investigation, using Monte Carlo simulations, of the amount by which quasars are expected to exceed radio galaxies in optical luminosity in the context of the `receding torus' model. We compare these simulations with the known behaviour of the \OIII~$\lambda$5007 and \OII~$\lambda$3727 emission lines and conclude that \OIII\ is the better indicator of the strength of the underlying non-stellar continuum. | It is widely believed that radio-loud quasars and radio galaxies differ from each other only in terms of the angle between their radio axes and the line of sight (Scheuer 1987; Barthel 1989). Quasars are observed fairly close to the line of sight ($\simlt 45\degree$) and therefore frequently exhibit the effects of beaming, such as superluminal motion and luminous flat-spectrum radio cores, whereas radio galaxies are observed with their axes close to the plane of the sky, and do not show these effects. However, quasars also possess a luminous non-stellar optical continuum and broad emission lines, which are absent in radio galaxies. It has therefore been proposed that there is material around the nucleus which lies preferentially in the plane perpendicular to the radio axis and obscures the central regions from view in radio galaxies. Because of the geometry of this material, it is often referred to as the ``torus'', but other geometries (e.g.\ a warped disk) are possible. Although the broad lines are hidden from direct view by this material, the narrow line region (NLR) is much larger in size and should be less strongly affected. It is therefore possible that narrow line luminosity could be an isotropic measure of the strength of the non-stellar continuum that is otherwise invisible in radio galaxies. Jackson \& Browne (1990) tested this idea by comparing the \OIII~$\lambda$5007 luminosities of quasars and radio galaxies with similar redshifts and extended radio luminosities. They reported that quasars are 5--10 times more luminous in this line than radio galaxies, and attributed the difference to higher extinction in radio galaxies. Hes, Barthel \& Fosbury (1993) performed a similar test using the \OII~$\lambda$3727 doublet and found that there was no measurable difference between the line luminosities of the two classes. Since \OII\ is at a shorter wavelength than \OIII\ and would be more greatly affected by a foreground screen of reddening, they suggested that most of the \OIII\ emission is produced close to the nucleus and is still obscured by the torus in radio galaxies, whereas \OII\ is produced farther out and is unaffected. Unfortunately, Jackson \& Browne's analysis was biased because, although the radio galaxies were selected from the 3CR-based complete sample of Laing, Riley \& Longair (1983; also called 3CRR), the quasars were drawn from a number of incomplete surveys and were therefore subject to uncertain selection effects, most notably the tendency to preferentially include optically bright objects. Since line and continuum luminosities are very well correlated in quasars, their result would be biased towards finding systematically higher line luminosities in the quasars. In addition, Jackson \& Browne's radio galaxy sample included some objects with very weak emission lines (Class~B optical spectra; Hine \& Longair 1979) that it is now believed may not belong to the unified scheme. An unbiased analysis significantly reduces the magnitude of Jackson \& Browne's result, but the quasars are still about twice as luminous in \OIII\ than the radio galaxies. This result is also seen in the higher \OIII/\OII\ ratios in broad-line objects than in narrow-line objects (e.g.\ Saunders et al.\ 1989; Tadhunter et al.\ 1993). At higher redshift, however, the \OIII\ luminosities of the two classes of object are comparable (Jackson \& Rawlings 1997). Rawlings \& Saunders (1991) note that the tendency for quasars to have a higher emission line luminosity than radio galaxies could be the result of a classification bias. This could explain the discrepancy in \OIII\ luminosities between the two classes, but appears to run counter to the similarity in the \OII\ line emission. In this {\it Letter\/}, we apply the simple receding torus model (Lawrence 1991; Hill, Goodrich \& Depoy 1996) to provide a quantitative explanation of the difference in \OIII\ luminosities between quasars and radio galaxies (and its redshift dependence), and explain why a similar effect is not seen in \OII. The free parameters involved in our explanation are constrained by observed quantities independent of the emission line luminosities. We conclude that the luminosity of the \OIII~$\lambda$5007 line is an excellent indicator of the strength of the underlying non-stellar continuum. | We have used a simple receding torus model, whose free parameters are constrained by observation, to show that low redshift 3CRR quasars should be, on average, about twice as luminous in their ionizing continua as radio galaxies of the same radio luminosity. This difference should also be seen in their \OIII, but not their \OII, emission line luminosities, in agreement with observation. For samples with a higher quasar fraction, such as the high redshift 3CRR objects, the difference in ionizing luminosities between quasars and radio galaxies should be smaller, and there should therefore be less of a difference in their \OIII\ luminosities, again in line with observation. This model leads to the conclusion that the \OIII~$\lambda$5007 emission line, and not the \OII~$\lambda$3727 doublet, is an unbiased indicator of the intrinsic optical--ultraviolet luminosity of both quasars and radio galaxies. | 98 | 3 | astro-ph9803023_arXiv.txt |
9803 | astro-ph9803215_arXiv.txt | We report an upper limit of $9\times 10^{12}$ cm$^{-2}$ on the column density of water in the translucent cloud along the line-of-sight toward HD 154368. This result is based upon a search for the C-X band of water near 1240 \AA\ carried out using the Goddard High Resolution Spectrograph of the Hubble Space Telescope. Our observational limit on the water abundance together with detailed chemical models of translucent clouds and previous measurements of OH along the line-of-sight constrain the branching ratio in the dissociative recombination of H$_3$O$^+$ to form water. We find at the $3\sigma$ level that no more than 30\% of dissociative recombinations of H$_3$O$^+$ can lead to H$_2$O. The observed spectrum also yielded high-resolution observations of the Mg II doublet at 1239.9 \AA\ and 1240.4 \AA, allowing the velocity structure of the dominant ionization state of magnesium to be studied along the line-of-sight. The Mg II spectrum is consistent with GHRS observations at lower spectral resolution that were obtained previously but allow an additional velocity component to be identified. | Translucent clouds have total extinctions $A_{\rm V}$ in the range of 2-5 magnitudes. The physical and chemical conditions that characterize translucent clouds are therefore intermediate between those in diffuse and in dense molecular clouds (Crutcher 1985; van Dishoeck \& Black 1988). Photodissociation rates are significantly smaller in translucent clouds than in diffuse clouds, and the column densities of molecules like CO, OH and CS are correspondingly larger. Translucent clouds can be observed through absorption lines of CN, CH, CH$^+$ and C$_2$ as well as through millimeter emission lines of CO and CS. Although they are an abundant constituent of dense molecular clouds (Jacq et al.\ 1988; Zmuidzinas et al.\ 1995; Gensheimer et al.\ 1996; van~Dishoeck \& Helmich 1996; Cernicharo et al.\ 1997) and are expected to be the dominant coolant of warm dense regions within such clouds (Neufeld \& Kaufman 1993; Neufeld, Lepp \& Melnick 1995), water molecules have not been detected in diffuse or translucent molecular clouds. In this paper, we report the results of a search for water in the translucent cloud along the line of sight to HD 154368. HD 154368 is an O9.5 Iab star at an estimated distance of 800 pc from the Sun (Snow et al.~1996) that is situated near the Sco OB 1 association (Blades \& Bennewith 1973). The line-of-sight toward HD 154368 lies about $\rm 14^o$ from the core of the $\rho$ Oph molecular cloud, which is located at a distance of 125$\pm$25 pc (de Geus, de Zeeuw, \& Lub 1989) and has a heliocentric radial velocity of $\rm -6.6\, km \,s^{-1}$. The main H I (e.g.\ Riegel \& Crutcher 1972) and Na I (e.g.\ Crawford, Barlow, \& Blades 1989) absorption features are observed at a radial velocity that is close to that of the $\rho$ Oph molecular cloud, a result that suggests that the main component of gas toward HD 154368 is probably not close to the star and is more likely to be the outer envelope of a dense molecular cloud only about 125 pc from the Sun. The color excess along the line of sight is E(B--V)=0.82 and most of the gas toward HD 154368 resides in two clouds centered near --3.26 (main component) and --20.95 km s$^{-1}$ (heliocentric). This line-of-sight has been observed extensively from the ground by means of narrow H I 21 cm absorption features (e.g.~Riegel \& Crutcher 1972); optical absorption lines of Na~I~D (e.g.~Crawford et al.~1989), CN B--X (0,0) and A--X (0,0), CH, CH$^+$; and near-infrared absorption lines of C$_2$ in the A--X Phillips system at 8750 \AA\ (van Dishoeck \& de Zeeuw 1984). The CH observations can be used to infer the total column density of H$_2$ along the line-of-sight (Gredel, van Dishoeck, \& Black 1993), a quantity that is needed to determine the atomic depletions. The red system of CN has been used by Gredel, van Dishoeck, \& Black (1991) to derive an electron density of 0.05-0.15 cm$^{-3}$ for the molecular component. Molecular emission lines are also detectable toward HD 154368. Data on $^{12}$CO $J=1-0$, $J=2-1$, and $J=3-2$ and $^{13}$CO $J=1-0$ have been presented by van Dishoeck et al.~(1991), and have been used to constrain the column density of CO and the density of H$_2$ in the molecular component. These authors confirmed the relatively low density $n_{\rm H}\approx 350$ cm$^{-3}$ derived independently from the C$_2$ absorption data. The $^{12}$CO $J=1-0$ distribution over a region of $30'\times 30'$ around the star is featureless, a result that suggests once more that the cloud and the star are not located close to one another. The velocity structure of the gas toward HD 154368 is well known through the Na I data obtained using the Ultra High Resolution Facility at the Anglo Australian Telescope (Snow et al.~1996) and high resolution Ca II observations (Crawford 1992). The Na I data indicates seven velocity components at --27.7, --20.95, --18.2, --14.75, --10.5, --3.26, and +5.62 km s$^{-1}$ where 96\% of the neutral sodium resides in the --3.26 km s$^{-1}$ feature. The Ca II results show five velocity components at --27.6, --20.9, --14.4, --4.3, and +6.8 km s$^{-1}$. These five coincide roughly with the Na I components, but their relative column densities are different. Approximately 50\% of the ionized calcium resides in the --4.3 km s$^{-1}$ feature with an even distribution over the remaining velocity components. A detailed discussion of all these observations and the implications they have for the physical conditions of the ambient medium can be found in Snow et al.~(1996). In this work, their results will be adopted and the main focus will be on the resulting OH and H$_2$O chemistry. | We have reported a $3\sigma$ upper limit of $9\times 10^{12}$ cm$^{-2}$ on the column density of water toward HD 154368. We have constructed detailed chemical models which incorporate many existing constraints on the physical conditions of the ambient medium. Together with the known column density of OH along the line of sight, we constrain the H$_2$O branching ratio for dissociative recombination of H$_3$O$^+$ to be smaller than 30\%. Our results are in agreement with the laboratory studies of Williams et al.\, which find an oxygen channel in the recombination of H$_3$O$^+$ and a small branching ratio for the production of water, but are mildly inconsistent with the laboratory results of Vejby-Christensen et al. The observed spectrum of HD 154368 also yielded high-resolution observations of the Mg II doublet at 1239.9 \AA\ and 1240.4 \AA, allowing the velocity structure of the dominant ionization state of magnesium to be studied along the line-of-sight. The Mg II spectrum is consistent with GHRS observations at lower spectral resolution that were obtained previously but allow an additional velocity component to be identified. Absorption by the Ge II line at 1237.059 \AA\ was also detected, the first detection of interstellar germanium along the line-of-sight to HD 154368. We acknowledge with gratitude the support of NASA grant NAGW-3147 from the Long Term Space Astrophysics Research Program and of grant GO-06739.01-95A from the HST Cycle 6 Program. In the final stages of this project MS was partially supported by NASA through grant HF-01101.01-97A, awarded by the Space Telescope Institute, which is operated by the Association of Universities for Research in Astronomy, Inc., for NASA under contract NAS 5-26555. Most of the simulations presented in this work were performed on the Cray YMP operated by the Netherlands Council for Supercomputer Facilities in Amsterdam. \newpage | 98 | 3 | astro-ph9803215_arXiv.txt |
9803 | astro-ph9803287_arXiv.txt | ASCA observations of NGC 4636 and a southern region have revealed extended X-ray emission to a radius of about 300 kpc from the galaxy. The symmetric nature of the observed surface brightness around NGC 4636 indicates its association to this galaxy rather than to the Virgo cluster. Model independent estimation of the gravitational mass profile shows a flattening at a radius of $20 \sim 35$ kpc, where the total mass reaches $\sim 6\times 10^{11} M_\odot$ and a mass-to-light ratio of 23. The mass still increases to larger radii, reaching $9\times10^{12} M_\odot$ with a mass-to-light ratio of 300 at $\sim$ 300 kpc from NGC~4636. These features suggest presence of a galaxy group surrounding NGC 4636. If such optically dark groups are common among X-ray bright ellipticals, it would explain the very large scatter in their X-ray luminosities with similar optical luminosities. | Spatial distribution of gravitational mass around elliptical galaxies has been estimated from radial distribution of X-ray emitting hot interstellar medium (ISM) (e.g. \cite{3}; \cite{4}). However, the outermost radius of the ISM distribution has so far remained undetermined due to a lack of sensitivity. As a result, the total gravitating masses of elliptical galaxies depend heavily on the assumed overall extent of the ISM\@. High sensitivity X-ray observations of bright galaxies are therefore important in determing the mass distribution in these systems. NGC~4636 is one of the nearest giant elliptical galaxies. We employ a distance of 17 Mpc (\cite{8}). Diffuse thermal X-rays from its ISM make it one of the X-ray brightest elliptical galaxies (e.g. \cite{3}; \cite{4}) except those located at the center of clusters. Although NGC~4636 is located in the Virgo Southern Extension (\cite{nol}), it is apparently clear of the large-scale X-ray emission from hot intra-cluster gas associated with the Virgo cluster (\cite{10}; \cite{11}). Trinchieri et al.\ (1994), using {\it ROSAT}, detected a largely extended X-ray emission up to 18$'$ from NGC~4636, and suggested it to be an extension of the Virgo cluster emission. Here, we present new results on the extended X-ray emission around NGC~4636 based on a very long observation towards the galaxy center and on an offset observation from ASCA. Spectral properties of NGC~4636 have been already reported by Matsushita et al.\ (1997). | ASCA observations have shown a very extended X-ray emission with a radius of at least $60'$ (300 kpc) surrounding NGC~4636, which is much larger than that detected by {\it ROSAT}\@. We studied radial temperature and density distributions of the gas, and determined gravitational mass profile with and without assuming a double-$\beta$ model. According to Fig.\ 3, the gravitational halo of NGC~4636 appears to terminate once at $R \sim 10-30$ kpc, where the total mass reaches several $ \times 10^{11} ~ M_\odot$. This inner component probably corresponds to the mass associated with the galaxy NGC~4636. Then, the total mass starts increasing again, forming a halo-in-halo structure whose total size is comparable to that of a galaxy group. The mass of the whole system and its mass-to-light ratio at 300 kpc reach $\sim 1 \times 10^{13}~ M_\odot$ and $\sim 300$, respectively, which are again comparable to those of galaxy groups (\cite{28}). At $R \ge 200$ kpc, the X-ray emitting plasma becomes the dominant form of baryons, with its mass reaching 5--8\% of the total gravitating mass. This value is considerably higher than those of individual elliptical galaxies, and again close to those found in galaxy groups and poor clusters (\cite{28}). These features all support the presence of a gravitational potential with a size of a galaxy group around NGC~4636. A further support is given by the observed abundance decrease in the X-ray emitting plasma (\cite{12}; \cite{15}; \cite{matusita97}; \cite{17}), from $\sim 1$ solar within $\sim 30$ kpc, to $\sim 0.2$ solar beyond $\sim 50$ kpc to a is typical level of groups and clusters of galaxies. Nolthenius (1993) identified % NGC~4636 as a member of Virgo Southern Extension F Cloud. However the 3-dimensional location of NGC~4636 is far offset from the center of the member galaxies in the cloud, and their velocity dispersion, 463 km s$^{-1}$, is much larger than that inferred from the observed gas temperature. The diameter of the cloud, 1.26 Mpc, is also much larger than the scale of the X-ray emission. Therefore, the relation between the X-ray halo and the galaxy cloud remains unclear, but it is very likely that there is some galaxy concentration around NGC~4636. Extensive studies of ASCA data have revealed (\cite{17}) that X-ray luminous elliptical galaxies preferentially possess large-scale X-ray halos of a few 100 kpc scale, just like NGC~4636. Such a galaxy may be regarded as a dominant member of a galaxy group even though its evidence is scarce in the optical data. Thus, the X-ray emitting plasma may provide a better tracer of the total mass distribution than the light emitting matter. We speculate that X-ray luminous elliptical galaxies may commonly possess such an extended emission with low surface brightness. X-ray luminosity of elliptical galaxies scatter by 2 orders of magnitude for the same optical luminosity, and its cause has long been a puzzle. The presence and absence of the extended emission can easily account for the large scatter in the total X-ray luminosity, and the low surface brightness of the extended emission, such as in NGC~4636, explains the previous undetection. We hope that systematic deep exposures from ASCA will solve this problem and bring us a complete understanding of the mass distribution around elliptical galaxies. \medskip K.M. acknowledges support by the Postdoctoral Fellowship of the Japan Society for Promotion of Science. \clearpage | 98 | 3 | astro-ph9803287_arXiv.txt |
9803 | astro-ph9803078_arXiv.txt | Recent hard X-ray spectroscopy of active galactic nuclei has strongly suggested that double-peaked, very broad Fe K emission arises from an accretion disk around the central engine. Model fitting of the observed Fe K emission line profile makes it possible to estimate a probable inclination angle of the accretion disk. In order to study the geometrical relationship between the accretion disk and broad emission-line regions (BLRs), we investigate the correlation between the inclination angle of the accretion disk and the velocity width of BLRs for 18 type-1 Seyfert galaxies. We found that there may be a negative correlation between them, i.e., Seyfert nuclei with a more face-on accretion disk tend to have larger BLR velocity widths, suggesting that the BLRs are not coplanar with respect to the accretion disk. The most probable interpretation may be that the BLRs arise from outer parts ({\it r} $\sim$ 0.01 pc) of a warped accretion disk illuminated by the central engine. | Given the current paradigm of active galactic nuclei (AGNs), the observed huge luminosities of AGNs are powered by gravitational accretion onto a supermassive black hole (e.g., Rees 1984; Blandford 1990; Antonucci 1993; Peterson 1997). The central engines are considered to be surrounded by dusty tori whose typical inner radii are on the order of $\sim$ 1 pc (e.g., Antonucci, Miller 1985; Pier, Krolik 1992, 1993). Therefore, in order to understand AGNs, it is very important to investigate the spatial structures of the inner $\sim$ 1 pc regions in which a supermassive black hole, an accretion disk, warm absorbers, and broad emission-line regions (BLRs) reside. The innermost constituent is the accretion disk; its typical radius is $\sim$ (10 --- 100) $R_{\rm S} \sim (10^{-4}$ --- $10^{-3})M_{8}$ pc where $R_{\rm S}$ is the Schwarzschild radius of a black hole and $M_{8}$ is the black-hole mass in units of $10^{8} M_{\sun}$. The existence of accretion disks in AGNs has been demonstrated by the recent X-ray spectroscopy, collected using {\it ASCA} (Tanaka et al. 1994). The {\it ASCA} X-ray spectra of type-1 AGNs show the presence of very broad Fe K$\alpha$ emission, whose line profile can be fitted well by some accretion-disk models (Tanaka et al. 1995; Fabian et al. 1995; Mushotzky et al. 1995; Iwasawa et al. 1996; Nandra et al. 1997; Reynolds 1997). An ionized accretion disk has also been detected by the recent radio continuum mapping in the archetypical, nearby Seyfert galaxy NGC 1068 (Gallimore et al. 1997). Another inner constituent is broad-line regions (BLRs), because their typical radii are of on the order of 0.01 pc (e.g., Peterson 1993). One of the most important questions related to the BLRs is how emission-line clouds are distributed around AGNs. Although there is still no definite consensus concerning the dynamical and spatial structure of BLRs, there are three alternative models: 1) the disk model (Shields 1979; see also Osterbrock 1989), 2) the high-velocity streamer model (Zheng et al. 1991), and 3) a pair of conical regions in which photoionized clouds are orbiting randomly with Keplerian motion (Wanders et al. 1995; Wanders, Peterson 1996, 1997; Goad, Wanders 1996). In particular, a recent detailed analysis of the reverberation mapping has strongly suggested that the BLRs of the type-1 Seyfert galaxy NGC 5548 has the third type of geometry (Wanders et al. 1995). It is, however, still not known which model is the most popular one. Since recent observations have shown that the BLRs are dominated by rotational motion, rather than the radial motion (e.g., Peterson 1993; Wanders et al. 1995), it is interesting to examine whether or not the rotational axis of the BLR is nearly the same as that of the accretion disk. If the disk model for BLRs would be correct, the BLRs may be coplanar with respect to the accretion disk. In fact, double-peaked BLRs have sometimes been considered to arise from accretion disks, themselves [e.g., P\'erez et al. (1988); Livio, Xu (1997) and references therein; see also, however, Gaskell (1996)]. Motivated by this, we investigate the relationship between the inclination angle of the accretion disk and the width of BLR statistically using published data. | The most important result in this study is that there is no obvious {\it positive} correlation between $\sin i_{\rm AD}$ with FWZI(H$\alpha$)/ $L^{1/4}_{\rm X}$. This suggests that {\it the BLRs are not coplanar with respect to accretian disks}. Some Seyfert nuclei in our sample show double-peaked BLRs (DBLRs). It has sometimes been considered that such DBLRs may arise from an accretion disk, itself (e.g., P\'erez et al. 1988). The DBLR emission profiles of the four Seyfert nuclei in our sample (NGC 3227, NGC 3783, NGC 5548, and 3C 120) were studied by Rokaki et al. (1992) using a standard geometrically-thin accretion- disk model; also, the inclination angles of the DBLRs ($i_{\rm BLR}$) were derived. We compare these inclination angles with those of the accretion disks in figure 2. This comparison also suggests a negative correlation between $i_{\rm AD}$ and $i_{\rm BLR}$, being consistent with our result. This strengthens our suggestion that the BLRs are not coplanar with respect to the accretion disks in the Seyfert nuclei studied here. Let us consider what kind of geometrical configuration can explain the non-coplanar property. The negative correlation means that the normalized velocity width increases with decreasing inclination angle; i.e., {\it Seyfert nuclei with a more face-on accretion disk tend to have larger BLR velocity widths}. There may be three alternative ideas to explain this property. One is the bipolar streamer model (e.g., Zheng et al. 1990). If we observe the accretion disk from a face-on view, the velocity width would be widest because the bipolar wind flows along our line of sight. However, this model has an intrinsic difficulty, as claimed by Livio and Xu (1996), because the emitting region on the receding flow (jet) is obscured from view by the accretion disk; the standard, optically thick accretion disk is opaque up to $\sim$ 1 pc, and, thus, the BLR component behind the disk cannot be seen, because the typical radial distance of BLRs from the central engine is on the order of 0.01 pc (e.g., Peterson 1993). The second idea is that BLRs are located in nearly the same plane as that of an accretion disk, but are orbiting with poloar orbits. If a two-sided jet is ejected with a highly inclined angle with respect to the {\it global} accretion disk, we can explain the negative correlation. Such a jet model is briefly described by Norman and Miley (1984). This idea is consistent with the recent reverberation mapping result for the BLRs of NGC 5548 because the most likely geometry of the BLRs of this galaxy is a pair of conical regions in which photoionized clouds are orbiting randomly with Keplerian motion (Wanders et al. 1995; Wanders, Peterson 1996, 1997; see also Goad, Wanders 1996). This model may also have the same obscuration problem as that for the above streamer model. However, if the BLR clouds are moving at randomly oriented orbits (Wanders et al. 1995), there may be no obscuration problem. The third idea is that BLRs arise from outer parts of a warped accretion disk. The disk model for BLR is the standard idea (Shields 1977; see also for a review Osterbrock 1989). It has been recently shown that accreting gas clouds probed by water-vapor maser emission at 22 GHz show evidence of significant warping (Miyoshi et al. 1995; Begelman, Bland-Hawthorn 1997). The warping of accretion disks can be driven by the effect of the radiation-pressure force (Pringle 1996, 1997). For typical AGN, the warping may occur at {\it r} $>$ 0.01 pc (Pringle 1997), which is at a similar distance as BLRs. Therefore, the warped-disk model can explain the observed negative correlation reasonably well. This model is schematically shown in figure 3. Since the degree of warping and the viewing angle are different from AGN to AGN, the negative correlation between $i_{\rm AD}$ and $i_{\rm BLR}$ may be blurred as obtained in our analysis, although the poor correlation may be also due to the large errors in the estimate of $i_{\rm AD}$. | 98 | 3 | astro-ph9803078_arXiv.txt |
9803 | astro-ph9803308_arXiv.txt | We present measurements of the oxygen abundances in 64 H{\sc ii} regions in 12 LSB galaxies. We find that oxygen abundances are low. No regions with solar abundance have been found, and most have oxygen abundances $\sim 0.5$ to 0.1 solar. The oxygen abundance appears to be constant as a function of radius, supporting the picture of quiescently and sporadically evolving LSB galaxies. | Low surface brightness (LSB) disk galaxies have all the characteristics of unevolved galaxies. Those discovered so far constitute a population of gas-rich, metal-poor galaxies with very low star formation rates (see the review by Bothun et al. 1997). Their surface brightnesses are a few magnitudes lower than the values commonly found for so-called normal galaxies. Most of them are rather late-type galaxies, with diffuse spiral arms. A direct probe of the evolutionary state of these galaxies is the metal abundance in the interstellar medium (ISM). A low abundance generally indicates only limited enrichment of the ISM and therefore (in a closed system) a small amount of evolution. Because of their low surface brightness, obtaining spectra of the stellar disks of LSB galaxies is difficult and requires large amounts of telescope time. Conclusions on metallicities must therefore be derived from spectra of H{\sc ii} regions. These usually are the brightest distinct objects in a LSB galaxy. Their bright emission lines make them more easily observable than the underlying continuum. The H{\sc ii} regions that are observed in LSB galaxies are usually giant H{\sc ii} regions, that are ionized by star clusters rather than by a few stars. The first measurements of the oxygen abundances in H{\sc ii} regions in LSB galaxies were presented in McGaugh (1994). He found, using an empirical oxygen abundance indicator, that LSB galaxies are low metallicity galaxies with typical values for the metallicity $Z<\frac{1}{3} Z_{\odot}$. It shows that low metallicities can occur in galaxies that are comparable in size and mass to the bright galaxies that define the Hubble sequence. As LSB galaxies are found to be isolated (Mo et al.\ 1994), this suggest that surface mass density and environment are as important for the evolution of a galaxy as total mass. In this paper we present a follow-up study of oxygen abundances in LSB galaxies. We confirm the results by McGaugh (1994) that LSB galaxies are metal-poor. We present two direct measurements of the oxygen abundance from measurements of the [O {\sc iii}]$\lambda 4363$ line, supplemented with a large number of empirically determined oxygen abundances. In addition, for those galaxies where sufficient data are available, we investigate the change in abundance with radius, and show that the measurements are consistent with no gradient. The steeper gradients found in HSB galaxies (Vila-Costas \& Edmunds 1992 [VCE], Zaritsky et al. 1994, Henry \& Howard 1995, Kennicutt \& Garnett 1996) are not present. It is worth noting that the exact form and magnitude of the Milky Way oxygen gradient has now been consistently reproduced in early-type stars, H{\sc ii} regions and planetary nebulae, which supports that the extra-galactic H{\sc ii} region gradients in HSB galaxies are real (Smartt \& Rolleston 1997). The lack of abundance gradients in LSB galaxies supports the picture of stochastic and sporadic evolution, where the evolutionary rate only depends on local conditions and not on the global properties of LSB galaxies as a whole. Section 2 describes the sample selection and observations. Section 3 presents the data, while in Sect.~4 the analysis is described. Section 5 discusses the abundances found. Section 6 discusses reddening towards the H{\sc ii} regions, while Sect.~7 concludes with presenting the gradients. In Sect.~8 the results are summarized. | The oxygen abundances in 64 H{\sc ii} regions in 12 LSB galaxies have been measured. Oxygen abundances are low. No region with solar abundance has been found, and most have oxygen abundances that are $\sim 0.5$ to 0.1 solar. No strong radial oxygen abundance gradients are found. The abundance seems to be constant, rather, as a function of radius, supporting the picture of quiescently and sporadically evolving LSB galaxies. | 98 | 3 | astro-ph9803308_arXiv.txt |
9803 | hep-ph9803471_arXiv.txt | \noindent From a phenomenological point of view, we study active-active and active-sterile flavour-changing (and flavour-conserving) oscillations of Dirac-Majorana neutrinos both in vacuum and in matter. The general expressions for the transition probabilities in vacuum are reported. We then investigate some interesting consequences following from particular simple forms of the neutrino mass matrices, and for the envisaged scenarios we discuss in detail neutrino propagation in matter. Special emphasis is given to the problem of occurrence of resonant enhancement of active-active and active-sterile neutrino oscillations in a medium. The peculiar novel features related to the Dirac-Majorana nature of neutrinos are particularly pointed out. | Today we have several indications in favour of non zero neutrino masses and mixing.\\ The solar neutrino problem, i.e. the observed deficit of solar neutrino fluxes \cite{SNP}, is a well established tool whose resolution requires (almost without doubt \cite{doubt}) neutrino physics beyond the (minimal) Standard Model. The acceptable solutions to this problem, in terms of vacuum \cite{Vsol} or matter \cite{Msol} flavour oscillations or spin and flavour oscillations \cite{SFsol} as well as in terms of active-sterile neutrino conversion \cite{Stsol}, all need non vanishing neutrino masses and mixing \cite{Vac, MSW, Akh, sterile}.\\ The second indication in favour of neutrino oscillations come from the observed deficit of atmospheric muon neutrinos with respect to electron neutrinos \cite{atmo} that can be explained in terms of \nm \rt \nt or \nm \rt \ne or even active-sterile neutrino transitions \cite{atsol}.\\ Laboratory direct searches for massive neutrinos only give, at present, upper limits on neutrino masses \cite{masses} and the same is valid for reactor and accelerator neutrino oscillations experiments \cite{reactor}, except for the LSND experiment \cite{LSND} whose results seem to be explained in terms of $\ov{\nm} \rt \ov{\ne}$ oscillations.\\ Hints for massive neutrinos also come from cosmology, looking at \nt as the most probable candidate for the hot component of the dark matter \cite{HDM} and from the observed peculiar velocities of pulsars \cite{Segre}. On the other hand, in Grand Unified Theories, which attempt to give a unified view of electroweak and strong interactions, massive neutrinos are predicted \cite{Buccella} together with other phenomena violating both lepton numbers and baryon number (such as, for example, proton decay). However, the most intriguing fact is that a simple scenario with only three massive neutrinos cannot account for the solar neutrino problem, the atmospheric neutrino anomaly and the LSND result. This is because the three squared masses differences $\Delta m^2$ for the three oscillation solutions to these problems are all distinct between them: the resonant MSW solution to the solar neutrino problem requires $\Delta m^2 \, \sim \, 10^{-5}$ eV$^2$, while for the atmospheric anomaly $\Delta m^2 \, \sim \, 10^{-2}$ eV$^2$ is needed and $\Delta m^2 \, \sim \, 1$ eV$^2$ for the LSND result. Many analyses \cite{analyses} have been conducted for giving a unified view of the three problems in terms of neutrino oscillation (taking into account also the limits from laboratory experiments) and a coherent picture seems to emerge with four massive neutrinos, namely the three known neutrinos plus a sterile neutrino. Note that four neutrino mass eigenstates are needed, but not necessarily more than three neutrino flavour eigenstates. This scenario is easily realized if one considers neutrinos as Dirac-Majorana particles described by the following general mass term in the electroweak lagrangian \cite{sterile}: \be -{\cal L}^{DM}_{m} \; = \; \sum_{l,l^{\prime}} \ov{\nu}_{l^{\prime}R} \; M^{D}_{l^{\prime}l} \; \nu_{l^{\prime}L} \; + \; \frac{1}{2} \; \sum_{l,l^{\prime}} \ov{\nu}^{c}_{l^{\prime}R} \; M^{1}_{l^{\prime}l} \; \nu_{l L} \; + \; \frac{1}{2} \; \sum_{l,l^{\prime}} \ov{\nu}^{c}_{l^{\prime}L} \; M^{2}_{l^{\prime}l} \; \nu_{l R} \; + \; h.c. \label{11} \ee Here $l,l^{\prime} = e, \mu , \tau$ label the three flavour eigenstates and $M_D$, $M_1$, $M_2$ are the Dirac and the two Majorana mass matrices which, in general, are hermitian and non diagonal (however, $M_1$ and $M_2$ have to be symmetric). To construct the mass term in (\ref{11}) we need the three known left-handed neutrinos (and their antiparticles) and other three right-handed sterile neutrinos (and their antiparticles) \footnote{Obviously, the generalization to more than three families is possible and straightforward}. After the diagonalization of (\ref{11}) we can obtain in general six mass eigenstates which are Majorana fields; so in this framework we can easily endow the above scenario with four massive neutrinos coming from the experiments.\\ Note that if neutrinos are really described by the mass term in (\ref{11}), the total lepton number is no longer conserved and peculiar phenomena, as neutrinoless double beta decay and neutrino-antineutrino oscillations can take place.\\ We stress that (\ref{11}) is predicted in many GUTs \cite{Buccella} in which the popular ``seesaw'' mechanism \cite{seesaw} can give rise to very small neutrino masses in a very natural way by supposing $M_1 \approx 0$ and $M_D \ll M_2$. However, this is not the only possibility; recently some models assuming $M_1 \simeq M_2$ have been proposed \cite{equal} for accounting the three experimental indications on neutrino oscillations discussed above. Here we further explore this last scenario and study flavour transitions of Dirac-Majorana neutrinos from a completely phenomenological point of view, adopting no particular model. This work is a generalization to flavour transitions of previous papers \cite{TE, previous} in which we studied peculiar oscillations of Dirac-Majorana neutrinos. We now assume, for simplicity, only two flavours, so $M_D$, $M_1$, $M_2$ in (\ref{11}) are $2 \times 2$ matrices in the flavour space. In the following section, the basic vacuum oscillations allowed by (\ref{11}) are studied and transition probabilities are explicitly given in the general case. Some very interesting consequences due to particularly simple forms of the mass matrix are also investigated. In section 3, given the effective hamiltonian of Dirac-Majorana neutrinos interacting with a medium, resonant matter oscillations are considered along with a qualitative discussion of the phenomenon with the aid of the level crossing diagram. Finally, in section 4, there are our conclusions and remarks. | In this paper we have studied the propagation both in vacuum and in matter of Dirac-Majorana neutrinos and analyzed active-active (flavour-changing) as well as active-sterile transitions, which are, in general, both possible.\\ For vacuum oscillations, in section 2 we have given the general expressions for the transitions probabilities for $ \nel \rightarrow \nml $, \nel \rt \ncml , $ \nel \rightarrow \ncel $ We have then discussed some interesting limiting cases for Dirac ($M_D$) and Majorana ($M_M$) mass matrices. For pure Dirac ($M_M=0$) or pure Majorana ($M_D = 0$) neutrinos obviously we recover the usual flavour oscillation formulae \cite{Vac}, while for both $M_D$ and $M_M$ non vanishing and diagonal the Pontecorvo formula \cite{Pontecorvo} for neutrino-antineutrino (active-sterile) oscillations is obtained \cite{TE}.\\ An interesting non trivial case is that with $M_D$ and $M_M$ given by (\ref{226}) or (\ref{235}) which implement the idea that neutrino mixing is essentially ruled only by the Dirac mass term while the Majorana mass term is diagonal or vice-versa, respectively. In both cases, neither pure flavour oscillations nor Pontecorvo oscillations are predicted, but only flavour-changing active-sterile transitions, such as $\nel \rightarrow \ncml$, are possible. Remarkably, the transition probability for these is identical in form to that for flavour oscillations for pure Dirac or Majorana neutrinos, and this holds both in vacuum and in matter. For the latter, the resonance condition is only shifted by the neutral current contribution of \nel to the effective potential. So, for example, the solution to the solar neutrino problem in terms of active-sterile neutrino oscillations proposed in \cite{Stsol} applies unmodified to the present scheme. \\ Another interesting case, even if a bit more complicate, has been analysed for the mass matrices in (\ref{236}) or (\ref{237}), which implements the idea that neutrino mixing is given by the Dirac and Majorana mass terms with the same strength. In this case, $ \nel \rightarrow \nml $, $ \nel \rightarrow \ncml $, $ \nel \rightarrow \ncel $ transitions are all possible and, in vacuum, the first two have the same transition probability, which is also identical in form to that obtained in the previous case, except for a constant suppression factor in the amplitude of oscillations of 1/4. Also in matter the pattern of neutrino transitions present in the general case is (qualitatively) reproduced in this peculiar scheme. In particular, all the flavour changing oscillations can be resonantly amplified while Pontecorvo \nel \rt \ncel matter oscillations have maximum amplitude only for a given electron to neutron number ratio (see eq. (\ref{313}) and the related footnote); the resonance conditions were discussed in section 3.1 . \\ Given the multiresonance structure of the oscillations pattern, it is then interesting to follow the evolution of a \nel , for example, in a varying density medium such as the Sun; this has been done in section 3.4 with the help of a level crossing diagram reported in Fig. 1. Several scenarios are possible according to the adiabaticity properties of level crossing near the resonance points. In particular, starting from a pure \nel beam at high density, to have a consistent conversion into \nml at low density the resonance for \nel \rt \ncml has to be crossed non adiabatically, while the passage through the one for \nel \rt \nml has to be adiabatic. However, we have also shown that at very low density, and then in the vacuum, it is more appropriate to deal with the Majorana combinations $\wt{n_\pm}$ in (\ref{212}) than with the pure flavour states \nel , \ncel , \nml , \ncml . This is strictly related to the Dirac-Majorana nature of neutrinos, which chooses Majorana eigenstates instead of pure flavour states as starting points. In this respect, we have to deal with ``generic'' flavour-changing or flavour-conserving transitions of Dirac-Majorana neutrinos without looking at the particular active neutrino or sterile antineutrino state. It is through the weak interactions, with which neutrinos are produced and detected, that a particular (active or sterile) component of the Majorana eigenstates is chosen.\\ The results obtained for the case of degenerate Dirac-Majorana mixing are qualitatively valid also in the general case in which all the entries of the Dirac and Majorana mass matrices are non zero and different between them. The main difference between the two cases is that in the general framework there are 3 mass parameters and two mixing angles ruling the evolution, while for the particular case studied in sections 2.3 and 3.4 there are only 2 mass parameters and 1 mixing angle (these parameters being not completely independent, because of relation (\ref{239b})). The presence of more degrees of freedom in the general case allows to consider some peculiar situations which are not possible otherwise. The most remarkable one is that in the general case the proportionality of the $\Sigma$ parameter to $\sin^2 \, 2 \theta_+$ (see eq. (\ref{239b})) is lost, so that the structure of the level crossing diagram at very low density can be altered. The eigenvalues of $H_m$ in (\ref{39}) for zero density (vacuum) are given by \be \frac{1}{8k} \, \left( \pm \, 2 \Delta m_+^2 \; + \; \Sigma \right) \ee \be \frac{1}{8k} \, \left( \pm \, 2 \Delta m_-^2 \; - \; \Sigma \right) \ee so that one can manipulate the mass parameters to modify the low density region of the level crossing diagram without grossly altering the region where the resonance points are present. In any case, there can be present no substantial modifications of the conclusions reached above. The oscillations of Dirac-Majorana neutrinos here studied with their peculiar features can be efficiently tested in astrophysics, in particular detecting solar or supernova neutrinos, and can have even profound implications in cosmology for the nucleosynthesis of light elements in the Universe. \vspace{1truecm} \noindent {\Large \bf Acknowledgements}\\ \noindent We express our sincere thanks to Prof. F. Buccella for very useful talks and his unfailing encouragement, and to Prof. E. Kh. Akhmedov for enlightening discussions with one of us (S.E.). | 98 | 3 | hep-ph9803471_arXiv.txt |
9803 | astro-ph9803234_arXiv.txt | A new cataclysmic variable is identified as the optical counterpart of the faint and hard X-ray source RX\,J0757.0+6306 discovered during the ROSAT all-sky survey. Strong double-peaked emission lines bear evidence of an accretion disc via an S--wave which varies with a period of 81$\pm 5$~min. We identify this period as the orbital period of the binary system. CCD photometry reveals an additional period of 8.52$\pm 0.15$ min. which was stable over four nights. We suggest that \rxj\ is possibly an intermediate polar, but we cannot exclude the possibility that it is a member of the SU UMa group of dwarf novae. | Within a project for the optical identification of a complete sample of 674 northern ROSAT All-Sky Survey (RASS) X-ray sources (which is a collaboration between the Max-Planck-Institut f\"ur extraterrestrische Physik, Garching, the Landessternwarte Heidelberg, Germany, and the Instituto Nacional de Astrof\'isica, Optica y Electronica, Mexico) several new cataclysmic variables were identified. A detailed description of the project is given by Zickgraf \etal\ (1997). The full catalogue with all identifications is published in Appenzeller \etal\ (1998). Here we report the identification of the RASS X-ray source RX\,J0757.0+6306 (= 1RXS J075700.5+630602). Cataclysmic variables (CVs) are close binary systems with a white dwarf primary accreting matter supplied by a late type main-sequence secondary star via an accretion disc or along magnetic field lines of the white dwarf. Magnetic CVs, where the white dwarf has a sufficiently strong magnetic field to affect the accretion trajectory, form two distinctive subclasses: the high-field polars, and the low-field intermediate polars (IPs). These subclasses are characterized by well-defined observational properties (Cropper 1990; Patterson 1994; Warner 1995). The polars are usually soft X-ray emitters and have near synchronously rotating WD, the IPs are harder X-ray sources and show a second periodicity due to the asynchronously rotating WD. In some cases a third period is observable, which is interpreted as the beat period between orbital and spin periods. Besides differences in the flux distribution and variability, the orbital period distribution of the various subclasses of CVs were also noticed to be different (Kolb 1995). Polars tend to cluster below the period gap (2 h $< P_{orb}<$ 3 h), while IPs are preferentially above the gap. Non-magnetic CVs are distributed almost equally. All subclasses however show deficiency of systems in the period gap and a short-period cutoff at $\rm P_{min}=80$ min (the minimum period). The statistically significant properties of the period distribution are presumed to have an evolutionary origin (Verbunt \& Zwaan 1981, Verbunt 1984, King 1988, Kolb and Ritter 1992). The rapidly increasing number of magnetic CVs discovered from the ROSAT data has a significant impact on the above mentioned distribution and its consequences. | A new cataclysmic variable is discovered with interesting features: \begin{enumerate} \item The orbital period of $81\pm 5$ min puts \rxj\ near the hydrogen burning period minimum where CVs experience a turning point of their evolution. Large flickering in the optical light curve and the observed spectral features of the object certainly show the presence of an accretion disc. \item The limited search in the Sonneberg all-sky patrol plates revealed that the system undergoes outburst activity. Another outburst was recorded (vsnet-alert No. 1379) shortly after the object's discovery was announced through the VSNET (vsnet-chat No 662). From the plate statistics we can assume that the system has rather frequent outbursts. The amplitude of the outbursts of about 4 mag are typical for dwarf novae systems, but not as large as in SU~UMa superoutbursts or the so called TOADs (tremendous outburst amplitude dwarf novae; Howell \etal\ 1995). \item There are periodic light variations with a period of 8.5 min in the light curve of the \rxj. We observed them directly on four out of five occasions. In the fifth night a periodic signal with a side-band frequency was detected in the power spectrum. Very recently, the 8.5\,min period was confirmed by R. Fried (vsnet-alert No 1387) from more prolonged observations. \end{enumerate} Thus, \rxj\ shows mixed characteristics, making its type classification uncertain. From purely spectroscopic characteristics one may conclude that the new CV is a dwarf nova. Its short orbital period suggests that instead it may belong to the SU~UMa class or TOADs. But the repetitive detection of high-frequency pulses with a clearly fixed period indicates that it deserves a classification as an intermediate polar. This still needs to be confirmed by checking the coherency of the photometric pulses and by the detection of X-ray pulses. Intermediate polars are CVs with the primary white dwarf rotating asynchronously due to its moderate magnetic field. Column accretion onto the magnetic poles results in the emission of high-energy radiation. This radiation is reprocessed elsewhere in the system into optical light which is modulated at periods shorter than the orbital period. The optical modulation can track the spin and/or the spin/orbit beat period of the binary (see the review by Patterson 1994). The presence of X-ray emission in the quiescent state of \rxj\ along with the moderate He~{\sc ii} 4686 \AA\ emission also argue in favor of a magnetic nature. The survey of non-magnetic CVs by van Teeseling \etal\ (1996) shows that the majority of X-ray emitting dwarf novae are of the SU UMa type, but they all are softer sources (HR1$\le 0.7$) than \rxj. There are a few long-period objects classified as non-SU UMa variables that are as hard as \rxj. These belong to the VY~Scl, Z~Cam, and UX~UMa subclasses. We do not have any evidence which support a classification of \rxj\ as any of these types. Hence, since the rest of the CVs which are X-ray sources are magnetic, we conclude that \rxj\ is most probably magnetic. Only two IPs have been detected with EUVE and only a few dwarf novae, all of the latter during outburst. AM Herculis stars, particularly those with high magnetic fields are detected using EUVE. Thus, the non-detection of \rxj\ with EUVE does not prove that it is not an IP. It may indicate that \rxj\ could have a weak magnetic field (B $< 8$ MG), but given that only two IPs have EUVE detections and because of the rather large distance of \rxj\ the non-detection is not considered unusual. On the other hand, the intensity of the He{\sc~ii} emission is not high enough to unambiguously classify it as a magnetic system. Silber (1992) set the following criteria for magnetic CVs: $20<$EW~(H$\beta)<40\AA$ and He{\sc ii}/H$\beta>0.4$. In our case, if the larger equivalent width could be attributed to a shorter orbital period, the He{\sc ii}/H$\beta$ ratio is definitely below this criterium ($\approx~0.15$). The existence of outbursts and the short orbital period of the system is in some discordance with the IP classification. Most IPs cluster above the 2--3 hour period gap, while short period magnetic CVs are usually polars. However, a weak field IP will remain an IP even when it evolves towards shorter periods. In IPs, accretion outside of the Alfven radius remains in the form of a disc, while accretion inside the Alfven radius is dominated by flow along magnetic field lines. Outburst activity is uncommon since the inner part of the disc is disrupted by the magnetic field. Nevertheless, neither outburst activity nor short orbital period exclude the possibility of \rxj\ to be classified as an IP. In addition, the presence of a large disc in a short period magnetic CV suggests that the magnetic field is weak. Otherwise it would be a polar. For such a weak field case it may not be surprising that \rxj\ appears to be an IP with some properties (i.e. outbursts) similar to non-magnetic CVs, yet the evidence that it is a magnetic CV is compelling. Therefore, we offer \rxj\ as a candidate for the shortest period intermediate polar. | 98 | 3 | astro-ph9803234_arXiv.txt |
9803 | astro-ph9803002_arXiv.txt | We show that pulsar velocities may arise from anisotropic neutrino emission induced by resonant conversions of massless neutrinos in the presence of a strong magnetic field. The main ingredient is a small violation of weak universality and neither neutrino masses nor magnetic moments are required. | 98 | 3 | astro-ph9803002_arXiv.txt |
||
9803 | astro-ph9803016_arXiv.txt | The $^9$Be\,{\sc ii} $\lambda$ 3131 \AA\ doublet has been observed in the solar-type stars 16 Cyg A \& B and in the late G-type star $\rho^1$ Cnc, to derive their beryllium abundances. 16 Cyg A \& B show similar (solar) beryllium abundances while 16 Cyg B, which has been proposed to have a planetary companion of $\sim 2$ $M_{\rm Jup}$, is known to be depleted in lithium by a factor larger than 6 with respect to 16 Cyg A. Differences in their rotational histories which could induce different rates of internal mixing of material, and the ingestion of a similar planet by 16 Cyg A are discussed as potential explanations. The existence of two other solar-type stars which are candidates to harbour planetary-mass companions and which show lithium and beryllium abundances close to those of 16 Cyg A, requires a more detailed inspection of the peculiarities of the 16 Cyg system. For $\rho^1$ Cnc, which is the coolest known object candidate to harbour a planetary-mass companion ($M > 0.85$ $M_{\rm Jup}$), we establish a precise upper limit for its beryllium abundance, showing a strong Be depletion which constrains the available mixing mechanisms. Observations of similar stars without companions are required to asses the potential effects of the planetary companion on the observed depletion. It has been recently claimed that $\rho^1$ Cnc appears to be a subgiant. If this were the case, the observed strong Li and Be depletions could be explained by a dilution process taking place during its post-main sequence evolution. | \label{sec1} In very recent years, several stars have been proposed to have planetary companions on the basis of measured precise radial velocity variations. This field of research is experiencing rapid development, and updated reviews of the present situation can be found in the proceedings of the workshop on {\it Brown Dwarfs and Extrasolar Planets} edited by Rebolo et al. (1998) and in The Extrasolar Planets Encyclopaedia\footnote{http://wwwusr.obspm.fr/planets/} by J. Schneider. Once a solar-type star has been suggested to harbour a planetary-mass companion, it is interesting to investigate any similarities with the Sun, as well as to find possible differences with respect to other single stars. Chemical abundances are among the most important parameters to be compared and, in particular, precise abundances of light elements such as lithium and beryllium (easy to destroy by $(p,\alpha)$ nuclear reactions when the temperature reaches $\sim 2.5\times 10^6$ and $\sim 3.5\times 10^6$ K, respectively) combined with the abundances of other elements which are not so readily destroyed in stellar interiors, should help to understand how the presence of planets may affect the chemical composition of their parent stars. Gonzalez (1997, 1998) has derived the overall metallicities as well as abundances of different elements (including lithium) for a wide sample of proposed parent stars, finding that four of the known systems show a metallicity significantly higher than the solar value. A peculiar system such as 16 Cyg A \& B, formed by twin solar-type stars of which only one has an orbiting planet (Cochran et al. 1997), is an especially suitable candidate to perform a detailed abundance study. Gonzalez (1998) found that both stars have a similar metallicity with a value slightly larger than solar, and confirmed independently a previous result of King et al. (1997a) that 16 Cyg B (the star with a suspected planet) is strongly depleted in lithium with respect to 16 Cyg A. The knowledge of their beryllium abundances is of potential value in quantifying the possible influence of a planetary companion on the mixing mechanisms operating in the stellar interior. $\rho ^1$ Cnc is a star with spectral type G8V, and is the coolest known object which is a candidate to have a planetary companion. Following Gonzalez (1998), this star falls into the group having roughly Jupiter-mass companions with small circular orbits and very metal-rich parent stars. Dominik et al. (1998) have shown recently that the planetary system of $\rho ^1$ Cnc also hosts a Vega-like disk of dust, evidenced by an infrared excess at 60 $\mu$m. The star is very depleted in lithium and its beryllium abundance could be compared with existing upper limits measured in younger stars with similar effective temperatures (Garc\'\i a L\'opez et al. 1995a). In this paper we derive the beryllium abundances of the 16 Cyg system and of $\rho ^1$ Cnc by comparing observations with spectral syntheses of the $^9$Be\,{\sc ii} $\lambda$ 3131 \AA\ doublet. We use those, together with their published lithium values, as well as with available abundances for other stars (with and without suggested planetary companions), and discuss briefly possible effects of planets on processes taking place in their structure and evolution. | \label{sec5} Beryllium abundances have been derived for the solar-like stars 16 Cyg A \& B and the cooler object $\rho ^1$ Cnc, for which there are published values of their lithium abundances. 16 Cyg B and $\rho ^1$ Cnc are candidates to be parents of extrasolar planets, and by measuring their Be abundances we aim at studying the potential dependence on the presence of planetary companions of detailed processes operating in their structure and evolution. 16 Cyg A \& B show very similar Be abundances, which are compatible with the solar value, while the lithium abundance of 16 Cyg B is at least a factor 6 smaller than that of 16 Cyg A. Different rates of mixing of material in their interiors associated with different angular momentum histories, as well as the hypothetical ingestion of a planetary companion by 16 Cyg A are discussed as potential explanations. The existence of two other solar-like parent stars, whose Li (and Be) does not show strong depletion, i.e. whose behaviour is like 16 Cyg A, the Sun and the majority of similar stars with Li and Be abundances available, implies that the 16 Cyg system requires special observational and theoretical attention. A low upper limit has been derived for the beryllium abundance of $\rho ^1$ Cnc. This is the first time a precise limit has been set and that such strong Be depletion has been observed in a late G-/early K-type MS star. This measurement clearly constrains the depletion predictions of the available mixing mechanisms, but requires observation of planet-free stars with similar age and spectral type to discard the potential effects of the planetary companion on the Li and Be depletions. Claims have also been made indicating that $\rho ^1$ Cnc appears to be a subgiant. If this were the case, its strong Li and Be depletions could be explained by a dilution process taking place during its post-MS evolution. | 98 | 3 | astro-ph9803016_arXiv.txt |
9803 | astro-ph9803199_arXiv.txt | We present a study of the polarizing power of the dust in cold dense regions (dark clouds) compared to that of dust in the general interstellar medium (ISM). Our study uses new polarimetric, optical, and spectral classification data for 36 stars to carefully study the relation between polarization percentage ($p$) and extinction ($A_V$) in the Taurus dark cloud complex. We find two trends in our $p-A_V$ study: (1) stars background to the warm ISM show an increase in $p$ with $A_V$; and (2) the percentage of polarization of stars background to cold dark clouds does not increase with extinction. {\it We detect a break in the $p-A_V$ relation at an extinction $1.3 \pm 0.2$ mag, which we expect corresponds to a set of conditions where the polarizing power of the dust associated with the Taurus dark clouds drops precipitously. This breakpoint places important restrictions on the use of polarimetry in studying interstellar magnetic fields.} | The polarization of background starlight has been used for nearly half a century to probe the magnetic field direction in the interstellar medium (ISM). The observed polarization is believed to be caused by dichroic extinction of background starlight passing through concentrations of aligned elongated dust grains along the line-of-sight. Although there is no general consensus on which is the dominant grain alignment mechanism (Lazarian, Goodman \& Myers 1997), it is generally believed that the shortest axis of the ``typical'' elongated grain tends to be become aligned to the local magnetic field. For this orientation, the observed polarization vector is parallel to the plane-of-the-sky projection of a line-of-sight-averaged magnetic field (Davis \& Greenstein 1951). The line-of-sight averaging inherent in background starlight polarimetric observations can make interpretation of the polarization produced by different field orientations and/or several independent dust clouds very complicated. Nonetheless, it was thought that if lines of sight with just one localized very dusty region (such as a dark cloud) between us and a background star could be found, surely the polarization would reveal the field associated with that dusty region. However, recent studies in the Taurus region (Goodman et al. 1992; Gerakines et al. 1995) and other parts of the sky (Creese et al. 1995; Goodman et al. 1995) have uncovered substantial evidence to show that dust inside cold dark clouds has lower polarizing power than dust in the general warm ISM. This means that the polarization of the light from a background star is a non-uniformly {\it weighted} line-of-sight average of the projected plane-of-the-sky field, and that grains in cold dark clouds are systematically down-weighted. The ultimate implication of this down-weighting is that above some (column?) density threshold, the polarization of background starlight gives no information about the magnetic field in dark clouds. It is the goal of this Letter to find and physically describe this threshold. The evidence that grains in cold dark clouds are inefficient polarizers of background starlight is multi-faceted. Eight years ago, Goodman et al. (1990) found that the smooth large-scale patterns apparent in polarization maps of dark cloud complexes (e.g. Vrba et al. 1976; Vrba et al. 1981; Moneti et al. 1984; Whittet et al. 1994) do not systematically change in response to the large density enhancements represented by the dark clouds. After this realization, it was hypothesized that perhaps optical polarimetry was incapable of seeing field changes which might occur only in the high-density, optically opaque, interiors of dark clouds. So, near-infrared polarimetry, which can probe the optically opaque cloud interiors was undertaken. The near-infrared observations showed that the mean direction and dispersion of the polarization vectors are virtually {\it identical} in the cloud interiors and their peripheries (Goodman et al. 1992; 1995). Thus, the presence of cold dark clouds still appeared to have no geometric effect on the polarization maps, implying either that: 1) the field is truly unaffected by the cloud; or 2) that background starlight polarimetry is somehow insensitive to the field in dark clouds. Polarization-extinction relations provide the best discriminant between these hypotheses. For grains of constant polarization efficiency, $p$ should rise with $A_V$. Using the near-infrared observations, Goodman et al. (1992, 1995) find that the {\it percentage of polarization does not rise with extinction} in cold dark clouds. The simplest interpretation\footnote{Note that increased field tangling inside dark clouds cannot explain the near-infrared polarimetric observations for two reasons. 1.) The dispersion in the distribution of position angle does not increase in the cloud interior (near-IR observations) relative to the periphery (optical observations). And, 2.) while it is true that the slope of a $p-A_V$ relation will diminish due to field tangling, it will remain positive even for highly tangled fields if all grains polarize equally well (see Jones 1989; Jones et al. 1992).} of this result is that dust in dark clouds adds plenty to the observed extinction, but has very little ``polarizing power" and so adds only a very small fraction to the observed net polarization. A number of factors, including poor grain alignment, grain growth, and/or changes in grain shape or composition, could be responsible for the low polarization efficiency exhibited by dust grains in cold dark clouds (see Goodman et al. 1995). Regardless of which factor(s) cause(s) the low polarization efficiency exhibited by by dust in dark clouds, the fact is that background starlight polarimetry does not reliably reveal the magnetic field {\it in} dark clouds. Based on the near-infrared studies, we expect that the fraction of grains with high polarization efficiency is relatively constant in the lower-density warm ISM, but drops precipitously in dark clouds. Therefore, we hypothesize that a breakpoint in the $p-A_V$ relationship might exist at the dark clouds' ``edges", which previous studies in the near-IR (Goodman et al. 1992; 1995; Gerakines et al. 1995) could not detect, due to their inability to measure low $A_V$'s accurately enough. In this Letter, we present our attempt to carefully study the $p-A_{V}$ relation near dark clouds, and thus offer a set of guidelines as to where the polarization maps can be taken as faithful representations of the magnetic field projected onto the plane of the sky, and where they cannot. | The breakpoint in the $p-A_{V}$ relation places important restrictions on the use of polarimetry in studying interstellar magnetic fields. Since the polarization efficiency of the dust inside dark clouds is very low, most of the polarization observed for lines of sight which pass through these extinction peaks is not due to the dark cloud; it is due to dust background and foreground to the cloud. Hence, one should not use background starlight polarimetry to map magnetic fields inside dark clouds. With the results of this study we can quantify the word ``inside." In regions like Taurus, {\it it is safe to interpret the polarization of background starlight as a representation of the plane-of-the-sky projected magnetic field up to the 1.3 $\pm$ 0.2 mag ``edge'' of the dark cloud}. In other words the linear relation between $p$ and $A_V$ that exists in the low-density ISM breaks down for stars background to the $\simgreat 1.3$ mag of extinction produced by a dense localized dusty region (i.e., dark cloud). After this edge polarization no longer rises with extinction, and thus cannot reveal the field structure in the dense region. We restate that this proscription only applies for stars background to cold dark clouds, as stars background to the warm ISM have not been shown to exhibit such behavior. | 98 | 3 | astro-ph9803199_arXiv.txt |
9803 | astro-ph9803150_arXiv.txt | We investigate the effect of gravitational lensing by matter distribution in the universe on the cosmic microwave background (CMB) polarization power spectra and temperature-polarization cross-correlation spectrum. As in the case of temperature spectrum gravitational lensing leads to smoothing of narrow features and enhancement of power on the damping tail of the power spectrum. Because acoustic peaks in polarization spectra are narrower than in the temperature spectrum the smoothing effect is significantly larger and can reach up to 10\% for $l<1000$ and even more above that. A qualitatively new feature is the generation of $B$ type polarization even when only $E$ is intrinsically present, such as in the case of pure scalar perturbations. This may be directly observed with Planck and other future small scale polarization experiments. The gravitational lensing effect is incorporated in the new version (2.4) of CMBFAST code. | Over the next few years a number of ground based, balloon and satellite experiments will measure CMB sky with an unprecedented accuracy and detail. The promise of a one percent precision on the measured power spectrum of CMB anisotropies requires a similar accuracy in the theoretical predictions, if we are to exploit all the information present in the data. The rewards will be rich: among other things this will allow an accurate determination of a number of cosmological parameters and testing of current structure formation theories \cite{parameters}. In principle such a program is possible, since the anisotropies were produced when the universe was still in the linear regime, which makes the calculations of model predictions very accurate. In practice there are a number of important effects that need to be included if this goal is to be realized. One of the most important among these is the gravitational lensing effect. As photons propagate through the universe from their last scattering to our detectors they are randomly deflected by the gravitational force exerted upon them by the inhomogeneous mass distribution. Previous work has shown that gravitational lensing has an effect on the temperature anisotropy power spectrum which is not insignificant \cite{others,uroslens}. The random deflections smear out the sharp features in the correlation function or power spectrum, leading to a suppression of acoustic oscillations. Gravitational lensing can also enhance power on the damping tail, causing it to decay less rapidly than predicted on very small angular scales \cite{bentonsilk}. Gravitational lensing effect on the temperature anisotropies has been discussed several times in the literature and the formalism to calculate it using the evolution of density power spectrum both in linear and nonlinear regime has been presented in \cite{uroslens}. In this paper we extend this calculation to the two linear polarization power spectra and to the cross-correlation spectrum between temperature and polarization. Because acoustic oscillations are narrower for polarization spectra than for temperature, one expects gravitational lensing effect to be more significant in the former and indeed our results confirm this. In addition, a qualitatively new effect is the mixing between $E$ and $B$ types of polarization, which changes the pattern of polarization. The outline of the paper is the following. In \S \ref{formalism} we develop the formalism: this section contains all the main analytic expressions needed for a numerical implementation of the effect. These have been numerically implemented in the new version of CMBFAST package (version 2.4) and require only a marginal increase in the CPU time for their evaluations. In \S \ref{estimate} we compute the effect for a typical cosmological model and address the question of direct observability of the effect. We present the conclusions in \S \ref{conclusions}. | 98 | 3 | astro-ph9803150_arXiv.txt |
|
9803 | astro-ph9803293_arXiv.txt | Optical multi-slit spectra have been obtained for 47 globular clusters surrounding the brightest Virgo elliptical NGC~4472 (M49). Including data from the literature, we analyze velocities for a total sample of 57 clusters and present the first tentative evidence for kinematic differences between the red and blue cluster populations which make up the bimodal colour distribution of this galaxy. The redder clusters are more centrally concentrated and have a velocity dispersion of 240 kms$^{-1}$ compared with 320 kms$^{-1}$ for the blue clusters. The origin of this difference appears to be a larger component of systematic rotation in the blue cluster system. The larger rotation in the more extended blue cluster system is indicative of efficient angular momentum transport, as provided by galaxy mergers. Masses estimated from the globular cluster velocities are consistent with the mass distribution estimated from X-ray data, and indicate that the M/L$_B$ rises to $50$~M/L$_\odot$ at 2.5 R$_e$. | The study of extragalactic globular cluster systems can provide important clues to the formation history of their host galaxies. This is particularly true for elliptical galaxies for which there are two currently popular paradigms. One paradigm is the standard monolithic collapse model in which elliptical galaxies form in a single burst of star formation at high redshift (e.g. Arimoto \& Yohii 1987). In contrast, hierarchical structure formation theories predict that spheroidal galaxies form continuously through a sequence of galaxy mergers (e.g. Cole \etal 1994; Kauffmann 1996). Ashman \& Zepf (1992) explored the properties of globular clusters in models in which elliptical galaxies are the products of the mergers of spiral galaxies, and showed that the greater specific frequency of globular clusters around ellipticals relative to spirals could be explained if globular clusters form during the mergers. They also predicted that elliptical galaxies formed by mergers will have two or more populations of globular clusters - a metal-poor population associated with the progenitor spirals, and a metal-rich population formed during the merger. In contrast, monolithic collapse models naturally produce unimodal metallicity distributions. The discovery that the globular cluster systems of several elliptical galaxies have bimodal colour (and by implication metallicity) distributions (Zepf \& Ashman 1993; Whitmore \etal 1995; Geisler \etal 1996) provides strong support for the merger model. Geisler \etal (1996) and Lee et al. (1998) also show that the red (metal-rich) cluster population is more centrally concentrated than the blue (metal-poor) population, as predicted by Ashman \& Zepf (1992). Recently, an alternative view has been presented by Forbes et al.\ (1997), who suggest that the bimodal color distributions may not be due to mergers, but to a multi-phase single collapse. Although the primary physical mechanism known to produce distinct formation episodes is mergers, it is important to attempt to distinguish between these competing models for the formation of globular cluster systems and their host elliptical galaxies. The kinematics of globular cluster systems may offer such a test of these models. In the multi-phase collapse picture, angular momentum conservation requires that the spatially concentrated metal-rich population rotates more rapidly than the extended metal-poor population. In contrast, simulations of merger models indicate that mergers typically provide an efficient means of angular momentum tranfer, and that the central regions have specific angular momentum that is lower than the outer regions (Hernquist 1993; Heyl, Hernquist and Spergel 1996). Studies of the kinematics of globular cluster systems therefore provide important constraints on the formation history of elliptical galaxies. They also provide useful probes of the mass distribution of elliptical galaxies at radii larger than can be reached by studies of the integrated light. The extended nature of globular cluster systems allows the dynamical mass determined from their velocities to be compared at similar radii to masses determined through studies of the hot X-ray gas. A recent example is the study of the M87 globular cluster system by Cohen \& Ryzhov (1997), who find that a rising mass-to-light ratio is required out to radii of $\sim 3 R_e$, in agreement with X-ray mass determinations. However, M87 occupies a privileged position at the center of the Virgo cluster, so it is critical to test whether the rising mass-to-light ratio, and the agreement with X-ray masses, is true for more typical cluster elliptical galaxies. In this paper we present a spectroscopic study of the globular cluster system of the elliptical galaxy NGC~4472 (M49). This is the brightest elliptical galaxy in the Virgo cluster and has been the subject of a detailed photometric study by Geisler \etal (1996). The only previously published spectroscopic data for the NGC~4472 globular cluster population is by Mould \etal (1990) who presented velocities and line strengths for 26 clusters. The outline of our paper is as follows: Section 2 discusses the sample selection together with our observations and data reduction; Section 3 discusses the kinematics of the metal-rich and metal-poor populations in the context of the merger model, and analyses the implications for the overall M/L ratio in NGC~4472. Finally, we present our conclusions in Section 4. | We have made a detailed spectroscopic study of the globular cluster system of NGC~4472, and have more than doubled the number of confirmed clusters to 57. Whilst this remains a statistically small sample, the data show several interesting properties when combined with the accurate colour/metallicity data from Geisler \etal (1996). When the complete sample is divided into a metal-rich (47\%) and a metal-poor (53\%) subset on the basis of their bimodal colour histogram, the metal-poor subset appears to have a broader distribution of velocities. We have investigated this further, and conclude that the most likely cause is a higher mean level of rotation in the metal-poor cluster system, which is consistent with that of the underlying stellar halo in amplitude and position angle (but with a much higher specific angular momentum). The metal-rich clusters on the other hand show only weak evidence for any rotation, and about an axis which is tilted $\sim 50^o$ from that of the other components. These results are qualitatively in agreement with the predictions of a model in which the metal-rich clusters are formed during the merger of two massive gas-rich galaxies, each with its own old metal-poor cluster population. The cluster system of NGC~4472 forms a dynamically hotter population than the stellar halo, but is consistent with being in dynamical equlibrium with the halo potential defined by the hot X-ray emitting plasma, and supports the presence of a dark $\sim 10^{12}{\rm M}_\odot$ halo in this giant elliptical galaxy. | 98 | 3 | astro-ph9803293_arXiv.txt |
9803 | hep-ph9803309_arXiv.txt | The two possible Mikheyev-Smirnov-Wolfenstein (MSW) solutions of the solar neutrino problem (one at small and the other at large mixing angle), up to now tested mainly through absolute neutrino flux measurements, require flux-independent tests both for a decisive confirmation and for their discrimination. To this end, we perform a joint analysis of various flux-independent observables that can be measured at the SuperKamiokande and Sudbury Neutrino Observatory (SNO) experiments. In particular, we analyze the recent data collected at SuperKamiokande after 374 days of operation, work out the corresponding predictions for SNO, and study the interplay between SuperKamiokande and SNO observables. It is shown how, by using only flux-independent observables from SuperKamiokande and SNO, one can discriminate between the two MSW solutions and separate them from the no oscillation case. | 98 | 3 | hep-ph9803309_arXiv.txt |
||
9803 | astro-ph9803250_arXiv.txt | This study shows one important effect of preexistent cosmic microwave background temperature fluctuations on the determination of the Hubble constant through Sunyaev-Zel'dovich effect of clusters of galaxies, especially when coupled with the gravitational lensing effect by the same clusters. The effect results in a broad distribution of the apparent Hubble constant. The combination of this effect with other systematic effects such as the Loeb-Refregier Effect seems to provide an explanation for the observationally derived values of the Hubble constant currently available based on the Sunyaev-Zel'dovich effect, if the true value of the Hubble constant is $60-80~$km/s/Mpc. It thus becomes possible that the values of the Hubble constant measured by other techniques which generally give a value around $60-80~$km/s/Mpc be reconciled with the SZ effect determined values of the Hubble constant, where are systematically lower than others and have a broad distribution. | The Sunyaev-Zel'dovich (SZ) effect of a cluster of galaxies on the cosmic microwave background (CMB) photons can be used to determine the distance to the cluster hence the Hubble constant ($H_0$), when analysed in conjunction with X-ray observations of the cluster (Cavaliere, Danese, \& De Zotti 1977; Gunn 1978; Silk \& White 1978; Birkinshaw 1979). For an excellent recent review on this subject and other SZ related topics, see Rephaeli (1995 and references therein). The accuracy of the Hubble constant determination depends upon the accuracy of several assumptions involving both sets of observations (radio and X-ray). Perhaps among the most important are the assumptions of sphericity, isothermality of clusters of galaxies (e.g., Inagaki \etal 1995). In this {\it Letter} we point out a completely separate effect on the determination of the Hubble constant due to preexistent, small-amplitude CMB temperature fluctuations before the photons undergo the SZ effect through a cluster. The effect is significantly amplified by the gravitational lensing of the CMB photons by the cluster, because the SZ observational beam size is typically comparable to Einstein radius of the source-lens system. This effect, when coupled with some systematic effects such as the one proposed by Loeb \& Refregier (1997) due to the systematic over-removal of background point radio sources in the beam, may provide an explanation for the observed distribution of $H_0$ determined by SZ effect. | We show that the background CMB fluctuations, especially when they are coupled with the gravitational lensing effect by clusters of galaxies, have one important effect on the determination of the Hubble constant through Sunyaev-Zel'dovich effect of the clusters (for the adopted set of characteristic numbers for the cluster, which seem fairly realistic compared to those of real clusters of interest): a broad distribution of the apparent Hubble constant is produced with a FWHM about $30\%$ of the apparent mean value. The combination of this effect with other systematic effects such as the Loeb-Refregier Effect seems to provide a reasonable explanation for the observationally derived values of the Hubble constant currently available, if the true value of the Hubble constant is $\sim 65~$km/s/Mpc. Thus, it becomes possible that the values of $H_0$ measured by other techniques which generally give a value around $60-80$km/s/Mpc [e.g., $73\pm 10~$km/s/Mpc ($1\sigma$) from Freedman, Madore, \& Kennicutt 1997 based on HST observations of Cepheids; $64\pm 6~$km/s/Mpc ($1\sigma$) from Riess, Press, \& Kirshner 1996 based on type Ia supernova multicolor light-curve shapes; $64\pm 13~$km/s/Mpc ($95\%$ confidence level) from Kundic \etal 1997 based on gravitational lensing time delay measurements; $70\pm 5~$km/s/Mpc ($1\sigma$) from Giovanelli 1997 using I-band Tully-Fisher relation; $81\pm 6~$km/s/Mpc ($1\sigma$) from Tonry 1997 using surface brightness fluctuations] be reconciled with the SZ effect determined values of $H_0$. It may be possible, at least in principle, that one can use a large sample of SZ measured Hubble constant to infer the fluctuations of the CMB at the relevant scales, when the Hubble constant is independently measured to high accuracy by methods such as that using detached eclipsing binaries (Paczynski 1997). | 98 | 3 | astro-ph9803250_arXiv.txt |
9803 | astro-ph9803299_arXiv.txt | {A mechanism of ultra-high energy cosmic ray acceleration in extragalactic radio sources, at the interface between the {\it relativistic} jet and the surrounding medium, is discussed as a supplement to the shock acceleration in `hot spots'. Due to crossing the tangential discontinuity of the velocity the particle can gain an amount of energy comparable to the energy gain at the shock crossing. However, the spectrum of particles accelerated at the jet side boundary is expected to be much flatter than the one formed at the shock. Due to this fact, particles accelerated at the boundary can dominate the overall spectrum at highest energies. In conditions characteristic to extragalactic jets' terminal shocks, the mechanism naturally provides the particles with $E \sim 10^{20}$ eV and complies with the efficiency requirements. The spectrum formation near the cut-off energy due to action of both the shock acceleration and the tangential discontinuity acceleration is modelled with the Monte Carlo particle simulations. It confirms that the upper energy limit can surpass the shock acceleration estimate.} | Jet-like outflows are observed in a number of astrophysical environments, starting from young stars embedded in their parent molecular clouds, up to active extragalactic objects. In the later case, a number of interesting observational phenomena are noted over orders of magnitude of linear sizes. In particular, at the smallest mili-arc-second scales, one often observes relativistic jet velocities, with flow Lorentz factors $\gamma_u$ reaching the values above $10$ (cf. Ghisellini et al. 1996). At larger scales, velocity measurements are more difficult, but without entrainment of large amount of matter near the active galactic nuclear source the jet flow velocity must be also relativistic. A possible loading of a jet with matter is expected to be appended by a substantial amount of turbulence (Henriksen 1987) and related jet kinetic energy dissipation. However, in the FR~II radio sources, there are often observed jets efficiently transporting energy to the far-away hot spots and any jet breaking mechanism can not act too effectively near the central core. Also, the existing hydrodynamical simulations of relativistic jets show for possibility of extended stable jet structures (Marti et al. 1995, 1997; G\'omez et al. 1995). Another argument suggesting the relativistic jet speed at all scales, may be based on the visible asymmetry of jets with respect to the nuclear source, if one believes the effect is caused by the high velocity of the essentially bi-symmetric outflow (cf. Bridle et al. 1994). Let us also note that the Meisenheimer et al. (1989) modelling of the shock acceleration process at extragalactic radio-source hot spots yields `the best-guess' jet velocities in the range ($0.1$, $0.6$) The relativistic movement of the jet leads to shock wave formation in places where an obstacle or perturbation of the flow creates a sudden velocity jump. For jets loaded with a cold plasma the highly oblique conical shocks are formed within the jet tube. These shocks can have a non-relativistic character, involving the velocity jump perpendicular to the shock surface much smaller than the overall jet velocity $U \sim c$. They lead to a limited kinetic energy dissipation and are usually claimed to be responsible for forming the so called `knots' along the jet. A much more powerful shock is formed at the final working surface of the jet. There, a substantial fraction of the jet energy is transferred into heating the jet's plasma, generating strong turbulence, boosting magnetic fields within the turbulent volume, and finally accelerating electrons and nuclei to cosmic ray energies. Rachen \& Biermann (1993) considered the process of particle acceleration to ultra-high energies (UHE) at such shocks. They show that given the favourable conditions the UHE particles up to $\sim 10^{20}$ eV can be formed. Then Rachen et al. (1993) show that assumption of UHE particle acceleration in extragalactic powerful radio sources is compatible with the current measurements of cosmic ray abundances and spectra at energies above $10^{17}$ eV. Additionally, the arrival directions of cosmic ray particles observed above $10$ EeV are correlated with the local galactic supercluster structure (Stanev et al. 1995; see, also, Medina Tanco et al. 1996, Sigl 1996, Sigl et al. 1995, 1996, Hayashida et al. 1996, Elbert \& Sommers 1995, Geddes et al. 1996 and Medina Tanco 1998). An alternative model involving the several-Mpc-scale non-relativistic shocks in galaxy clusters is proposed by Kang et al. (1996; see also Kang et al. 1997). As noted by us (Ostrowski 1990; henceforth Paper~I) a tangential discontinuity of the velocity field can also provide an efficient cosmic ray acceleration site if the considered velocity difference $U$ is relativistic and the sufficient amount of turbulence on both its' sides is present. The problem was extensively discussed in the early eighties by Berezhko with collaborators (see the review by Berezhko 1990) and in the diffusive limit by Earl et al. (1988) and Jokipii et al. (1989). In the present paper we consider the process of ultra high energy cosmic ray acceleration in relativistic jets including the possibility of such boundary layer acceleration. As the considerations of Rachen \& Biermann (1993) treat the acceleration process at relativistic shock in a somewhat simplified way (see, also Sigl et al. 1995), in the first part of the next section (section 2.1) we review this process in some detail in order to understand the inter-relations between the conditions existing near the shock, the accelerated particle spectrum and the particle's upper energy limit. Then, in section (2.2), we present a short description of the basic physical model for the considered acceleration process acting at the jet boundary. We show (section 2.3) that in the conditions characteristic for relativistic jets in extragalactic radio sources, particles with energies above $10^{20}$ eV can be produced in this process without extreme parameter fitting. The required efficiency is discussed in section (2.4). We confirm the estimates presented previously for the shock acceleration, showing that the UHE particle flux observed at the Earth can be reproduced as a result of acceleration processes in jets of nearby powerful radio sources. In section 3 we discuss the problem of the particles' spectrum. With the use of Monte Carlo simulations, we consider the action of both processes acting near the terminal shock in a relativistic jet. Modification of the spectrum due to varying boundary conditions and jet velocity is discussed for the case of ($e^-$, $p$) jets expected to occur in the powerful FRII radio sources (cf. Celotti \& Fabian 1993). The derived particle's upper energy limits are above the shock acceleration estimates and the spectrum modification at highest energies can resemble the observed above 10 EeV `ankle' structure. A short summary and final remarks are presented in section 4. A preliminary report about this work was presented in Ostrowski (1993b, 1996). For the discussion that follows, we consider the jet propagating with the relativistic velocity, $U \sim c$. We use $c$ = 1 units. | 98 | 3 | astro-ph9803299_arXiv.txt |
|
9803 | astro-ph9803066_arXiv.txt | We present new Keck observations of giant arcs in the cluster Abell 2390. High resolution two-dimensional spectra of two arcs show metal lines at $z=4.040\pm 0.005$ with Ly$\alpha$ emission spatially separated from the stellar continuum. In addition, spectroscopy along the notorious `straight' arc reveals two unrelated galaxies, at z=0.913 and z=1.033, with absorption lines of MgII and FeII at z=0.913 seen against the more distant object, indicating the presence of a large gaseous halo. | New high resolution spectra are presented for the high-redshift $z=4.04$ lensed system behind A2390. Figure 1 shows a section of the spectrum aligned along the long axes of the two main arcs. This clearly shows that the Ly$\alpha$ emission is spatially-separated from the continuum light and redshifted with respect to the interstellar lines, indicating an outward flow of enriched gas thought to be typical of starburst galaxies (Lequeux et al. 1995). All stellar and interstellar features are common to both spectra, confirming these arcs are images of one single highly-magnified galaxy. Note the Ly$\alpha$ absorption is seen only in the southern portions of both arcs, coincident with the stellar continuum and with no associated Ly$\alpha$ emission, indicating absorption of this line where the HI column is high. A lens model for this system is discussed in Frye \& Broadhurst (1998). Keck infrared observations were also taken to study the stellar populations of this galaxy, by comparing flux ratios on opposite sides of the 4000 \AA \ break (Bunker, et al. these proceedings). | 98 | 3 | astro-ph9803066_arXiv.txt |
|
9803 | astro-ph9803316_arXiv.txt | We have carried out K-band speckle observations of a sample of 114 X-ray selected weak-line T\,Tauri stars in the nearby Scorpius-Centaurus OB association. We find that for binary T\,Tauri stars closely associated to the early type stars in Upper Scorpius, the youngest subgroup of the OB association, the peak in the distribution of binary separations is at 90 A.U. For binary T\,Tauri stars located in the direction of an older subgroup, but not closely associated to early type stars, the peak in the distribution is at 215 A.U. A Kolmogorov-Smirnov test indicates that the two binary populations do not result from the same distribution at a significance level of 98\%. Apparently, the same physical conditions which facilitate the formation of massive stars also facilitate the formation of closer binaries among low-mass stars, whereas physical conditions unfavorable for the formation of massive stars lead to the formation of wider binaries among low-mass stars. The outcome of the binary formation process might be related to the internal turbulence and the angular momentum of molecular cloud cores, magnetic field, the initial temperature within a cloud, or - most likely - a combination of all of these. We conclude that the distribution of binary separations is not a universal quantity, and that the broad distribution of binary separations observed among main-sequence stars can be explained by a superposition of more peaked binary distributions resulting from various star forming environments. The overall binary frequency among pre-main-sequence stars in individual star forming regions is not necessarily higher than among main-sequence stars. | Taurus-Auriga is the star forming region which has been most thoroughly surveyed for pre-main-sequence binary and multiple systems (see Mathieu 1994 for a review). For separations from 15 A.U.\ to 2000 A.U., the binary frequency among T\,Tauri stars in Taurus is 1.9 times as high as among nearby main-sequence stars (K\"ohler \& Leinert 1998). Extrapolating over the whole range of separations yields a binary frequency of 100\%, i.\,e., each T\,Tauri star in Taurus should be member of a binary or multiple systems. This apparent overabundance of binaries among pre-main-sequence stars is puzzling. One possible explanation is a decrease in the binary frequency as a function of the age of a stellar population (Patience et al.\ 1998). However, a T\,association like Taurus might not be the typical birthplace for low-mass stars, as up to 80\% of all low-mass stars could originate in OB associations (Miller \& Scalo 1978; see also Zinnecker et al.\ 1992). Scorpius-Centaurus is the most nearby OB association at a distance of about 145 parsec (de Zeeuw et al.\ 1998). It consists of three subgroups (cf.\ Figure 1) with ages ranging from 5 to 13 Myr (de Geus et al.\ 1989). Upper Centaurus-Lupus (UCL) is the oldest subgroup of the association. Star formation started here 13 Myr ago and subsequently progressed throughout the parental giant molecular cloud (e.g.\ Blaauw 1991 and references therein). Based on observations with the EINSTEIN X-ray satellite, Walter et al.\ (1994) identified 28 weak-line T\,Tauri stars (WTTS) in Upper Scorpius (US), the youngest subgroup. 10 of these have been surveyed by Ghez et al.\ (1993) for binary or multiple systems, and three binaries have been detected. The EINSTEIN fields covered only a small fraction of US (Fig.\ 2) and the 28 WTTS did not allow for a statistical meaningful study of binary frequencies and separations. A search for visual binary stars among 74 ROSAT selected WTTS and post-T\,Tauri stars in US (Kunkel et al., in prep) was carried by Brandner et al.\ (1996). This survey was sensitive to binary separations down to 0\farcs8, and revealed a rather high binary frequency in the region located between US and UCL (`US-B') and an apparent absence of wide visual binary stars in the center of US (`US-A'). In order to identify closer binary systems and to get a better statistics on possible spatial variations of binary star properties, we have carried out a speckle survey of 114 WTTS in US based on the lists by Walter et al.\ (1994) and Kunkel et al.\ (in prep). \begin{figure*}[htb] \centerline{\plotfiddle{fig1.ps}{11cm}{0}{90}{90}{-520}{-310}} \figcaption{The spatial distribution of 532 proper motion members of the Scorpius-Centaurus association based on HIPPARCOS measurements (adapted from de Zeeuw et al.\ 1998). The boundaries between the subgroups Upper Scorpius (US), Upper Centaurus Lupus (UCL), and Lower Centaurus Crux (LCC) are indicated by dotted lines. Star formation has progressed from the oldest subgroup UCL towards the younger subgroups US and LCC. The two adjacent fields of our survey for binary T\,Tauri stars, centered on US (US-A, solid lines) and at the interface between US and UCL (US-B, dashed lines) are outlined. \label{fig1}} \end{figure*} | We have shown that the distributions of binary separations among weak-line T\,Tauri stars in two adjacent fields (`US-A' and `US-B') in the Scorpius-Centaurus OB association are clearly distinct from each other and considerably more peaked than the (broad) distribution of binary separations observed among main-sequence field stars. In US-A, the WTTS are closely associated with B type stars, whereas in US-B only a few early type stars are present. We conclude that the same physical conditions which facilitate the formation of massive stars also facilitate the formation of closer binaries among low-mass stars, whereas physical conditions unfavorable for the formation of massive stars lead to the formation of wider binaries among low-mass stars. The outcome of the binary formation process might be determined by the critical density at which the magnetic field support breaks down, the internal turbulence and the angular momentum of molecular cloud cores, the initial temperature within a cloud, or - most likely - a combination of all of these. We further conclude that the distribution of binary separations is not a universal quantity. Instead, both the peak and the width of the distribution might vary from one star forming region to the next. The broad distribution of binary separations observed among main-sequence field stars can be understood as a superposition of binary populations originating in various star forming environments with very distinct peaks in the distribution of binary separations. The apparent overabundance of binaries among T\,Tauri stars in the Taurus-Auriga T\,association might be explained by the fact that the distribution of binary separations there is strongly peaked towards $\approx$ 30\,A.U. Extrapolating from this very pronounced peak over the whole range of possible binary separations then leads to an erroneously high estimate of the overall binary frequency. | 98 | 3 | astro-ph9803316_arXiv.txt |
9803 | astro-ph9803120_arXiv.txt | We investigate the non-gaussian properties of cosmic-string-seeded linear density perturbations with cold and hot dark matter backgrounds, using high-resolution numerical simulations. We compute the one-point probability density function of the resulting density field, its skewness, kurtosis, and genus curves for different smoothing scales. A semi-analytic model is then invoked to provide a physical interpretation of our results. We conclude that on scales smaller than $1.5{(\Omega h^2)}^{-1}$Mpc, perturbations seeded by cosmic strings are very non-gaussian. These scales may still be in a linear or mildly non-linear regime in an open or $\Lambda$-universe with $\Gamma=\Omega h \lsim 0.2$. | \label{intro} At present, the two main candidates for the origin of cosmic structure are inflation and topological defects (for a review, see Vilenkin \& Shellard 1994). Although both scenarios may produce a power spectrum of density perturbations consistent with observations, they have very different predictions regarding the statistical properties of the density field. While most inflationary models produce gaussian random-phase initial conditions, defect models produce non-gaussian perturbations particularly on small scales. New results from cosmic-string-seeded structure formation using high-resolution simulations (\markcite{ASWAs,ASWAl}Avelino {\it et al.}~1997, 1998) were encouraging for models with $\Gamma = \Omega h = 0.1$--$0.2$ (see also Battye {\it et al.}~1997); both the mass fluctuation amplitude at $8 h^{-1}$Mpc, $\sigma_8$, and the power spectrum shape of cosmic-string-induced cold dark matter fluctuations, ${\cal P}(k)$, were consistent within uncertainties with observational data (Peacock \& Dodds 1994; Viana \& Liddle 1996). However, because cosmic strings induce non-gaussian density perturbations on small scales, the power spectrum alone is insufficient to describe all the statistical properties of such a density field. This is even more important in open or $\Lambda$-models because in those models the characteristic scales of the density field are shifted to larger scales relative to a flat model with $\Lambda=0$. In this Letter we investigate the non-gaussian properties of the linear density field induced by cosmic strings using higher-order statistics such as the skewness and the kurtosis of a one-point probability density function (PDF), as well as genus statistics. The non-gaussian properties we reveal provide a significant observational signature for cosmic string-seeded structure formation models on small length-scales. We note that previous analytic work has investigated the string-induced velocity field on scales above several $h^{-1}$Mpc, which was inferred to be gaussian (Vachaspati 1992; Moessner 1995), and that some of the features we study here were also observed in global topological defect models, notably for textures (Park, Spergel, \& Turok 1991). Past work on genus statistics in the context of topological defects was made using toy models which incorporated some important features of the models in question (Brandenberger, Kaplan \& Ramsey 1993; Albrecht \& Robinson 1995; Avelino 1997). Our first step in the present analysis was to perform high-resolution numerical simu\-lations of cosmic string networks in an expanding universe (Allen \& Shellard 1990) from which we subsequently computed the causally-sourced density field with either a cold or hot dark matter (CDM or HDM) background. The cosmic string simulations had a dynamic range extending from before the radiation-matter transition at $0.4 \eta_{\rm eq}$ through to deep into the matter era $8.4 \eta_{\rm eq}$, where $\eta_{\rm eq}$ is the conformal time at radiation-matter density equality. The structure formation simulation boxes contained $256^3$ grid-points and their physical volume was in the range $(4$--$100 h^{-1}$Mpc$)^3$. A much more detailed description of these methods is given by Avelino {\it et al.}~(1997, 1998). | We conclude that on length scales smaller than\ $1.5 {(\Omega h^2)}^{-1}{\rm Mpc}$ perturbations seeded by cos\-mic strings are very non-gaussian, especially in the context of a CDM model. In an open or $\Lambda$-universe with $\Gamma=\Omega h \sim 0.15$, this scale will be shifted to $10 h^{-1}$Mpc, which may still be in the linear or mildly non-linear regime, thus potentially providing a strong empirical test for cosmic string models. It has been suggested that such non-gaussianity may imply that it is difficult in string models to deduce the parameter $\beta=\Omega_0^{0.6}/b$ from observations of density and velocity fields (van de Bruck, 1997). However, our results indicate otherwise on large scales because we find that string perturbations are very similar to gaussian-random phase fluctuations, when smoothed on sufficiently large scales. | 98 | 3 | astro-ph9803120_arXiv.txt |
9803 | astro-ph9803193_arXiv.txt | Radio observations establish the B/A magnification ratio of gravitational lens 0957+561 at about 0.75. Yet, for more than 15 years, the {\it optical} magnfication ratio has been between 0.9 and 1.12. The accepted explanation is microlensing of the optical source. However, this explanation is mildly discordant with (i) the relative constancy of the optical ratio, and (ii) recent data indicating possible non-achromaticity in the ratio. To study these issues, we develop a statistical formalism for separately measuring, in a unified manner, the magnification ratio of the {\it fluctuating} and {\it constant} parts of the light curve. Applying the formalism to the published data of Kundi\'c et al. (1997), we find that the magnification ratios of fluctuating parts in both the g and r colors agrees with the magnification ratio of the constant part in g-band, and tends to disagree with the r-band value. One explanation could be about $0.1$ mag of consistently unsubtracted r light from the lensing galaxy G1, which seems unlikely. Another could be that 0957+561 is approaching a caustic in the microlensing pattern. | \label{sect:intro} The gravitational lens system 0957+561 has by now been observed at optical and radio wavelengths for nearly twenty years (Walsh, Carswell, and Weymann 1979; Porcas et al. 1979). Radio studies have definitively established that the B/A magnification ratio of the lens, measured at the core of the radio images (which lies at the location of the optical point source images), is close to 0.75; some recent measurements are $0.75\pm 0.02$ (Garrett et al. 1994), $0.752\pm 0.028$ (Conner et al., 1992). Note that while there is some controversy about the radio magnification ratio at the location of the radio jet, as opposed to the core (see, e.g., Garrett et al. 1994), only the core value interests us here. Because the variability of the quasar in the optical is larger than in the radio, measurement of the B/A magnification ratio in the optical requires that the light curves be shifted by the correct time delay $\tau$ before the ratio is taken. Thus, the earliest determinations of B/A were incorrect. For example, Young et al. (1980) obtained a ratio of 0.76, comfortingly -- yet erroneously -- close to the radio magnification. However, at least from Vanderriest et al. (1989) on, who used a value for $\tau$ quite close to definitive recent determinations (Kundi\'c et al. 1997), it has been clear that the B/A magnification ratio of the optical continuum in the B and A point sources is quite different from 0.75, and moreover has remained at least fairly constant for the full history of observation. Smoothing over observing seasons, Vanderriest et al. (1989) obtained a B/A ratio varying between about 0.9 and 1.05 over the observing seasons early-1980 through early-1986 (times referenced to A component), with a single best-fit value of 0.97. It is debatable whether the variation around the best-fit value is actual time variation of the lens magnification ratio (as distinct from time variation in the quasar luminosity, n.b.) or observational artifact. However, it does seem quite likely that the magnification ratio varied by no more than about $\pm 8$\% during this time. More recently, the value recently obtained by Kundi\'c et al. (1995, 1997) for the 1995 season (A component) is $1.12\pm 0.01$ in g-band (with the error bar, a 95\% confidence limit, depending somewhat on the method of reduction used). So, it is quite plausible (and not contradicted by other measurements in the literature) that the optical B/A remained in the range 0.9 to 1.12 from 1980 through 1995, and possible that the variation has been considerably smaller than this range. The discrepancy between the optical and radio magnification ratios has long been understood as due to microlensing (as predicted by Chang and Refsdal 1979, and Gott 1981). The proper radius of the Einstein ring from a 0.5 $M_\odot$ star at the lens galaxy redshift $z=0.36$, illuminated by the quasar at redshift $z=1.41$, is about $2\times 10^{16}\,h^{-1/2}$cm. Since the radio emission region is much larger than this scale, it averages spatially over the microlensing pattern and is magnified by the macrolens ratio of 0.75. If the optical magnification indeed differs by $\sim 30$\% from the macrolens value, then the optically emitting region must be smaller, or at most a few times larger, than the Einstein ring scale. This accords nicely with (e.g.) the size of an accretion disk smaller than 100 Schwarzschild radii around a $10^9$M$_\odot$ black hole (a scale of $3\times 10^{16}$cm). This Einstein ring radius is only marginally, however, in accord with the apparent constancy of the microlensed magnification ratio: Since the Earth, the microlensing star (or stars, the effect being collective), and the quasar each have (3-dimensional) peculiar velocities of at least 300 km/s, the Earth should move through $\sim 100$\% microlensing variations in $\sim 10$ yr (see Kochanek, Kolatt, and Bartelmann, 1996 for related calculations). So, the observed microlensing is about a factor 10 {\it too constant}, and one is invited to speculate on whether something other than luck is the reason. Another invitation to speculation is the fact that Kundi\'c et al. obtain rather different magnification ratios in their r- and g-band data, with the r ratio being $1.22\pm 0.02$, with, again, the error bar depending on the method of analysis used. By any interpretation of the error bars, however, the r and g results are strongly discrepant. (Again note that there is no assumption that the fluctuations themselves have the same amplitude in the two colors, but only that the magnification {\it ratio} should be the same.) Either a full $0.1$ mag of r-band galaxy light has escaped Kundi\'c et al.'s careful subtraction in the B image, or something else is going on in the lens magnification ratio. With these two hovering peculiarities (possible excess time-constancy, and possible non-achromaticity), it seems useful to try to get additional information on the magnification ratio. This paper therefore asks the questions: Is the optical magnification ratio the same for the source region that produces the {\it fluctuations} in quasar light as it is for the source region that produces the {\it constant} light? And, does the magnification ratio of the fluctuations (which we may call the ``AC'' magnification ratio or ``ACMR'') agree more closely with the r- or g-band magnification ratio previously measured (here called the ``DC'' magnification ratio or ``DCMR'')? The answers to these questions can help diagnose the following situations: (i) If, as is true in many models, the size of the emitting region is much smaller than the microlensing scale, then all the magnification ratios should have the same value. (ii) If there is a problem with r-band galaxy subtraction -- or any other constant source of flux added to one lens component and not the other -- then the ACMR should represent the ``true'' microlens magnification ratio, and we might further expect it to be close to the g-band DCMR (where galaxy subtraction is a much smaller effect). (iii) If the optically emitting quasar accretion disk has a scale comparable to the microlensing scale, and has (as seems almost inevitable) color gradients, then the r and g ACMRs, and r and g DCMRs, might all be distinct. Indeed, the two ACMRs and two DCMRs then provide four distinct windows on the convolution of the accretion disk source with the microlensing pattern. Conceptually, one measures a DCMR and an ACMR as follows: Shift one of the light curves (A,B) in time by $\tau$ to undo the lens delay. Fit each light curve by a constant value plus a residual time-varying part. The ratio of the constant values is the DCMR. Now, for the two time-varying residuals, fit for a model that makes the B residual a constant times the A residual. The best-fitting constant is the ACMR. This conceptual formulation, while simple, is actually not quite right. In the next section, we will give a statistical formulation of the problem that is more complete, and also more directly applicable to unevenly sampled data. In Section 3, we discuss some implementation details, and in Section 4 we apply the formulation to the published data of Kundi\'c et al. (1995, 1997). Section 5 is discussion and conclusions. | While these data, in this analysis, do not support any very definitive conclusions, we may make the following remarks: Occam's razor would seem to indicate that the r-band light curve of Kundi\'c et al. has about $0.1$ mag of residual, unsubtracted, constant light, as perhaps from unmodeled small-scale variations in the lens galaxy surface brightness. If this is the case, then all the data are compatible with a single magnification ratio for both colors and for both the fluctuating and constant pieces. This in turn suggests an accretion disk scale much smaller than the microlensing scale, in accord with theoretical prejudice. The utility of the ACMRs is that, taken together, they strongly favor the hypothesis that the g-band magnification ratio is the correct one, and that nothing more exotic is going wrong. We note, however (per E. Turner, private communication), that the galaxy G1 is something like 2 magnitudes fainter than component B in r band; thus the amount of unsubtracted light would need to be comparable to the total brightness of G1, which seems quite unlikely. It is up to the observers, not us, to decide whether Occam's razor should rule in this case. If $0.1$ mag of residual is not possibly present, then we must conclude that the accretion disk scale is comparable to the microlensing scale, and that the constant r-band part of the disk is {\it more} strongly magnified than (at least some of) the other three regions. It seems likely on physical grounds that the fluctuating regions should be smaller than the constant regions, and that the g-band regions should be smaller than the r-band regions (temperature decreasing outward in the disk). For the larger (r-band and constant) region to have a higher magnification ratio than an enclosed smaller region, the larger region must extend to a place where the magnification is a superlinear function of position in the sky. This might indicate at least a fair chance of the B image passing through a caustic in the near ($\sim 10$ year) future. This possibility, as well as the reconciliation of the relative constancy of the magnification ratio over the last 15 years, will be explored by Monte Carlo simulations in another paper (Press and Kochanek, in preparation). | 98 | 3 | astro-ph9803193_arXiv.txt |
9803 | astro-ph9803008_arXiv.txt | We obtain restrictions on the universal baryon fraction, $f_{\rm B} \equiv \Omega_{\rm B}/\Omega_0$, by assuming that the observed microlensing events towards the Large Magellanic Cloud are due to {\em baryonic} MACHOs in the halo of the Galaxy and by extracting a bound to the total mass of the Milky Way from the motion of tracer galaxies in the Local Group. We find a lower bound $f_{\rm B} > 0.29^{+0.18}_{-0.15}$. Consistency with the predictions of primordial nucleosynthesis leads to the further constraint on the total mass density, $\Omega_0 \alt 0.2$. | It is a Herculean task to inventory the contents of the Universe (e.g., \cite{fhp}). A more modest goal might be to pin down the baryonic fraction of the total mass, $f_{\rm B}$ (e.g., \cite{wnef}, \cite{xrc}). If objects can be identified which are likely to provide a ``fair" sample of $f_{\rm B}$, we may avoid the daunting prospect of having to identify all the guises baryons may assume. Large clusters of galaxies offer a very promising site (\cite{wnef}; \cite{xrc}; \cite{xray1}; \cite{xray2}). To test the estimates of the systematic errors in $f_{\rm B}$ derived from X-ray cluster data, it would be of value to measure $f_{\rm B}$ in a completely different system, provided a case could be made that it will provide a ``fair" sample. Suppose, for example, we could estimate the baryonic mass associated with the Galaxy. If we could also measure the corresponding ``dynamical" mass, we could obtain an independent estimate of $f_{\rm B}$ whose systematic uncertainties (and dependence on the Hubble parameter) differ from those which accompany the X-ray cluster determinations. In this paper we focus on the Local Group of galaxies (LG), using the MACHO mass estimates (\cite{MACHO}) for a lower bound on the baryonic mass and relying on LG dynamics to constrain the total mass estimate. Microlensing experiments (\cite{MACHO}) suggest that roughly half the mass in the halo of our Galaxy, out to the distance of the Large Magellanic Cloud (LMC), may be in the form of Massive Compact Halo Objects (MACHOs). One can imagine several exotic possibilities for the nature of the MACHOs. They could be very dense clusters of non-baryonic dark matter with special properties that allow them to clump inside their Einstein ring radii (\cite{kt94}), or they could be primordial black holes. Neither of these possibilities is especially well motivated and each has its intrinsic difficulties, but neither can be excluded a priori. Stellar remnants such as old white dwarfs\footnote{Neutron stars and black holes of stellar origin cannot constitute a significant halo fraction in view of the constraints arising from the observed metallicity and helium abundances (\cite{ros90}).} appear to offer a more natural candidate (\cite{MACHO}) which, however, is not without its problems too [e.g., white dwarfs require a rather narrow initial mass function in order to avoid overproducing low-mass stars or supernovae (\cite{AL96})]. Dense and cold baryonic gas clouds have also been considered as a viable alternative for the observed gravitational microlenses (\cite{cbc}; \cite{cbc2}). Finally, it must be kept in mind that the observed microlensing may be due to objects which are not in the halo of the Galaxy. If the MACHOs are, indeed, stellar remnants (or cold baryonic gas clouds) in the halo of the Galaxy, then the mass of baryons within 50 kpc of the Galactic center is $M_{\rm B} (50~{\rm kpc}) \geq M_{\rm MACHO} = 2.0^{+1.2}_{-0.7} \times 10^{11}M_\odot$ (\cite{MACHO}). The purpose of the present paper is to extract information on the universal baryon fraction from this number assuming the MACHOs are revealing baryonic matter in the Galaxy halo, and from the dynamics of the Local Group of galaxies. The constraint we obtain may be compared to the one derived from X-ray galaxy clusters (see, e.g., \cite{xrc}; \cite{xray1}; \cite{xray2}), but it relies on different observations in a completely different physical system on a vastly different scale and, interestingly, has a different dependence on the Hubble parameter ($H_0 \equiv 100h $~km~s$^{-1} $~Mpc$^{-1}$). The value of $M_{\rm B} (50~{\rm kpc})$ derived from microlensing experiments is approximately 50\% of the total mass of the Galaxy out to this distance. The latter mass, presumably the sum of baryons and cold dark matter, is derived dynamically (see, e.g., \cite{koch}). However, on the basis of this we cannot conclude that the primordial baryon fraction is $f_{\rm B} \approx 0.5$. Baryons are ``strongly'' interacting particles, while for the (non-baryonic) cold dark matter all interactions except gravitational can be neglected. Consequently, the density profile of the baryonic matter does not necessarily follow the density profile of the cold dark matter, and baryonic matter may be more (or less) concentrated towards the center of the gravitational well. However, we may be able to estimate the primordial baryon fraction if we take the ratio of baryons (as revealed by the MACHOs) to the total mass on some larger scale, which should be sufficiently large so that the matter inflow or outflow across the boundary of the region is negligible. The total mass of matter residing in such a larger region can be found dynamically; however, we cannot measure the mass of baryons separately on such larger scales. Although the baryonic halo may be expected to extend outside of the 50 kpc scale (in the form, e.g., of MACHOs, diffuse gas, satellite Galaxies, etc.), by neglecting these extended baryons we can obtain a lower bound on $f_{\rm B}$. Indeed, while in the past there might have been violent processes of baryon ejection from the Galaxy accompanying, e.g., supernova explosions, analogous ejecta of cold dark matter is not expected. Therefore, by neglecting the unknown ejected component of baryons we will be on the ``safe side'' in our inequality for $f_{\rm B}$, which, we emphasize, does rely on our assumption that MACHOs are baryonic matter in the halo of the Galaxy. For the larger reference scale we can choose the current turnaround radius for the LG. Initially, every shell of the Galaxy's building material expands with the Universe. Gradually, this expansion slows down and eventually a gravitationally bound shell separates from the general expansion. This shell stops expanding and then collapses (\cite{gg72}). The radius of this first stopping point is the turnaround radius. With the passage of time shells that are more and more distant and less and less bound turn around sequentially, i.e., the turnaround radius propagates outward with time (for details see, e.g., \cite{si}; \cite{si2}; \cite{stw}, 1997). There is one shell that is turning around now, at present; the corresponding distance of this shell from the center of mass of the system is the current turnaround radius. Collisionless cold (non-baryonic) dark matter is restricted to remain within this radius, which is just what we want for the larger reference scale. This picture of infall is valid independent of the assumption of spherical symmetry (the turnaround sphere will become a turnaround surface); for the model to be tractable analytically, we do assume spherical infall. | There remain several uncertainties in our LG baryon fraction estimate. One possibility which would weaken or even eliminate our constraint is if some of the observed microlensing events towards the LMC were due to an intervening satellite galaxy between us and the LMC, or due to debris in the LMC tidal tail (\cite{z96}; \cite{z97}). However, the MACHO collaboration concluded (\cite{fg}) that if the lenses were in a foreground galaxy, it must be a particularly dark galaxy; see also (\cite{AG97}). Moreover, the first observation of a microlensing event in the direction of the Small Magellanic Cloud (SMC) (\cite{smc}), implies an optical depth in this direction roughly equal to that in the direction of the LMC. This makes it unlikely that a dwarf galaxy or a stellar stream between us and the LMC is responsible simultaneously for the observed microlensing towards the LMC and the SMC (\cite{fg}; \cite{AG97}). Recently, however, Gates et al. (1997) found Galactic models which explain the current microlensing data by a dark extension of the thick disk, reducing the MACHO fraction. It is to be anticipated that as more microlensing data are accumulated, these uncertainties will be resolved. We note that even in the absence of baryonic MACHOs there is still a limit, albeit much weaker, to $f_{\rm B}$ from LG dynamics. The mass of baryons in the disk of the Galaxy provides a lower bound to $M_{\rm B}$ which is smaller by a factor of $\sim 3$ than the microlensing estimate we have used (\cite{fhp}). Our lower bound to $f_{\rm B}$ would be reduced by this factor while our upper bound to $\Omega_0$ would be increased by the same factor. In summary, if the observed microlensing events are the result of baryonic MACHOs in the Galaxy halo, then the dynamics of the LG may be used to infer a {\it lower} bound to the universal baryonic mass fraction: $f_{\rm B} > 0.29^{+0.18}_{-0.15}\, t_{10}^2$. If primordial nucleosynthesis is used to provide an {\it upper} bound to the present baryonic density, we obtain an {\it upper} bound to the present total mass density: $\Omega_0 \alt 0.2$ (with an extereme upper bound derived using nucleon-to-photon ratio based on the lithium abundance being $\Omega_0 \alt 0.47$). | 98 | 3 | astro-ph9803008_arXiv.txt |
9803 | astro-ph9803187_arXiv.txt | We write a non-relativistic Lagrangian for a hierarchical universe. The equations of motion are solved numerically and the evolution of the fractal dimension is obtained for different initial conditions. We show that our model is homogeneous at the time of the last scattering, but evolves into a self-similar universe with a remarkably constant fractal dimension. We also show that the Hubble law is implied by this model and make an estimate for the age of the universe. \vfill\eject | \indent This is a decisive time for cosmology since our theories for the large scale structure of the universe are being seriously challenged by the ever-growing amount of data. The CfA1 redshift survey (de Lapparent, Geller \& Huchra, 1986, 1988) was the first to reveal structures such as filaments and voids on scales where a random distribution of matter was expected. The most remarkable feature of these structures is the so-called ``great wall" which is a coherent sheet of galaxies extended over an area of at least $60\times 170 $ Mpc (Geller \& Huchra 1989). Later on, deep pencil beam surveys (Broadhurst et al. 1990), the redshift surveys based on IRAS catalogue (Efstathiou et al. (1990a, 1990b), Saunders et al. 1991, Fisher et al 1996), the deep wide angle survey SSRS (de Costa 1988, 1994) and some others have shown inhomogeneities at scales where the galaxy-galaxy and cluster-cluster correlations were believed to be negligible. Of particular significance for the future are the two extensive redshift surveys, SLOAN and 2dF, about to commence, which aim to trace the 3-dimensional distribution of over one million galaxies across the northern and southern skies. The first quantitative study of the cosmic inhomogeneity lead to the well-known $1.8$ power law behaviour of the galaxy-galaxy correlation function (Groth \& Peebles 1977, Peebles 1980, Davis \& Peebles (1983a, 1983b)). Although this law has been consistently identified in different catalogues, the break away from it at larger scales and a crossover to homogeneity has not yet been established (Davis 1996, Pietronero 1987, Coleman \& Pietronero 1992, Pietronero 1996). Whether there is a crossover to homogeneity or not, the power law nature of the two-point and higher-order correlation functions is itself suggestive of some kind of scaling behaviour at least in some range. The simplest structure that obeys such a scaling law is a single fractal. A theoretical model describing a non-analytic inhomogeneous scale-invariant universe is non-existent. The most-extensively-studied inhomogeneous cosmological model is Tolman spacetime (Tolman 1934, Bondi 1947). Tolman's dust solution has been used to model a hierarchical cosmology compatible with the observational analysis of the redshift surveys (Bonnor 1974, Ribeiro (1992a, 1992b), 1993). Recently, the Einstein equation for a scale-invariant spherically symmetric inhomogeneous, but isotropic, universe, which allows a non-vanishing pressure has been solved (Abdalla \& Mohayaee 1997). However, the results obtained in this way are perturbative, assume a preferred center for the universe and violate the linearity of the Hubble law. A self-similar universe avoiding such difficulties can only be constructed for a non-analytic distribution of matter. It is rather a difficult task to construct a fractal metric and to solve Einstein equation for a self-similar universe. However, many cosmological phenomena can be accurately described by the Newtonian gravity, especially in the present matter-dominated era. In this work, we construct a self-similar universe whose dynamics is governed by the Newtonian gravity. We divide the universe into $k$ spherical clusters each of which contains $k$ subclusters which in their turn contain $k$ sub-subclusters. This clustering cascades down all the way to the level of the galaxies which are at the lowest rung on the clustering ladder. The mass and radius of each cluster can be used to define the fractal dimension of our model. We write the kinetic and potential energies of each cluster in terms of its center of mass energy and the internal energies of its subclusters. The thermal energy is obtained by requiring the total entropy of the canonical ensemble of the clusters and their subclusters to remain constant. The final Lagrangian is formulated in terms of two dynamical parameters: radius of the largest cluster and its ratio to the radius of its subclusters. The radius of the largest and smallest clusters, the ratio of the mass to the critical mass contained in a sphere of radius $20$ Mpc, the number of subclusters in each cluster and the ratio that characterises the relative significance of thermal and gravitational energies are left as free parameters. By fixing these to different observational values, we are able to solve the equations of motion numerically using a Pascal program. From the solutions, we can trace the evolution of the fractal dimension and verify the linearity of the velocity-distance relationship over time scales comparable to the age of the universe. The results are remarkable. We observe that for different initial conditions a nearly homogeneous universe with a fractal dimension close to 3 evolves into a universe with a fractal dimension of the order of 2 at the present time. This fractal dimension fluctuates slightly about the value of 2 over the future times but remains on the average constant. We also show that for insignificant thermal energies, the Hubble law is closely obeyed by our model. We also make an estimate for the age of our self-similar universe. This is one of the most challenging problems of Friedmann cosmology since the observed age of the old stars in the globular clusters is far bigger than the value estimated for the age of the universe in the nearly flat standard model. The age of the universe obtained in our model is related to the radius at which the crossover to homogeneity occurs. The farther the crossover radius the older is the universe. This article is organized as follows. In Section 2, we formulate our clustering model. In Section 3, we obtain the kinetic, the potential and the thermal energies, write the Lagrangian and the equations of motion. In Section 4, we solve these equations numerically and discuss the validity and limitations of our Newtonian approximation. In Section 5, we show different plots of the fractal dimension for different choices of the initial condition. Hubble law is discussed in Section 6. In Section 7, we study the evolution of the scale factor and obtain a value for the age of the universe in our model. Section 8 is devoted to the conclusion. | \indent We have constructed a model for a nonrelativistic fractal universe. Our model starts off homogeneous and evolves rapidly to a self-similar universe with a remarkably constant fractal dimension of about 2. The homogeneity at the earlier times explains the isotropy of the microwave background radiation. We have also shown that the Hubble law is closely obeyed by our model for small thermal energies. We have estimated the age of the universe and have shown that it complies with the corresponding observational results. It remains an open problem to extend our model to multi-fractals and to the relativistic regime. {\bf Acknowledgements} We thank R. Mansouri and M. Khorrami for useful discussions. This work has been partially supported by Conselho Nacional de Desenvolvimento Cient\'\i fico e Tecnol\'ogico, CNPq, Brazil, and Funda\c c\~ao de Amparo \`a Pesquisa do Estado de S\~ao Paulo (FAPESP), S\~ao Paulo, Brazil. | 98 | 3 | astro-ph9803187_arXiv.txt |
9803 | astro-ph9803134_arXiv.txt | We have completely mapped the Galactic globular cluster NGC~1851 with large-field, ground-based $VI$ CCD photometry and pre-repair $HST$/WFPC1 data for the central region. The photometric data set has allowed a $V $ vs. $ (V-I) $ colour--magnitude diagram for $\sim$ 20500 stars to be constructed. From the apparent luminosity of the horizontal branch (HB) we derive a true distance modulus $(m-M)_0$ = 15.44 $\pm$ 0.20. An accurate inspection of the cluster's bright and blue objects confirms the presence of seven ``supra-HB'' stars, six of which are identified as evolved descendants from HB progenitors. The HB morphology is found to be clearly bimodal, showing both a red clump and a blue tail, which are not compatible with standard evolutionary models. Synthetic Hertzsprung--Russell (HR) diagrams demonstrate that the problem could be solved by assuming a bimodal efficiency of the mass loss along the red giant branch (RGB). With the aid of Kolmogorov--Smirnov statistics we find evidence that the radial distribution of the blue HB stars is different from that of the red HB and subgiant branch (SGB) stars. We give the first measurement of the mean absolute $I$ magnitude for 22 known RR~Lyr variables ($<M_{I}({\rm RR})> = 0.12 \pm 0.20$ mag at a metallicity [Fe/H]~=~--1.28). The mean absolute $V$ magnitude is $<M_{V}(\rm RR)> = 0.58 \pm 0.20$ mag, and we confirm that these stars are brighter than those of the zero-age HB (ZAHB). Moreover, we found seven new RR~Lyr candidates (six $ab$ type and one $c$ type). With these additional variables the ratio of the two types is now $N_c$/$N_{ab} = 0.38$. From a sample of 25 globular clusters a new calibration for $\Delta V_{\rm bump}^{\rm HB}$ as a function of cluster metallicity is derived. NGC~1851 follows this general trend fairly well. From a comparison with the theoretical models, we also find some evidence for an age--metallicity relation among globular clusters. We identify 13 blue straggler stars, which do not show any sign of variability. The blue stragglers are less concentrated than the subgiant branch stars with similar magnitudes for $r>80$ arcsec. Finally, a radial dependence of the luminosity function, a sign of mass segregation, is found. Transforming the luminosity function into a mass function (MF) and correcting for mass segregation by means of multi-mass King--Michie models, we find a global MF exponent $x_0=0.2\pm 0.3$. | Galactic globular clusters (GGC) are dynamically evolved objects. In order to understand the interplay between the internal dynamical processes and the influence of the Galactic potential, we must study a sample of GGCs comprising objects whose concentration, position in the Galaxy, luminosity and metallicity cover the whole observed range. The mass function and the radial profile must be determined for each cluster, in order to carry out a detailed dynamical analysis. The introduction of large-size CCDs has made this kind of investigations possible. With these detectors it is also possible to obtain deep photometry for the nearest globulars, and therefore to probe their mass functions over large mass intervals, in order to reach those MS stars which are more sensitive to dynamical effects (e.g. Pryor et al. 1986). A rich sample of stars is also essential in order to reveal and study the shortest-lived (and hence poorly known) phases of the stellar evolution (Renzini \& Buzzoni 1986). Furthermore, the interactions between the single stars affect their evolution (e.g. Djorgovski et al. 1991). To establish the reliability of the stellar evolutionary models, we must therefore ascertain to what extent a GC colour--magnitude diagram and luminosity function is changed by the interactions among its stars. For the above reasons, our group started a project aimed at studying a number of globular clusters covering a wide range of the relevant parameters. NGC~1851 ($\alpha_{2000} = 5^{\rm h} 14^{\rm m} 6^{\rm s}.30$; $\delta_{2000} = -40^\circ 2\arcmin 50\farcs00$) has been selected for its position and its concentration. Its galactocentric distance, which is about twice that of the Sun, and its distance of 7.1~kpc from the Galactic plane (Djorgovski 1993) make it a typical halo object. Its concentration $c = 2.24$ is one of the highest in the list of Trager et al. (1995). A recent measurement of the cluster's proper motion has confirmed that NGC~1851 has halo-type kinematics (Dinescu et al. 1996). According to these authors, the space velocities of the cluster are $U=256\pm35$~km s$^{-1}$, $V=-195\pm26$~km s$^{-1}$, $W=-122\pm30$~km s$^{-1}$, $\Pi=195\pm37$~km s$^{-1}$ and $\Theta=167\pm37$~km s$^{-1}$. Past photometric studies of the cluster are given in Alcaino (1969, 1971, 1976), Stetson (1981, hereafter S81), Sagar et al. (1988), Alcaino et al. (1990) and Walker (1992a, hereafter W92). The most exhaustive analysis is that of W92. His main results are that: (1) the cluster core, although unresolved, appears to be blue; (2) the HB is bimodal, showing both a red clump and an extended blue tail; (3) there are no radial trends in the relative numbers of red horizontal branch (RHB), blue horizontal branch (BHB) and red giant branch (RGB) stars for 48\arcsec~ $< r <$ 190\arcsec~; (4) the RGB ``bump'' is at $V$ = 16.15 $\pm$ 0.03 mag; (5) the population ratio $R = N({\rm HB})/N({\rm RGB})$ has a value 1.26 $\pm$ 0.10, which corresponds to a helium abundance $Y$ = 0.23 $\pm$ 0.01 (computed by means of the R-method; e.g. Renzini 1977); (6) there are six blue straggler (BS) stars and six supra-RHB stars $[$15.7 mag $<$ $V$ $<$ 16.0 mag; 0.6 $< (B-V) <$ 0.8$]$ in the region 120\arcsec~ $< r <$ 220\arcsec~, and there is evidence of segregation only for the BS stars, so an origin for supra-RHB stars from BS stars is not supported by W92 data; (7) no significant abundance spread is found from the colour width of the main sequence (MS); and finally (8) an age of 14 $\pm$ 1 Gyr results from the $\Delta$~($B$$-$$V$) method (Sarajedini \& Demarque 1990; VandenBerg et al. 1990). We have now obtained new {\sl large-field} CCD $V$, $I$ photometry for NGC~1851. The new data set makes it possible to re-analyse the stellar content of the cluster with a much richer sample and, for the first time, allows a comprehensive study of its dynamical properties. However, the central regions of the cluster cannot be studied with this ground-based material, due to the extreme crowding of the core. To overcome this limitation, pre-repair {\it Hubble Space Telescope} ({\it HST}) images have been retrieved from the archives and reduced in order to sample the central stellar content, in particular the radial distribution of the HB stars. For the sake of comparison, the photometric catalogue of W92 has been also used. \begin{figure}[t] \psfig{figure=n1851_neg_.ps,width=8.8cm} \caption[]{ The observed NTT/EMMI fields, sketched over a POSS field.} \label{field_map} \end{figure} \begin{small} \begin{table}[t] \caption[]{Log of NTT/EMMI observations.} \label{observs} \begin{tabular}{cccccccc} \noalign{\smallskip} \hline \hline \noalign{\smallskip} Nr. & Field & $t_{\rm exp}$(s) & Filter & Date & FWHM $[\arcsec]$ \\ \noalign{\smallskip} \hline \noalign{\smallskip} 1 & 6 & 50 & $V$ & 1993 Feb 18 & 1.2 \\ 2 & 6 & 70 & $I$ & 1993 Feb 18 & 1.2 \\ &&&&&\\ 3 & 5 & 45 & $I$ & 1993 Feb 18 & 1.2 \\ 4 & 5 & 10 & $I$ & 1993 Feb 18 & 1.4 \\ 5 & 5 & 10 & $V$ & 1993 Feb 18 & 1.1 \\ 6 & 5 & 30 & $V$ & 1993 Feb 18 & 1.3 \\ &&&&&\\ 7 & 2 & 30 & $V$ & 1993 Feb 18 & 1.0 \\ 8 & 2 & 40 & $I$ & 1993 Feb 18 & 1.2 \\ &&&&&\\ 9 & 3 & 60 & $I$ & 1993 Feb 18 & 1.2 \\ 10 & 3 & 44 & $V$ & 1993 Feb 18 & 1.1 \\ &&&&&\\ 11 & 1 & 45 & $V$ & 1993 Feb 18 & 1.0 \\ 12 & 1 & 60 & $I$ & 1993 Feb 18 & 1.2 \\ &&&&&\\ 13 & 4 & 60 & $I$ & 1993 Feb 18 & 1.1 \\ 14 & 4 & 45 & $V$ & 1993 Feb 18 & 1.1 \\ &&&&&\\ 15 & 7 & 45 & $V $ & 1993 Feb 18 & 1.1 \\ 16 & 7 & 60 & $I$ & 1993 Feb 18 & 1.2 \\ \noalign{\smallskip} \hline \noalign{\smallskip} 17 & 8 & 70 & $I$ & 1993 Feb 19 & 1.1 \\ 18 & 8 & 55 & $V$ & 1993 Feb 19 & 1.1 \\ &&&&&\\ 19 & 9 & 55 & $V$ & 1993 Feb 19 & 1.2 \\ 20 & 9 & 70 & $I$ & 1993 Feb 19 & 1.2 \\ \noalign{\smallskip} \hline \noalign{\smallskip} 21 & back & 120 & $V$ & 1993 Dec 10 & 1.0 \\ 22 & back & 180 & $I$ & 1993 Dec 10 & 1.1 \\ \noalign{\smallskip} \hline \end{tabular} \end{table} \end{small} | \label{thediscussion} We have presented new large-field CCD photometry for $\sim20500$ stars in the Galactic halo globular cluster NGC~1851, from both groundbased observations and a pre-repair {\it HST} field. The photometric catalogue has been used to build a $V$ vs. ($V$--$I$) CMD, which has been analysed in detail. An extensive comparison of our data set with the predictions of the stellar models has also been performed. The effects of the dynamical evolution over the main sequence mass function have been investigated by means of a completeness-corrected luminosity function and the radial-count profile. \paragraph{The evolved stellar content} With an accurate inspection of the cluster bright-blue objects, and a comparison with the numbers predicted from the background field and the Galactic count models, we have confirmed the presence of seven ``supra-HB'' stars in the CMD of NGC 1851. We have shown that six of the ``supra-HB'' stars could be evolved descendants from HB progenitors (post-HB or planetary nebulae). We have shown that standard evolutionary models are not able to reproduce the observed bimodal distribution of stars along the HB. Synthetic HR diagrams demonstrate that the problem could be solved by assuming that the efficiency of the RGB mass loss actually encompasses values going from 0.25 to 0.48. We have found evidence that the radial distribution of the blue HB stars is different from that of red HB and SGB stars. The BHB stars are significantly more concentrated than the SGB stars for $r>100$ arcsec. Though this distribution cannot be easily interpreted in terms of dynamical evolution, it might be related to the anomalous distribution of the BSs (see below). All the 27 known variable stars have been identified, and 26 have been measured in both colours (the remaining one being saturated). Twenty-two of them are RR~Lyr variables. For the first time, our photometry has allowed the mean absolute $I$ magnitude of the RR~Lyr variables to be obtained at a metallicity [Fe/H]~=~--1.28. The RR~Lyr are brighter than the ZAHB in the $V$ band, in accordance with the relation given by Carney et al. (1992). The positions and the photometry for seven new RR~Lyr candidates have been given. With these additional variables the ratio of the two types is now $N_c$/$N_{ab} = 0.38$, which reduces the current estimate N$_c$/$N_{ab} = 0.47$ (Wehlau et al. 1982). Thirteen BS stars have been identified outside the inner 80 arcsec. They do not show any sign of variability. We have investigated the radial distribution of the BSS. For $r>80$ arcsec, the BSs are less concentrated than the SGB stars with the same $V$ magnitude. We argue that the distribution of the BSs in the outer envelope of NGC 1851 might be similar to the distribution found by Ferraro et al. (1997) for the BSs in the envelope of M3. We have considered a sample of 25 globular clusters and have derived a new calibration for the $\Delta V_{\rm bump}^{\rm HB}$ parameter as a function of cluster metallicity, and we have found that NGC~1851 follows this general trend fairly well. From a comparison with the corresponding slopes predicted by the isochrones library from Bertelli et al. (1994), we have found that perhaps an age--metallicity relation actually exists among globular clusters, with the metal poorest possibly being older. \paragraph{Dynamical status of NGC~1851} We have been able to derive a complete LF down to $V \simeq 23.5$ mag for stars in the region 190\arcsec~ $< r <$ 650\arcsec~, and down to $V \simeq 22$ mag in the region 120\arcsec~ $< r <$ 189\arcsec~. The external LF is steeper than the internal one, and we have interpreted this result as a sign of mass segregation. By using the most updated mass--luminosity relations we have obtained MFs which can be well fitted by power laws with distinct exponents $x$. The observed value for the external MF is $x = 1.52 \pm 0.18$, which is steeper than the value $0.89 \pm 0.20$ found for the internal one. The global MF has been determined correcting the two observed mass functions for the effects of mass segregation, as predicted by the multi-mass King--Michie model which best fits the observed light profile of NGC 1851. The two values for the slope of the MF are compatible with the model if a global MF exponent $x_0=0.2\pm 0.3$ is adopted. This value for the global MF slope is marginally smaller (MF flatter) than what would be expected from the relation between the slope of the MFs and the position in the Galaxy and the metallicity of the cluster proposed by Djorgovski et al. (1993). This might indicate that NGC 1851 has had a stronger gravitational interaction with the Galactic disc than the average of the Galactic GCs with similar position and metallicity. \paragraph{} The above results indicate that NGC~1851 is a cluster where the dynamical evolution has affected both its evolved and unevolved stellar content. While the single findings are not of high statistical significance (mostly due to the small size of the stellar samples), taken together they give a coherent picture. Stellar encounters have led to mass segregation, as shown by the MF, which is steeper and steeper going from external to internal regions. They have probably contributed to the creation of the observed group of blue straggler stars, and possibly have triggered the formation of a blue tail in the HB. The internal dynamics of NGC~1851 has therefore influenced the evolution of its stars, introducing effects not reproducible by standard models. In turn, the dynamical evolution induced by the external gravitational field of the Galaxy has also very probably contributed to the modification of the present-day stellar population of NGC 1851, as strongly suggested by the anomalously flat global mass function. | 98 | 3 | astro-ph9803134_arXiv.txt |
9803 | astro-ph9803302_arXiv.txt | The shape of the radial velocity and light curves of 24 long-period ($30 \leq P \leq 134$ d) Cepheids in the Magellanic Clouds shows a progression with the period. The sequences of the radial velocity and light curves are based only on a small sample of stars; however, evident changes of the shape can be seen in Cepheids with period between 90 and 134 d. The Fourier parameter--period diagrams for the radial velocity curves show trends which remind in part those of Cepheids with period near 10 d. The plausible interpretation is a resonance, probably $P_0/P_1=2$ between the fundamental and the first overtone mode. The possible importance of this phenomenon for the study of stellar structure and evolution in relatively far galaxies is emphasized \footnote {Tables 2 and 3 are only available in electronic form at the CDS via anonymous ftp to cdsarc.u-strasbg.fr (130.79.128.5)}. | \begin{table*} \caption[]{List of the analyzed long-period Cepheids in Magellanic Clouds} \begin{flushleft} \begin{tabular}{lllllllllll} \hline\noalign{\smallskip} HV & $P_L$ & $N_L$ & Ref$_L$ & ord$_L$ & $\sigma_L$ (mag) & $P_{RV}$ & $N_{RV}$ & Ref$_{RV}$ & ord$_{RV}$ & $\sigma_{RV}$ (km~s$^{-1}$) \\ 904 L & 30.400 & 58 & 10 & 6 & .034 & & & & & \\ 1002 L & 30.4694 & 60 & 1 - 6 & 7 & .032 & & & & & \\ 899 L & & & & & & 31.027 & 49 & 7 & 7 & 1.7 \\ 2294 L & 36.530 & 98 & 2 - 6 & 8 & .032 & & & & & \\ 879 L & 36.817 & 31 & 4 & 5 & .048 & 36.782 & 43 & 7 & 6 & 1.4 \\ 909 L & 37.5828 & 61 & 1 - 4 & 4 & .047 & 37.510 & 64 & 7 & 5 & 1.1 \\ 11182 S & 39.1941 & 44 & 2,3,5 & 3 & .039 & & & & & \\ 2257 L & & & & & & 39.294 & 48 & 7 & 6 & 1.7 \\ 2195 S & 41.7988 & 39 & 1,2,5 & 5 & .049 & & & & & \\ 2338 L & 42.2153 & 55 & 2,4,5 & 7 & .048 & 42.184 & 59 & 7 & 8 & 1.4 \\ 837 S & 42.739 & 55 & 2 - 5 & 6 & .037 & 42.673 & 86 & 9 & 10 & 1.6 \\ 877 L & 45.107 & 78 & 2 - 5 & 3 & .041 & & & & & \\ 900 L & 47.5417 & 62 & 1 - 5 & 5 & .039 & 47.544 & 55 & 8 & 6 & 1.2 \\ 2369 L & 48.3311 & 108 & 1 - 6 & 7 & .034 & & & & & \\ 824 S & 65.8306 & 62 & 1,4 & 5 & .030 & 65.755 & 40 & 8 & 5 & 1.7 \\ 11157 S & & & & & & 69.06 & 62 & 9 & 5 & 1.2 \\ 834 S & 73.399 & 106 & 2 - 6 & 7 & .035 & 73.648 & 53 & 8 & 7 & 1.3 \\ 2827 L & 78.858 & 55 & 3,4 & 3 & .026 & 78.626 & 43 & 7 & 4 & 1.3 \\ 829 S & & & & & & 85.577 & 45 & 8 & 6 & 1.0 \\ 5497 L & 99.078 & 67 & 3 - 6 & 3 & .026 & 99.040 & 41 & 8 & 3 & 1.2 \\ 2883 L & 109.277 & 58 & 2 - 4,6 & 5 & .032 & 109.071 & 52 & 8 & 3 & 2.1 \\ 2447 L & 117.941 & 68 & 3 - 6 & 3 & .022 & 118.841 & 39 & 8 & 2 & 1.3 \\ 821 S & 127.490 & 117 & 1 - 6 & 4 & .040 & 127.083 & 50 & 8 & 4 & 1.4 \\ 883 L & 133.893 & 98 & 1 - 6 & 4 & .045 & 134.75 & 92 & 9 & 6 & 2.7 \\ \noalign{\smallskip} \hline \end{tabular} \end{flushleft} Ref.: 1. Gascoigne \& Kron (1965); 2. Madore (1975); 3. Van Genderen (1977, 1983); 4. Martin \& Warren (1979); 5. Eggen (1977); 6. Freedman et al. (1985); 7. Imbert et al. (1985); 8. Imbert et al. (1989); 9. Imbert (1994); 10. Sebo \& Wood (1995). \end{table*} The longer period Cepheids are particularly important in the context of the primary distance scale because they are bright enough to be visibile at a great distance. However, due to their low number, they have been poorly studied both observationally and theoretically. Simon \& Kanbur (\cite{sk}) considered 50 Cepheids in Galaxy and IC 4182 with period $P$ less than 70 d, compared them with hydrodynamical pulsation models and concluded that a detailed comparison between theory and observations must await a more extensive and accurate sample of observed stars. Antonello \& Morelli (\cite{am}) analyzed all the available photometric $V$ data of galactic Cepheids with period less than 70 d looking for possible resonance effects; they noted some small features in the Fourier parameters--period diagrams which were ascribed tentatively to expected resonances. Aikawa \& Antonello (\cite{aa}) tried to reproduce these observations with nonlinear models, but their conclusion was that the increasing nonadiabaticity of the pulsation with period probably reduces the effectiveness of resonance mechanisms. Finally, Simon \& Young (\cite{sy}) studied long period Cepheids in the period range $10 \leq P \leq 50$ d in Magellanic Clouds looking for galaxy-to-galaxy differences in the Cepheid distributions. Resonances between pulsation modes, which were studied essentially in Cepheids with $P$ less than about 30 d, represent a powerful comparison tool between observations and theoretical model predictions, because they affect the shape of the curves of pulsating stars in specific period ranges. The comparison of the Fourier parameters of observed and theoretical light and radial velocity curves allows to probe the stellar interior and to put constraints on the stellar physical parameters. After the work of Simon \& Lee (\cite{sl}) on the resonance $P_0/P_2=2$ at $P_0 \sim 10$ d between the fundamental and second overtone mode in classical bump Cepheids, several papers by various authors were devoted to this topic, from both the observational and theoretical point of view. For example, Buchler \& Kovacs (\cite{bk}) and Moskalik \& Buchler (\cite{mb}) studied the general effects of 2:1 and 3:1 resonances in radial stellar pulsations and discussed the possible astrophysical implications, Petersen (\cite{pe}) discussed the possible two- and three-mode resonances in Cepheids, and Antonello (\cite{an}) looked for the expected effects in short period Cepheids. Recent reviews on galactic and Magellanic Cloud Cepheids pulsating in fundamental and first overtone mode and on the problems raised by the comparison with the pulsational models are those by Buchler (\cite{bu}) and Beaulieu \& Sasselov (\cite{bs}). The resonance effects in Magellanic Cloud Cepheids cannot be reproduced by models constructed using current input physics and reasonable mass--luminosity relations; in particular the case of first overtone Cepheids characterized by $P_1/P_4=2$ (Antonello et al. \cite{apr}) has proven to be rather difficult for theorists. We mention in passing also the recent resonance $P_2/P_6=2$, studied in the models of hypothetical second overtone mode Cepheids (Antonello \& Kanbur \cite{ak}). In the present work we have considered the long-period Cepheids in Magellanic Clouds, with available photometric and radial velocity data which were suitable for Fourier decomposition. The initial purpose of the work was simply to extend the comparison between theory and observations to Cepheids with the longest known periods, but the probable discovery of a new resonance effect suggested to publish the present Letter in advance of the comparison with the hydrodynamical models (Antonello \& Aikawa, in preparation). | The analysis of the radial velocity and light curves of the long-period Cepheids in the Magellanic Clouds indicates the presence of a progression which we interpret tentatively as an effect of the resonance $P_0/P_1=2$ between the fundamental and the first overtone mode. Since the longest period Cepheids are also the brightest, the present results could be of some importance for the study of stellar structure and evolution in far galaxies, because the stars with $P \sim $ 100 d ($M_V \sim -7$ mag) are about three magnitudes brighter than those at 10 d, in which the well known resonance $P_0/P_2=2$ is occurring. The disadvantage is the low number of such stars. Few Cepheids with $P > 80$ d have been found in relatively nearby galaxies (NGC 6822, IC 1613, NGC 300; see e.g. Madore \cite{mm}), while in the Magellanic Clouds there are just a few of stars in comparison with a total of some thousand Cepheids. Presently the Hubble Space Telescope Key Project on the Extragalactic Distance Scale is optimized for the detection of Cepheids with period between 3 and 60 d (e.g. Ferrarese et al. \cite{fe}), therefore it is not possible to derive reliable conclusions about the number of long-period stars. We just note that in NGC 925 Silbermann et al. (\cite{sil}) found 4 stars with probable $P > 80$ d over a total of 80 Cepheids. Assuming that a sufficient number of such stars is detected and our interpretation is correct, the comparison of the observed resonance effect with nonlinear model predictions will allow to test the input physics and put constraints on the physical parameters of the stars in relatively far galaxies in the same way as it is occurring for the Galaxy and Magellanic Cloud Cepheids with shorter periods. | 98 | 3 | astro-ph9803302_arXiv.txt |
9803 | astro-ph9803244_arXiv.txt | The observed evolution of the galaxy cluster X-ray integral temperature distribution function between $z=0.05$ and $z=0.32$ is used in an attempt to constrain the value of the density parameter, $\Omega_{0}$, for both open and spatially-flat universes. We estimate the overall uncertainty in the determination of both the observed and the predicted galaxy cluster X-ray integral temperature distribution functions at $z=0.32$ by carrying out Monte Carlo simulations, where we take into careful consideration all the most important sources of possible error. We include the effect of the formation epoch on the relation between virial mass and X-ray temperature, improving on the assumption that clusters form at the observed redshift which leads to an {\em overestimate} of $\Omega_0$. We conclude that at present both the observational data and the theoretical modelling carry sufficiently large associated uncertainties to prevent an unambiguous determination of $\Omega_{0}$. We find that values of $\Omega_{0}$ around 0.75 are most favoured, with $\Omega_{0}<0.3$ excluded with at least 90 per cent confidence. In particular, the $\Omega_{0}=1$ hypothesis is found to be still viable as far as this dataset is concerned. As a by-product, we also use the revised data on the abundance of galaxy clusters at $z=0.05$ to update the constraint on $\sigma_8$ given by Viana \& Liddle \shortcite{VL}, finding slightly lower values than before. | The number density of rich clusters of galaxies at the present epoch has been used to constrain the amplitude of mass density fluctuations on a scale of $8\,h^{-1}\,{\rm Mpc}$ (Evrard 1989; Henry \& Arnaud 1991; White, Efstathiou \& Frenk 1993a; Viana \& Liddle 1996, henceforth VL; Eke, Cole \& Frenk 1996; Kitayama \& Suto 1997). This is usually referred to as $\sigma_{8}$, where $h$ is the present value of the Hubble parameter, $H_{0}$, in units of $100\;{\rm km}\,{\rm s}^{-1}\,{\rm Mpc}^{-1}$. However, the derived value of $\sigma_{8}$ depends to a great extent on the present matter density in the Universe, parameterized by $\Omega_{0}$, and more weakly on the presence of a cosmological constant, $\Lambda$. The cleanest way of breaking this degeneracy is to include information on the change in the number density of rich galaxy clusters with redshift \cite{FWED}, the use of X-ray clusters for this purpose having been proposed by Oukbir \& Blanchard \shortcite{OB} and subsequently further investigated \cite{HM,OB97}. Several attempts have been made recently, with wildly differing results \cite{Henry,FBC,Grossetal,BB,Ekeetal,Retal}. The best method to find clusters of galaxies is through their X-ray emission, which is much less prone to projection effects than optical identification. Further, the X-ray temperature of a galaxy cluster is at present the most reliable estimator of its virial mass. This can then be used to relate the cluster mass function at different redshifts, calculated for example within the Press--Schechter framework \cite{PS,BCEK}, to the observed cluster X-ray temperature function. We can therefore compare the evolution in the number density of galaxy clusters seen in the data with the theoretical expectation for large-scale structure formation models, which depends significantly only on the assumed values of $\Omega_{0}$ and $\lambda_{0}\equiv\Lambda/3H^{2}_{0}$, the latter being the contribution of $\Lambda$ to the total present energy density in the Universe. However, the X-ray temperature of a cluster of galaxies is not an easily measurable quantity, as compared to its X-ray luminosity. A minimum flux is required, so that there is a high enough number of photons for the statistical errors in the temperature determination to be reasonably small. Because of this, although estimates of the present-day cluster X-ray temperature function have been available since the early 90's \cite{ESFA,HA}, the change in the cluster X-ray temperature function as we look further into the past has been much more difficult to determine. Estimates for the X-ray temperatures of individual clusters with redshifts as high as 0.3 have been available for some years (e.g.~see David et al.~1993), but only with the advent of the {\em ASCA} satellite has it been possible to measure X-ray temperatures for clusters of galaxies around that redshift in a systematic way, and to go to even higher redshifts. The evolution of the cluster X-ray luminosity function with redshift, though easier to determine, provides much weaker constraints on $\Omega_{0}$ and $\lambda_{0}$, due to the fact that the X-ray luminosity of a galaxy cluster is not expected to be a reliable estimator of its virial mass (e.g.~Hanami 1993). Though it could in principle provide some indication of the change of the cluster X-ray temperature function with redshift, the problem is that not only is there considerable scatter in the present-day cluster X-ray temperature verses luminosity relation \cite{Davetal,Fetal}, but it is also not known how the relation may change with redshift, though recently it has been argued that at least up to $z=0.4$ it does not seem to evolve \cite{MScharf,AF,Retal,SBO}. The deepest complete X-ray sample of galaxy clusters presently available is the one obtained from the {\em Einstein Medium Sensitivity Survey} ({\em EMSS}) \cite{Getal,Hetal}. This sample is restricted to objects with declination larger than $-40^{\rm o}$ and is flux-limited, with $F_{{\rm obs}}\geq1.33\times10^{-13}\;{\rm erg}\,{\rm cm}^{-2}\,{\rm s}^{-1}$, where $F_{{\rm obs}}$ is the cluster flux in the 0.3 to 3.5 keV band which falls in a $2'.4\times2'.4$ {\em EMSS} detect cell. It presently contains 90 clusters of galaxies, after a few misidentifications were recently removed \cite{GioiaL,Netal}. This is the only complete galaxy cluster catalogue beyond a redshift of 0.3, and as such unique in providing the means to distinguish between different possible values for $\Omega_{0}$ and $\lambda_{0}$. However, until the recent effort by Henry \shortcite{Henry}, very few X-ray temperatures were known for those galaxy clusters in the {\em EMSS} sample with redshifts exceeding 0.15 (see Sadat et al.~1998 for a recent compilation). Henry \shortcite{Henry} used {\em ASCA} to observe all galaxy clusters in the {\em EMSS} cluster sample with $0.3\leq z \leq 0.4$ and $F_{{\rm obs}}\geq2.5\times10^{-13}\;{\rm erg}\,{\rm cm}^{-2}\,{\rm s}^{-1}$. The resulting sub-sample of 10 clusters has a median redshift of 0.32, and the data obtained for each cluster, the X-ray flux, luminosity and temperature, can be found in Table 1 of Henry (1997). We will use this data together with the present-day (median redshift 0.05) cluster X-ray temperature function. We work within the extended Press--Schechter formalism proposed by Lacey \& Cole (1993, 1994), which allows an estimation of the formation times of dark matter halos. We will assume the dark matter to be cold, and consider the cases of an open universe, where the cosmological constant is zero, and a spatially-flat universe, such that $\lambda_{0}=1-\Omega_{0}$. | {}From the above analysis, we conclude that {\em at present} it is not possible to reliably exclude any interesting value for $\Omega_{0}$ on the basis of X-ray cluster number density evolution alone, due to the limited statistical significance of the available observational data and to uncertainties in the theoretical modelling of cluster formation and evolution. However, we do find that values of $\Omega_{0}$ below 0.3 are excluded at least at the 90 per cent confidence level. Values of $\Omega_{0}$ between 0.7 to 0.8 are those most favoured, though not strongly. These results are basically independent of the presence or not of a cosmological constant. Our conclusions support those of Colafrancesco, Mazzotta \& Vittorio \shortcite{CMV}, who tried to estimate the uncertainty involved in the estimation of the cluster X-ray temperature distribution function at different redshifts based on its present-day value. They found this uncertainty, given the still relatively poor quality of the data, to be sufficiently large to preclude the imposition of reliable limits on the value of $\Omega_{0}$. Our results disagree with those of Henry \shortcite{Henry} and Eke et al.~\shortcite{Ekeetal}, as they found the preferred $\Omega_{0}$ to lie between 0.4 to 0.5, with the $\Omega_{0}=1$ hypothesis strongly excluded. This disagreement is mainly the consequence of our focus on the threshold X-ray temperature of 6.2 keV, while they draw their conclusions based on the analysis of the results obtained for several threshold X-ray temperatures. Further below we will repeat our calculations assuming a threshold X-ray temperature of 4.8 keV, and we will find that when we calculate the joint probability of some value for $\Omega_{0}$ being excluded on the basis of the results concerning either one or both threshold X-ray temperatures of 6.2 keV and 4.8 keV, the favoured value for $\Omega_{0}$ decreases to around 0.55. Some of the reasons for our choice of deriving the conclusions solely based on the results obtained for the 6.2 keV threshold were mentioned at the end of subsection 3.1 and others will be detailed below. Other less important contributions to the difference between our results and those presented by Henry \shortcite{Henry} and Eke et al.~\shortcite{Ekeetal} are the different assumed normalization for the virial mass to X-ray temperature relation, and the corrections in the expected values in the Universe for both $N(>6.2\,{\rm keV},\,0.05)$ and $N(>6.2\,{\rm keV},\,0.32)$ due to the uncertainties in the X-ray cluster temperature measurements. Note that changing the mean of the bootstrap distribution obtained for $N(>6.2\,{\rm keV},\,0.32)$ to its theoretically-expected overall value in some $\Omega_{0}$ universe and then calculating the exclusion level on the estimated value for $N(>6.2\,{\rm keV},\,0.32)$ in the Universe given the dataset in Henry \shortcite{Henry}, rather than just using the original bootstrap distribution to impose an exclusion level on the theoretically-expected overall value for $N(>6.2\,{\rm keV},\,0.32)$ in that $\Omega_{0}$ universe, does not seem to make much difference. This is a reflection of the fact that the bootstrap distributions recovered do not have a strongly asymmetric shape. Our disagreement with Eke et al.~\shortcite{Ekeetal} on the level of exclusion of the $\Omega_{0}=1$ hypothesis is also due to our much larger assumed uncertainty in the theoretically-expected overall value for $N(>6.2\,{\rm keV},\,0.32)$. For the $\Omega_{0}=1$ hypothesis to be favoured, one requires the lowest possible observed value for $N(>6.2\,{\rm keV},0.32)$. This is best achieved if, for the sample of 10 galaxy clusters used in its calculation, the X-ray temperatures turn out to be on average lower than the assumed mean, and the X-ray fluxes higher. A higher ratio between the extended and detect cell fluxes for the {\em EMSS} at $z=0.32$ would also help. On the theoretical side, the higher one decides the expected value for $N(>6.2\,{\rm keV},0.32)$ is, the more compatible with the data the $\Omega_{0}=1$ hypothesis becomes. This can be best achieved if, in decreasing order of importance, the cluster virial mass at fixed X-ray temperature is being underestimated, $\delta_{{\rm c}}$ is lower than the canonical value 1.7 and $f$, the assembled fraction of a cluster virial mass after which the X-ray temperature does not change significantly, is assumed greater than 0.75. However, the single most important factor in determining the theoretically-expected overall value for $N(>6.2\,{\rm keV},0.32)$ is the present-day normalization for the dispersion of the density field, $\sigma_{8}$, which in turn results from the observational value for the present density $N(>6.2\,{\rm keV},\,0.05)$. Although we worked with all X-ray clusters that make up the dataset in Henry \shortcite{Henry}, and even estimated the effect of also considering the 5 clusters with lower X-ray fluxes present in the {\em EMSS} in the redshift bin from 0.3 to 0.4, in fact we only used the abundance of clusters with X-ray temperatures in excess of 6.2 keV to constrain $\Omega_{0}$. We mentioned some of the reasons for this choice in Section~3. Nevertheless, we decided to repeat the same calculations for a threshold X-ray temperature of 4.8 keV. This value also well represents the mean curve going through the observed cumulative X-ray temperature distribution function at both $z=0.05$ and $z=0.32$. \begin{figure} \centering \leavevmode\epsfysize=5.4cm \epsfbox{zclus_fig4.eps}\\ \caption[Figure4]{The absolute exclusion levels for different values of $\Omega_{0}$ in both the open and spatially-flat cases, when the threshold X-ray temperature of 4.8 keV is used.} \end{figure} The results regarding the best-fit value for $\Omega_{0}$, presented in Figure~4, are somewhat different from those we obtained when the threshold X-ray temperature was assumed to be 6.2 keV. This is particularly true if the correction for the possibility of any of the 5 clusters with the lowest X-ray fluxes in the $0.3<z<0.4$ EMSS sub-sample having X-ray temperatures in excess of 4.8 keV is included, as can be seen in Figure 5 for the open case. While the standard analysis without these 5 X-ray clusters prefers a value for $\Omega_{0}$ between 0.4 to 0.5, when the correction for the scatter in the relation between the cluster X-ray temperature and luminosity is included, in the way described in subsection 3.3, the preferred value for $\Omega_{0}$ decreases to about 0.3. Now the $\Omega_{0}=1$ hypothesis is excluded at more than the 95 per cent confidence level, with or without the correction. At the 90 per cent confidence level, one finds that $\Omega_{0}>0.8$ is excluded without the correction, being this limit lowered to 0.7 when the correction is included. One can also estimate the joint probability of some $\Omega_{0}$ value being excluded on the basis of the results relative to either one or both X-ray temperature thresholds. Assuming the data used in the calculations for the two thresholds is independent, the results then imply that the favoured value for $\Omega_{0}$ is close to 0.55 (0.50 if the incompleteness correction is included) and the $\Omega_{0}=1$ hypothesis is excluded at the 99 per cent level. This agrees very well with the results of Henry \shortcite{Henry} and Eke et al.~\shortcite{Ekeetal}, leading us to believe that the main difference between our analysis and theirs is our decision to draw our conclusions solely based on the exclusion levels obtained for the X-ray temperature threshold of 6.2 keV. A further potential problem one must consider when working with clusters whose observed X-ray temperature is as low as 4.8 keV is the possibility that the energy in the intracluster gas has increased as a result of (pre-)heating by supernovae and starbursts in the cluster galaxies. In fact this is the leading hypothesis (e.g. Navarro, Frenk \& White 1995; Markevitch 1998) put forward to explain the discrepancy between the observed slope of the X-ray temperature--luminosity relation, close to 0.3, and the expected value of 0.5 if clusters evolve in a self-similar way \cite{Kaiser}. Following Eke et al.~\shortcite{Ekeetal}, we assume that in a cluster whose observed X-ray temperature is 4.8 keV, 17 per cent of its energy, that is 0.8 keV per intracluster gas particle, was due to (pre-)heating produced by processes occurring inside the cluster galaxies. This is approximately the amount of energy that gets injected into the intracluster gas particles in the simulation of Metzler \& Evrard \shortcite{ME}, where a galaxy cluster's X-ray temperature, which would otherwise be 5.6 keV, increased to 6.4 keV. Note however that in the scheme proposed by Eke et al.~\shortcite{Ekeetal} a cluster this large would not be (pre-)heated to the extent simulated by Metzler \& Evrard \shortcite{ME}, as in their proposal Eke et al.~\shortcite{Ekeetal} assume that the energy gained by each intracluster gas particle due to (pre-)heating decreases as a galaxy cluster becomes larger, being close to zero for galaxy clusters with X-ray temperatures exceeding 6.2 keV. \begin{figure} \centering \leavevmode\epsfysize=5.4cm \epsfbox{zclus_fig5.eps}\\ \caption[Figure5]{The absolute exclusion levels for different values of $\Omega_{0}$ only for the open case, when the threshold X-ray temperature of 4.8 keV is used. The full curve includes a correction (FC) for the possibility of the 5 clusters with lowest fluxes in the {\em EMSS} located between $z=0.3$ and $z=0.4$ having X-ray temperatures in excess of 4.8 keV. The dashed curve includes a correction (HC) for the possibility of (pre-)heating of the intracluster medium due to processes within the cluster galaxies. The dotted curve includes both corrections.} \end{figure} The above assumption means that the observed values for $N(>4.8\,{\rm keV},\,z)$, when $z=0.05$ and $z=0.32$, should now be compared with the theoretically-expected values for $N(>4.0\,{\rm keV},\,z)$ at those redshifts. The resulting exclusion levels on the value of $\Omega_{0}$ can be seen in Figure 5 for the open case. There is little difference compared to the results in Figure 4 that follow from the standard no-heating calculation. The lower value for $\sigma_{8}$, required to match theory and observations at $z=0.05$, more than compensates for the expected increase in the number of galaxy clusters with X-ray temperatures in excess of 4.8 keV at $z=0.05$, in effect bringing this number down. In fact, the standard no-heating calculation for a threshold X-ray temperature of 4.8 keV requires a value for $\sigma_{8}(\Omega_{0})$, so that the observed value for $N(>4.8\,{\rm keV},\,0.05)$ is reproduced, that is less than 3 per cent below that required by the $>6.2$ keV data, quoted in equation (\ref{final1}). On the other hand, including the (pre-)heating correction, the required $\sigma_{8}(\Omega_{0})$ value drops to 19 per cent below that preferred by the $>6.2$ keV data. Though the coincidence between the $\sigma_{8}$ values obtained for the two X-ray temperature thresholds 4.8 keV and 6.2 keV under the no-heating assumption may be accidental, it could indicate that (pre-)heating was relatively unimportant at least for the galaxy clusters observed at $z=0.05$ with X-ray temperatures exceeding 4 keV. If (pre-)heating was more important in the past than today, then the required $\sigma_{8}(\Omega_{0})$ value would be that obtained through the standard no-heating hypothesis, but the comparison at $z=0.32$ would include the (pre-)heating correction. This would push the theoretically-expected value for $N(>4.8\,{\rm keV},\,0.32)$ up, favouring higher values for $\Omega_{0}$. This is not as far-fetched as it may seem, given that it is well known that the star-formation rate peaks before $z=1$ (e.g. Madau, Ferguson \& Dickinson 1998; Baugh et al. 1998), and consequently so does the rate of supernovae Type II (the rate of supernovae Type Ia peaks a few Gyr later) and the probability of starbursts. The results for the 4.8 keV threshold X-ray temperature are close to those found by Eke et al.~\shortcite{Ekeetal}, leading us to believe that their exclusion levels for $\Omega_{0}$ are dominated by the information associated with the threshold X-ray temperatures 4.0 keV and 5.0 keV. In our view, the analysis for these X-ray temperature thresholds carries with it a sufficient number of uncertainties, due to the problems mentioned above, so as to render the constraints imposed on $\Omega_{0}$ not very trustworthy. Only the data regarding clusters with X-ray temperatures in excess of about 6 keV seems sufficiently free of modelling problems so as to be potentially useful in constraining $\Omega_{0}$. Another possible complication has arisen from recent work by Blanchard, Bartlett and Sadat \shortcite{BBS} who use a sample of 50 galaxy clusters with mean redshift of 0.05, which were identified through the {\em ROSAT} satellite, to estimate the cumulative X-ray temperature distribution function at $z=0.05$. They claim the number density of galaxy clusters at $z=0.05$ with X-ray temperatures exceeding 4 keV is being {\em underestimated} when the Henry \& Arnaud cluster sample is used. Through the X-ray cumulative temperature distribution function at $z=0.05$ they obtain, they then estimate $\Omega_{0}$ using the {\em EMSS} cluster abundance in the redshift bin $0.3<z<0.4$ and the X-ray temperature data gathered in Henry \shortcite{Henry}. They find the favoured value for $\Omega_{0}$ to be 0.75, while $\Omega_{0}<0.3$ is excluded at more than the 95 per cent level. These results coincide very well with ours when only the 6.2 keV threshold X-ray temperature is considered, thus perhaps implying that the discrepancy between the favoured value for $\Omega_{0}$ found when different X-ray temperature thresholds are considered may arise from a underestimation of the cumulative distribution function at $z=0.05$ for X-ray temperatures below about 6 keV. Unfortunately, due to uncertainties associated both with the observational measurements and the theoretical modelling of cluster evolution, the presently-available data on galaxy clusters with X-ray temperatures exceeding about 6 keV is not able to strongly discriminate between cosmologies with different values for $\Omega_{0}$. And in any case, the data available is probably not yet statistically significant. More is needed to support or disclaim the preliminary conclusions that can be obtained from it. In particular there are some oddities with the sub-sample of {\em EMSS} galaxy clusters observed by Henry, such as the strange redshift distribution, strongly clustered around $0.32$, and the unexpectedly low X-ray temperature of MS2137.3, that makes one have some doubts about how representative this dataset is of the Universe. Within the next few years, with the launch of the {\em XMM} satellite, possibly in late 1999, a significant increase in the quantity and quality of the available data is expected to occur \cite{Romer}. It should then be possible to place stronger constraints on $\Omega_{0}$ on the basis of the evolution of the galaxy cluster X-ray temperature function. This would be helped by improvements in the theoretical modelling of cluster evolution, perhaps based on the high-resolution hydrodynamical $N$-body simulations on cosmological scales expected in the near future. | 98 | 3 | astro-ph9803244_arXiv.txt |
9803 | astro-ph9803323_arXiv.txt | BATSE, Ulysses, and TGRS and KONUS on WIND detected four gamma-ray events within 1.8 days during 1996 October 27-29, consistent with coming from the same location on the sky. We assess the evidence that these events may be due to a series of bursts from a single source by calculating the probability that such a clustering in position and in time of occurrence might happen by chance. The calculation of this probability is afflicted by the usual problem of a posteriori statistics. We introduce a clustering statistic, which is formed from the "minimum circle radius" (i.e. the radius of the smallest circle that just encloses the positions of all the events) and the minimum time lapse (i.e. the time elapsed between the first and last event). We also introduce a second clustering statistic, which is formed from the "cluster likelihood function" and the minimum time lapse. We show that the use of these statistics largely eliminates the "a posteriori" nature of the problem. The two statistics yield significances of the clustering of $3.3\times 10^{-4}$ and $3.1\times 10^{-5}$, respectively, if we interpret the four events as four bursts, whereas the clustering is not significant if we interpret the four events as only three bursts. However, in the latter case one of the bursts is the longest ever observed by BATSE. | 98 | 3 | astro-ph9803323_arXiv.txt |
||
9803 | astro-ph9803115_arXiv.txt | { We have used the results of recent smoothed particle hydrodynamic simulations of colliding stars to create models appropriate for input into a stellar evolution code. In evolving these models, we find that little or no surface convection occurs, precluding angular momentum loss via a magnetically-driven stellar wind as a viable mechanism for slowing rapidly rotating blue stragglers which have been formed by collisions. Angular momentum transfer to either a circumstellar disk (possibly collisional ejecta) or a nearby companion are plausible mechanisms for explaining the observed low rotation velocities of blue stragglers Under the assumption that the blue stragglers seen in NGC 6397 and 47 Tuc have been created solely by collisions, we find that the majority of blue stragglers cannot have been highly mixed by convection or meridional circulation currents at anytime during their evolution. Also, on the basis of the agreement between the predictions of our non-rotating models and the observed blue straggler distribution, the evolution of blue stragglers is apparently not dominated by the effects of rotation. } \newpage | Blue stragglers lie roughly along an extension of a star cluster's main sequence (MS) and are generally bluer and brighter than the turn-off (TO) stars. While several possible mechanisms for their formation have been proposed (e.g. delayed star formation, binary mass transfer, binary coalescence, stellar collisions; see Livio, 1993, and Stryker, 1993, for reviews), \nocite{livio93}\nocite{stryker93} the collision scenario, in which initially unbound stars come into contact and merge during a dynamical encounter, has received the most attention recently (e.g. Sandquist \hbox{\em et al.} 1997; Ouellette \& Pritchet 1996; Sills, Bailyn \& Demarque 1995). \nocite{sbh97}\nocite{op96}\nocite{sbd95} Blue stragglers formed by collisions are of particular interest because their formation rate tells us about the stellar interaction rate in their cluster environment and the dynamical history of the cluster itself (Bailyn, 1995).\nocite{bailyn95} There are at least two ways to investigate the formation rate of blue stragglers by collisions. One is to perform N-body simulations, modelling the stellar environment and determining the collision rate directly. This requires knowledge of the stellar interaction cross-section, local stellar density, mass function, and binary fraction. The second is to model the structure of the merger remnants and to evolve them using a stellar evolution code. Just as the age of a cluster is found by comparing theoretical isochrones, based on standard stellar models, with an observed colour-magnitude diagram (CMD), comparison of the evolution of the merger models to the observed numbers and distribution of blue stragglers in the CMD can tell us about the distribution of lifetimes, and the production rates, of these stars. In this paper, we develop and evolve models of collisional merger remnants and compare the predictions of these models with the observed blue straggler distributions in NGC 6397 and 47 Tuc. | We have evolved stellar models which are appropriate for stars created during collisions between equal and unequal mass stars. In doing so we have used the results of SPH simulations of colliding polytropes, most importantly the predictions of entropy stratification. To produce these models, we have made several assumptions to facilitate the incorporation of the results of the SPH simulations of collisions between {\em polytropic} stars into models of real stars; these assumptions include neglecting mass loss, rotation, and shock heating during the collision. Because we have restricted our attention to head-on parabolic collisions, we believe these approximations to be reasonably valid; mass loss and rotational velocity are shown to be small in this case by the simulations of LRS; shock heating is shown to be small and, most importantly, should not affect the distribution of helium-rich gas in the merger remnant (Sec~\ref{sec:shock}), although it is likely that shock heating will affect the distribution of material near the surface of the star. The form of the energy injection term $\epsilon_x$ which we use to expand our merger remnant models is constrained by the results of SPH, but its form is nonetheless not crucial as long as significant mixing is not artificially introduced (via the Vogt-Russell Theorem; Vogt 1926, Russell 1927). In comparing the predictions of these models with the blue stragglers seen in two dense clusters, NGC 6397 and 47 Tuc, we have also assumed that all of the blue stragglers in these clusters have been formed through collisions. The apparent agreement between the predictions of our unequal mass merger models and the observed blue straggler distributions NGC 6397 and 47 Tuc suggests that \begin{itemize} \item little or no mixing occurred during the formation of these blue stragglers, either of fresh hydrogen to the cores, or of helium to the surfaces. As explained earlier, such mixing would produce a significant blueward scatter in the distribution of blue stragglers in the CMD --- as it is, our models with the best agreement with the observations, the unequal mass mergers, produce rather red stars, relative to the clusters' ZAMS. Regardless of the formation mechanism for the blue stragglers in these clusters, it is apparent from their location redward of the ZAMS that they are formed as {\em evolved} stars (Ouellette \& Pritchet, 1996). \item there may be some form of efficient AML mechanism acting to slow the rotation of these stars. Blue stragglers which are formed by collisions are expected to be rapidly rotating: the affect of rapid rotation on the observed distribution of blue stragglers in the CMD should result in a poor agreement with our non-rotating models, and yet the agreement appears to be quite good. The thin, short-lived convective envelopes seen in our models precludes a magnetically driven stellar wind AML mechanism to explain the slow rotation rates and apparent lack of rotational effects; however, angular momentum transfer to a circumstellar disk or nearby companion are still plausible. \end{itemize} The lack of surface convection in our most massive models and the thin convective envelopes seen in our less massive models means that surface abundances of collisionally merged blue stragglers might not be altered from those of the parent stars, unless some amount of hydrodynamical mixing occurs during the collision. Additionally, meridional circulation currents, which should be confined to the stellar envelope except in the most extreme cases, may act to mix the surface layers, regardless of the amount of convection. The extreme blue stragglers observed in both NGC 6397 and 47 Tuc occupy the region of the CMD which should be populated by highly mixed stars. It is possible that these blue stragglers are the few that were born rotating rapidly enough that the meridional circulation currents are able to penetrate more deeply in to the star; if so, these stars should still be rapidly rotating. It is, however, possible that some fraction of these stars are merely photometric errors. | 98 | 3 | astro-ph9803115_arXiv.txt |
9803 | astro-ph9803265_arXiv.txt | s{Fourteen classical double radio galaxies with redshifts between zero and two were used to determine the cosmological parameters $\Omega_m$, $\Omega_{\Lambda}$, and $\Omega_k$, where these are the normalized values of the mean mass density, cosmological constant, and space curvature at the present epoch. A low value of $\Omega_m$ is obtained, and $\Omega_m = 1$ is ruled out with 97.5 \% confidence. The low value of $\Omega_m$ determined using the radio source method described here is also indicated by several independent tests. Thus, it appears that either a cosmological constant, or space curvature, is significant at the present epoch. This means that the universe is undergoing, or has recently undergone, a transition away from a state of matter domination and into a state where either a cosmological constant or space curvature is determining the expansion rate of the universe. The low value of $\Omega_m$ presented here and by Guerra \& Daly (1998) means that we can state with 97.5 \% confidence that the universe will continue to expand forever. } | The structure and fate of the universe can be described by the cosmological parameters $\Omega_m$, $\Lambda$, and $k$, where $\Omega_m$ is the average density of matter in the universe at the present time divided by the critical density, $\Lambda$ is the current value of the cosmological constant (generally, though not always, taken to be time-independent), and $k$ describes the global geometry of the universe. The normalized values of $\Lambda$ and $k$ are denoted $\Omega_{\Lambda}$, and $\Omega_k$ (e.g. Peebles 1993). Assuming that any relativistic component, such as the microwave background radiation, has a negligible contribution to the total mass-energy density at the present time, the following equation must be satisfied: $\Omega_m$ + $\Omega_{\Lambda}$ + $\Omega_k$ = 1. As originally discussed by Dicke (1970), and later by Peebles (1993), each term evolves differently with redshift, so it is unlikely that two terms will be comparable at any given time. However, if $\Omega_m < 1$, then, the universe is undergoing a transition away from a state of matter domination into a state where either space curvature or the cosmological constant is dominant. Following Dicke's argument, it is rather unlikely that all three terms will be significant at the present time. The cosmological parameters can only be believably determined when several independent methods of estimating the parameters all yield similar values, both within each category of test, and comparing results from different categories. At present, there are three main categories of estimates of $\Omega_m$: (1) local, low-redshift dynamical tests; (2) tests that depend on the coordinate distance to high-redshift sources through the angular size distance or the luminosity distance; and (3) tests using the fluctuations of the microwave background radiation on different angular scales. At the present time, values of $\Omega_m$ have been determined using methods (1) and (2); results obtained using method (1) are mentioned below. Method (2) has been applied to radio galaxies with redshifts between zero and two by Daly (1994, 1995), Guerra \& Daly (1996, 1998), and is applied here to radio sources with redshifts between zero and two. Method (2) has also been applied to supernovae by Garnavich {\it et al.} (1998) and Perlmutter {\it et al.} (1998), who study sources to a redshift of about one. At some point in the future, there may be independent constraints from measurements of fluctuations of the microwave background radiation on different angular scales. Several local, low-redshift estimates of $\Omega_m$ indicate that it is significantly less than unity (Hudson {\it et al.} 1995; Shaya, Peebles, \& Tully 1995; Carlberg {\it et al.}\ 1997; Bahcall \& Fan 1998). These tests are interesting and important, though they are all subject to caveats. Many depend on whether the mass is distributed like the light, known as ``biasing," and many local measurements only indicate the amount of mass that is clustered on the scales of galaxies and clusters of galaxies. The concordance of many local, low-redshift tests suggests that the amount of mass that clusters with galaxies is significantly less than the critical value. Estimates of cosmological parameters that utilize the coordinate distance to sources at high redshift (category [2] defined above), such as tests involving angular size distance or luminosity distance, are fundamentally different from local, low-redshift tests. The value of $\Omega_m$ estimated through the coordinate distance, the angular size distance, or the luminosity distance, is truly the global value of $\Omega_m$. The test is independent of any biasing of matter relative to light, of how or whether the mass is clustered, of the nature of the dark matter (e.g. baryonic, cold dark matter, hot dark matter, etc.), and of the origin of fluctuations that lead to galaxy formation (e.g. cosmic strings, textures, adiabatic or isocurvature fluctuations, etc.). Two tests currently being used to determine global cosmological parameters through the coordinate distance to high-redshift sources (category [2] defined above) are the radio source method, and the supernova method. | The constraints on cosmological parameters obtained here arise primarily from the relative position of the data points as a function of redshift, and are not sensitive to any particular point, or to the normalization of the curves (which is left as a free parameter in the fits). For example, if the one low-redshift point near z = 0 is excluded, the results do not change (see Guerra \& Daly 1998). It is clear from figure 1a that the radio source model is working extremely well; the data fall right along the cosmological curves all the way from zero redshift to a redshift of about two. It is clear from figure 1b that, given this data, it is very unlikely that $\Omega_m=1$. The results presented here are very similar to those obtained by the supernovae groups. Any potential problems, such as dust extinction or evolution, are completely different for the two methods. The fact that they yield nearly identical results suggests that both are correct. In addition, the low value of $\Omega_m$ obtained using this method agrees with the low values indicated by local dynamical tests, which further suggests that $\Omega_m$ is low, and the universe will continue to expand forever. We are in the process of obtaining radio information on 67 additional radio galaxies with redshifts between zero and two. With these additional sources we will be able to constrain cosmological parameters to very high accuracy. | 98 | 3 | astro-ph9803265_arXiv.txt |
9803 | astro-ph9803053_arXiv.txt | We show that the kinematics of the shells seen around some elliptical galaxies provide a new, independent means for measuring the gravitational potentials of elliptical galaxies out to large radii. A numerical simulation of a set of shells formed in the merger between an elliptical and a smaller galaxy reveals that the shells have a characteristic observable kinematic structure, with the maximum line-of-sight velocity increasing linearly as one moves inward from a shell edge. A simple analytic calculation shows that this structure provides a direct measure of the gradient of the gravitational potential at the shell radius. In order to extract this information from attainable data, we have also derived the complete distribution of line-of-sight velocities for material within a shell; comparing the observed spectra of a shell to a stellar template convolved with this distribution will enable us to measure the gradient of the potential at this radius. Repeating the analysis for a whole series of nested shells in a galaxy allows the complete form of the gravitational potential as a function of radius to be mapped out. The requisite observations lie within reach of the up-coming generation of large telescopes. | The gravitational potentials of elliptical galaxies have proved difficult to derive, especially at large radii. In disk galaxies, both stars and gas have relatively simple orbit structures dominated by nearly circular orbits, and it is therefore straightforward to deduce the underlying gravitational potential (at least in the disk plane) from the observed kinematics in these systems. However, the corresponding orbital structure in elliptical galaxies is much more complicated, making the derivation of the gravitational potential from observed kinematics in such a system far from simple. In fact, basic kinematic observations of projected density and line-of-sight velocity dispersion are not sufficient to solve unambiguously for both the functional form of the gravitational potential and the distribution of stellar orbits in elliptical galaxies (Binney \& Mamon 1982). This ambiguity has recently been partially resolved by using the extra kinematic information that can be derived from the shapes of line profiles in high quality spectral data (e.g. Gerhard 1993, Carollo et al.\ 1995). However, these results are restricted to the inner few effective radii of the stellar light; it is hard to extend these methods to much larger radii because of the low surface brightnesses of galaxies further out. Independent mass measurements using gravitational lensing (e.g. Kochanek 1995) are restricted to still smaller radii. The only technique that reaches to large radii comes from mapping the distribution of hot gas around some ellipticals (Buote \& Canizares 1997). This method relies on the assumption of hydrostatic equilibrium in the analysis of X-ray intensity and temperature profiles, from which the gravitational potential may be deduced. The requisite data are hard to obtain with current X-ray telescopes, and the analysis is subject to possible systematic uncertainties if the hot gas is not in a single phase. Furthermore, many of the X-ray halos may be associated with the group hosting the elliptical rather than with the galaxy itself. It is therefore of interest to develop alternative probes of the mass distribution in the outer parts of elliptical galaxies. If we wish to measure the gravitational potential of an elliptical galaxy unambiguously using kinematic tracer particles, then we need a set of test bodies whose orbital structure is simple and observationally well-constrained. In this paper, we show that the large, regular systems of faint shells seen in some elliptical galaxies (Schweizer 1980, Malin \& Carter 1980) offer just such a tracer. These shells are believed to be the remnants of a small galaxy after a head-on collision with a larger system -- any other collision geometry does not result in extensive, nested shell systems. From the geometry of the merger, we know that all the stars in each shell must be on radial orbits with very similar energies. As we show below, this particularly-simple orbital structure means that the kinematics of the shells provide a useful new probe of the gravitational potential out to large radii in elliptical galaxies. | In this paper, we have set out to show how the kinematics of the shells produced when a small galaxy merges with a large system can be used to measure the underlying gravitational potential. As mentioned earlier in the paper, although the minor merger model is the most widely accepted interpretation of extensive shell systems, several other possible explanations have been advanced. Thus, not only would the detection of the characteristic kinematic structure illustrated in Fig.~\ref{fig-shellxw} allow us to measure the gravitational potential, but it would also confirm the origin of the shells themselves. An alternative method for using shells to constrain the form of the gravitational potential was originally proposed by Quinn (1984). He suggested that use be made of the distribution of shells with radius. Recall that, at any given time, the photometrically-observed shells are delineated by those stars that have completed integer number of radial oscillations. The ratio of the radii of successive shells therefore gives the oscillation periods at different energies, which is directly related to the form of the potential. The constraints on the potential obtained from such analyses are non-dimensional; they could, for example, provide information about the power-law index of the potential, but they cannot yield a mass normalization. By contrast, the kinematic method described here relies on measured velocities, which can be translated directly into the fully-dimensional gravitational potential. The shell counts have also been shown to be susceptible to uncertainties associated with dynamical friction acting on the shredding satellite, since different shells will have followed substantially different orbits, with different numbers of passages through the center of the host galaxy. Quinn's method also requires that we know how many radial oscillations the outermost shell has executed, which can only be inferred indirectly. The method discussed here, on the other hand, only uses information from each shell singly, and only uses the fact that phase wrapping is an efficient mechanism for concentrating stars of very nearly equal energies. Our analysis has assumed that the merger occurred on an exactly radial orbit. To test the importance of this assumption, we have repeated our simulations in the isochrone potential for satellite galaxies on slightly non-radial orbits. These calculations revealed that even in systems where the encounter is sufficently off-axis to produce shells that do not appear regular and aligned, the line profiles of individual shells are still accurately reproduced by the radial model calculations. Thus, the kinematics of an aligned shell system will almost certainly be well-modelled by the simple radial orbit approximation. Furthermore, altering the viewing angle of the shell system tends to diminish the shell visibility before altering the velocity structure, again suggesting that by selecting aligned, regular shell systems, we will preferentially find systems to which the above analysis is applicable. It should be noted, however, that the tests we have carried out have all been based on the isochrone potential, since the orbit calculations are computationally simple in a potential of this form. It would be interesting to carry out a more extensive numerical study of shells in other potentials in order to check that they, too, are insensitive to the assumption of radial orbits. Any flattening of the host galaxy's potential is also a possible problem. It may introduce orientation-dependent effects, as well as modify the orbit structure of the stars making up each shell. Simulations of the type described earlier in flattened potentials show that regular shell systems only occur for nearly radial encounters in spherical potentials: a degree of flattening as small as 5\% in the potential will cause significant beating between the motions in the various directions, and leads to shells with alternately large and small opening angles. The velocity structure of such shells is also rather disturbed, and it is not clear that the velocity profiles will provide much information about the overall potential. However, the detection of a regular uniform shell system provides us with strong {\it a priori} evidence that the potential must be very close to spherically-symmetric. Shells are photometrically faint structures, and so obtaining the kinematic measurements that we advocate here will not be simple. It is, however, noteworthy from a comparison of Figs.~\ref{fig-shellxy} and \ref{fig-shellxw} that the kinematic observation of these systems significantly enhances their contrast against any background. It is also worth noting that the brightnesses of observed shells do not decrease very rapidly with radius, and so, unlike other techniques for measuring gravitational potentials using stellar kinematics, the difficulty of applying this method does not increase prohibitively with radius. We have carried out signal-to-noise ratio calculations for some of the brighter shell galaxies such as NGC~3923, and have ascertained that data of the requisite quality could be obtained with a couple of nights' integration using a 4-metre telescope. Clearly, projects of this nature will prove quite feasible with the up-coming generation of large telescopes. | 98 | 3 | astro-ph9803053_arXiv.txt |
9803 | astro-ph9803047_arXiv.txt | For faint photometric surveys our ability to quantify the clustering of galaxies has depended on interpreting the angular correlation function as a function of the limiting magnitude of the data. Due to the broad redshift distribution of galaxies at faint magnitude limits the correlation signal has been extremely difficult to detect and interpret. We introduce a new technique for measuring the evolution of clustering. We utilize photometric redshifts, derived from multicolor surveys, to isolate redshift intervals and calculate the evolution of the amplitude of the angular 2-pt correlation function. Applying these techniques to the the Hubble Deep Field we find that the shape of the correlation function, at $z=1$, is consistent with a power law with a slope of $-0.8$. For $z>0.4$ the best fit to the data is given by a model of clustering evolution with a comoving \rn = 2.37 \Mpc\ and $\epsilon = -0.4^{+0.37}_{-0.65}$, consistent with published measures of the clustering evolution. To match the canonical value of \rn = 5.4 \Mpc, found for the clustering of local galaxies, requires a value of $\epsilon = 2.10^{+0.43}_{-0.64}$ (significantly more than linear evolution). The log likelihood of this latter fit is 4.15 less than that for the \rn = 2.37 \Mpc\ model. We, therefore, conclude that the parameterization of the clustering evolution of $(1+z)^{-(3+\epsilon)}$ is not a particularly good fit to the data. | The evolution of the clustering of galaxies as a function of redshift provides a sensitive probe of the underlying cosmology and theories of structure formation. In an ideal world we would measure the spatial correlation function of galaxies as a function of redshift and type and use this to compare with the predictions of different galaxy formation theories. Observationally, however, our ability to efficiently measure galaxy spectra falls rapidly as a function of limiting magnitude and consequently we are limited to deriving spatial statistics from small galaxy samples and at relatively bright magnitude limits (e.g.\ $I_{AB} < 22.5$, Le Fevre et al.\ 1996, Carlberg et al.\ 1997). To increase the size of the galaxy samples and thereby reduce the shot noise the standard approach has been to measure the angular correlation function, i.e.\ the projected spatial correlation function (Brainerd et al.\ 1996, Woods and Fahlman 1997). While this allows us to extend the measure of the clustering of galaxies to fainter magnitude limits ($R<29$, Villumsen et al.\ 1997) it has an associated limitation. For a given magnitude limit the amplitude of the angular correlation function is sensitive to the width of the galaxy redshift distribution, N(z). At faint magnitude limits N(z) is very broad and consequently the clustering signal is diluted due to the large number of randomly projected pairs. In this letter we introduce a new approach for quantifying the evolution of the angular correlation function; we apply photometric redshifts (Connolly et al.\ 1995, Lanzetta et al.\ 1996, Gwyn and Hartwick 1996, Sawicki et al.\ 1997) to isolate particular redshift intervals. In so doing we can remove much of the foreground and background contamination of galaxies and measure an amplified angular clustering. We discuss here the particular application of this technique to the Hubble Deep Field (HDF; Williams et al.\ 1996). | Photometric redshifts provide a simple statistical means of directly measuring the evolution of the clustering of galaxies. By isolating narrow intervals in redshift space we can reduce the number of randomly projected pairs and detect the clustering signal to high redshift and faint magnitude limits. Applying these techniques to the HDF we can characterize the evolution of the angular 2 pt correlation function out to $z=1.6$. For redshifts $0.4<z<1.6$ we find that the amplitude of the angular correlation function is best parameterized by a comoving \rn=2.37 \Mpc\ and $\epsilon = -0.4^{+0.37}_{-0.65}$. To match, however, the canonical local value for the clustering length, \rn=5.4 \Mpc, requires $\epsilon = 2.1^{+0.4}_{-0.6}$, significantly more than simple linear growth. It must be noted that while these results are in good agreement with those from published photometric and spectroscopic surveys (Le F\`{e}vre et al 1996, Hudon and Lilly 1996) there are two caveats that should be considered before applying them to constrain models of structure formation. The small angular extent of the HDF (at a redshift, $z=1$, the field-of-view of the HDF is approximately 0.9 \Mpc) means that fluctuations on scales larger than we probe will contribute to the variance of the measured clustering (Szapudi and Colombi 1996). Secondly, the requirement that the HDF be positioned such that it avoids bright galaxies ($I_{814}<20$) biases our clustering statistics by artificially suppressing the number of low redshift galaxies (a bias that will be present in most deep photometric surveys). Therefore, the clustering evolution in the HDF may not necessarily be representative of the general field population. Given this, there is enormous potential for the application of this technique to systematic wide angle multicolor surveys, such as the Sloan Digital Sky Survey, | 98 | 3 | astro-ph9803047_arXiv.txt |
9803 | astro-ph9803271_arXiv.txt | We argue that due to various restrictions cosmic strings and monopole-string networks are not likely to produce the observed flux of ultra-high energy cosmic rays (UHECR). Among the topological defects studied so far, the most promising UHECR sources are necklaces and monopolonia. Other viable sources which are similar to topological defects are relic superheavy particles. All these sources have an excess of pions (and thus photons) over nucleons at production. We demonstrate that in the case of necklaces the diffuse proton flux can be larger than photon flux, due to absorption of the latter on radiobackground, while monopolonia and relic particles are concentrated in the Galactic halo, and the photon flux dominates. Another signature of the latter sources is anisotropy imposed by asymmetric position of the sun in the Galactic halo. In all cases considered so far, including necklaces, photons must be present in ultra-high energy radiation observed from topological defects, and experimental discrimination between photon-induced and proton-induced extensive air showers can give a clue to the origin of ultra-high energy cosmic rays. | The observation of cosmic ray particles with energies higher than $10^{11}~GeV$ \cite{EHE} gives a serious challenge to the known mechanisms of acceleration. The shock acceleration in various astrophysical objects typically gives maximal energy of accelerated protons less than $(1-3)\cdot 10^{10}~GeV$ \cite{NMA} (see however \cite{Bier97}). The unipolar induction can provide the maximum energy $1\cdot 10^{11}~GeV$ only for extreme values of the parameters \cite{BBDGP}. Much attention has recently been given to acceleration by ultrarelativistic shocks \cite{Vie},\cite{Wax}. The particles here can gain a tremendous increase in energy, equal to $\Gamma^2$, at a single reflection, where $\Gamma$ is the Lorentz factor of the shock. However, it is known (see e.g. the simulation for pulsar relativistic wind in \cite{Hosh}) that particles entering the shock region are captured there or at least have a small probability to escape. Topological defects (TD) (for a review see \cite{Book}) can naturally produce particles of ultrahigh energies (UHE) well in excess of those observed in cosmic rays (CR). In most cases the problem with topological defects is not the maximum energy but the fluxes. However, in some cases the predicted fluxes are comparable with observations. Usually, UHE particles appear at the decays of superheavy (SH) particles produced by TD. (We shall refer to these SH particles as $X$-particles). Examples discussed in the literature include ejection of X-particles from superconducting strings, emission of X-particles from cusps or intersections of ``ordinary'' strings, and production of such particles in monopole-antimonopole annihilations. Metastable SH particles can also be relics of an earlier epoch, produced by a thermal or some other mechanism in the early Universe. A rather exceptional mechanism of UHE particle production is given by radiation of accelerated monopoles connected by strings. In this case a monopole can radiate gluons with very large Lorentz factors and with virtualities of the order of the monopole acceleration. A common signature of all extragalactic UHECR is the Greisen-Zatsepin-Kuzmin (GZK) cutoff \cite{GZK}. It reveals itself as a steepening of the spectrum of UHE protons and nuclei due to their interaction with microwave radiation. The steepening starts at $E \approx 3\cdot 10^{10}~GeV$. Apart from this steepening there is another signature of interaction of extragalactic CR with microwave radiation: a bump in the spectrum preceeding the cutoff. The bump is a consequence of the proton number conservation in the spectrum: protons loose energy and are accumulated before the cutoff. In this paper we will discuss the signatures of UHECR from TD distinguishing them from particles produced by astrophysical accelerators. We will confine ourselves here to the case of the conventional primary particles, protons and photons, and will not consider the other UHE signal carriers discussed in the literature such as neutrinos \cite{bz,bordes}, Lightest Supersymmetric Particles \cite{cfk,BeKa}, relativistic monopoles \cite{wk} and vortons \cite{bona}. Throughout the paper we shall use the following numerical values and abbreviations: the dimensionless Hubble constant $h=0.65$, the Cold Dark Matter density in terms of critical density $\Omega_{CDM}=0.2 h^2$, the size of Dark Matter halo $R_h=100~kpc$, UHECR - for Ultra High Energy Cosmic Rays and UHE - for Ultra High Energy, TD - for Topological Defect, SH - for Superheavy, CDM - for Cold Dark Matter, SUSY - for Supersymmetry, LLA - for Leading Logarithmic Approximation, AGN - for Active Galactic Nucleus, GC and AC - for Galactic Center and Anticenter, respectively. | We studied observational constraints on various TD models, as well as possible signatures of TD as sources of the observed UHE radiation ($E\geq 10^{10}~GeV$). The most stringent constraint is due to electromagnetic cascades. It depends on the energy spectrum of particles from decays and on astrophysical quantities which determine the development of the cascade (most notably on the flux of intergalactic infra-red/optical radiation and on intergalactic magnetic field ). The SUSY-QCD spectrum makes the cascade constraint weaker, because this spectrum predicts less higher energy particles and more low energy particles as compared with ordinary QCD spectrum. In case of very large $m_X$ it means that for a given UHECR flux, less energy is transferred to the e-m cascade radiation. There are considerable uncertainties in the extragalactic flux of infra-red radiation and extragalactic magnetic field. The conservative limit on the energy density of the cascade radiation imposed the latest EGRET data is $\omega_{cas} \approx 2\cdot 10^{-6}~eV/cm^3$. It could be $3 - 5\cdot 10^{-6}~eV/cm^3$ with astrophysical uncertainties mentioned above. The further progress in the study of origin of EGRET extragalactic flux and in calculation of SUSY-QCD spectrum, in the pessimistic case, can exclude such TD as e.g. necklaces as the sources of observed UHECR. Another important constraint arises from the fact that at ultra-high energies, the proton attenuation length $R_p(E)$ and the photon absorption length $R_{\gamma}(E)$ are both small compared to the Hubble radius. Models in which the typical distance between defects is $D>>R_p$ are disfavored. In such models, the observed spectrum would have an exponential cutoff, unless a source is accidentally close to the observer. In the latter case the flux would be strongly anisotropic. Finally, in many cases TD give UHECR fluxes lower than the observed ones. We showed here that this is the case for monopole-string networks. Superconducting and ordinary cosmic strings probably belong to this category as well, although some loopholes still remain to be closed. With all these constraints taken into account, it appears that only necklaces, monopolonium and relic SH particles survive as potential UHE sources. The most important observational signature of TD as sources of UHE CR is the presence of photon-induced EAS. For all known mechanisms of UHE particle production the pions (and thus photons) dominate over nucleons. At energies lower than $1\cdot 10^{12}~GeV$, protons have considerably larger attenuation length than photons and the observed proton flux can be dominant. Nevertheless, even in this case photons reach an observer from sources located inside the sphere of radius $R_\gamma (E)$ (assuming that $R_\gamma >D$). Unlike protons, photons propagate rectilinearly, indicating the direction to the sources. {\em Necklaces} with a large value of $r=m/\mu d > 10^7$ have a small separation $D <R_\gamma$. They are characterized by a small fraction, $R_{\gamma}/R_p$, of photon-induced EAS at energies $10^{10} - 10^{11}~GeV$. This fraction increases with energy and becomes considerable at the highest energies. For smaller values of $r \sim 10^4 - 10^6$, when the separation is larger than $R_{\gamma}$ but still smaller than $R_p$, most of UHE particles are expected to be protons (with a chance of incidental proximity of a source, seen as a direct gamma-ray source). Thus, in all cases necklaces are characterized by an excess of proton-induced showers. However, some fraction of photon-induced showers is always present, and it can be large at the highest energies. {\em Monopolonium}, decaying vortons {\em and SH relic particles} are characterized by an enhanced density in the Galactic halo. They give a photon-dominated flux without a GZK cutoff. Due to asymmetric position of the Earth in the Galaxy, this flux is anisotropic. The largest flux is expected from the direction of the Galactic Center, where the density of sources is the largest. Unfortunately, the Galactic Center is not seen by gigantic arrays , such as Akeno, Fly's Eye, Haverah Park, and Yakutsk array. However, these detectors can observe a minimum in the direction of the Galactic anticenter in particles with energies $E>1\cdot 10^{10}~GeV$, as compared with the direction perpendicular to the Galactic Plane. A flux from the Virgo cluster might be another signature of this model. The search for photon induced showers is not an easy experimental task. It is known (see e.g. Ref.\cite{AK}) that in the UHE photon-induced showers the muon content is very similar to that in proton-induced showers. However, some difference in the muon content between these two cases is expected and may be used to distinguish between them observationally. A detailed analysis would be needed to determine this difference. The Landau-Pomeranchuk-Migdal (LPM) effect \cite{LPM} and the absorption of photons in the geomagnetic field are two other important phenomena which affect the detection of UHE photons \cite{AK,Kasa}; (see \cite{ps} for a recent discussion). The LPM effect reduces the cross-sections of electromagnetic interactions at very high energies. However, if the primary photon approaches the Earth in a direction characterized by a large perpendicular component of the geomagnetic field, the photon is likely to decay into electron and positron \cite{AK,Kasa}. Each of them emits a synchrotron photon, and as a result a bunch of photons strikes the Earth atmosphere. The LPM effect, which strongly depends on energy, is thus suppressed. If on the other hand a photon moves along the magnetic field, it does not decay, and LPM effect makes shower development in the atmosphere very slow. At extremely high energies the maximum of the showers can be so close to the Earth surface that it becomes "unobservable" \cite{ps}. We suggest that for all energies above the GZK cutoff the showers be analyzed as candidates for being induced by UHE photons, with the probability of photon splitting in the geomagnetic field determined form the observed direction of propagation, and with the LPM effect taken into account. The search for photon-induced showers can be especially effective in the case of Fly's Eye detector which can measure the longitudinal development of EAS. The future Auger detector will have, probably, the highest potentiality to resolve this problem. | 98 | 3 | astro-ph9803271_arXiv.txt |
9803 | astro-ph9803101_arXiv.txt | We searched for cluster X-ray luminosity and radius evolution using our sample of 201 galaxy clusters detected in the 160~deg$^2$ survey with the \ROSAT\/ PSPC (Vikhlinin et al.\ 1998). With such a large area survey, it is possible, for the first time with \ROSAT\/, to test the evolution of luminous clusters, $L_x>3\times10^{44}\,$\ergpersec\ in the 0.5--2~keV band. We detect a factor of 3--4 deficit of such luminous clusters at $z>0.3$ compared to the present. The evolution is much weaker or absent at modestly lower luminosities, 1--$3\times10^{44}\,$\ergpersec. At still lower luminosities, we find no evolution from the analysis of the $\log N - \log S$ relation. The results in the two upper $L_x$ bins are in agreement with the {\em Einstein\/} EMSS evolution result (Gioia et al.\ 1990a, Henry et al.\ 1992) while being obtained using a completely independent cluster sample. The low-$L_x$ results are in agreement with other \ROSAT\/ surveys (e.g.\ Rosati et al.\ 1998, Jones et al.\ 1998). We also compare the distribution of core radii of nearby and distant ($z>0.4$) luminous (with equivalent temperatures 4--7~keV) clusters, and detect no evolution. The ratio of average core radius for $z\sim0.5$ and $z<0.1$ clusters is $0.9\pm0.1$, and the core radius distributions are remarkably similar. A decrease of cluster sizes incompatible with our data is predicted by self-similar evolution models for high-$\Omega$ universe. | The cluster evolution rate is a strong test of cosmological parameters (e.g., White \& Rees 1978, Kaiser 1986, Eke, Cole \& Frenk 1996). It is best to study evolution using X-ray selected samples of distant clusters which are much less affected by projection than the optically selected samples (van Haarlem et al.\ 1997). Of all the interesting cluster parameters such as mass, velocity dispersion, and temperature, the X-ray luminosity is the most accessible to measurements with present-day instruments, and most of the earlier studies were focused on evolution of the cluster X-ray luminosity function. A strong evolution of cluster luminosities at $z\sim0.1$ was reported from the EXOSAT survey (Edge et al.\ 1990), but was later disproved by the \ROSAT\/ All-Sky Survey (Ebeling et al.\ 1997). At higher redshifts, negative evolution of the cluster X-ray luminosity function was first reported by Gioia et al.\ (1990a) using the {\em Einstein}\/ Extended Medium Sensitivity Survey (EMSS; Gioia et al.\ 1990b, Stocke et al.\ 1991). Gioia et al.\ and later Henry et al.\ (1992) compared the cluster luminosity functions below and above $z=0.3$. They found that while the number of the low luminosity clusters does not evolve, there is a significant deficit of luminous, $L_x(0.3-3.5\mbox{~keV})>5\times10^{44}\,$\ergpersec, clusters at high redshift. This EMSS result was questioned recently. Nichol et al.\ (1997) reanalyzed the EMSS cluster sample using \ROSAT\/ X-ray and new optical observations and argued that the evolution reported in the original EMSS papers was not significant. Several groups pursued independent searches for distant clusters in archival \ROSAT\/ PSPC observations. Collins et al.\ (1997) found that the redshift distribution of 35 clusters detected in their 17~deg$^2$ survey is consistent with no evolution. This contradicted the earlier claim by Castander et al.\ (1995) of a strong evolution in a similar sample; however, the latter authors used an X-ray source detection algorithm not optimized for the cluster search. Jones et al.\ (1998) presented the $\log N - \log S$ relation for 46 clusters from their 16~deg$^2$ survey and found that this relation is consistent with no evolution of the $L_x<2\times10^{44}\,$\ergpersec\ (0.5--2~keV band) clusters. Rosati et al.\ (1998) derived cluster luminosity functions up to $z\sim 0.8$ from their sample of 70 clusters detected in a 33~deg$^2$ survey, and found no evolution at low luminosities, $L_x<3\times10^{44}\,$\ergpersec. However, none of these \ROSAT\/ surveys covers an area large enough to probe the evolution of the luminous clusters, and their no-evolution claims do not contradict the EMSS results. Our 160~deg$^2$ survey (Vikhlinin et al.\ 1998, hereafter Paper~I) is the first \ROSAT\/ survey comparable with the EMSS in sky coverage for distant clusters. We are able to test, and confirm, the EMSS evolution results even with the incomplete redshift data currently at hand. Eventually, when the spectroscopic work is complete, we will be able to characterize the luminosity evolution more accurately. In this Letter, we also show that the cluster X-ray core radii do not evolve between $z\sim0.5$ and now. Throughout the paper, we use definitions $f_{-14}$ and $L_{44}$ for flux and luminosity in the 0.5--2~keV energy band in units of $10^{-14}\,$\ergs\ and $10^{44}\,$\ergpersec, respectively. We also use $H_0=50$~km~s$^{-1}$~Mpc$^{-1}$ and $q_0=0.5$. | We present a first \ROSAT\/ analysis of the evolution of luminous, $L_x>3\times10^{44}\,$\ergpersec\ distant clusters. We find a significant, factor of 3--4, decrease in the number of such clusters at $z>0.3$, confirming the detection of evolution in the EMSS (Gioia et al.\ 1990a, Henry et al.\ 1992). At lower luminosities, 1--$3\times10^{44}\,$\ergpersec, the evolution is undetectable, with a decrease in number by a factor of only $1.3\pm0.2$. This is also consistent with the EMSS and other \ROSAT\/ surveys (e.g.\ Rosati et al.\ 1998). The absence of evolution of low luminosity clusters is also supported by the analysis of the $\log N - \log S$ distribution from which we find that the cluster volume emissivity, dominated by low-luminosity objects, does not evolve. The observed evolution can be reproduced by a model in which the characteristic luminosity decreases with redshift, but the comoving number density of clusters increases. Such models arise naturally in the hierarchical cluster formation scenario (e.g.\ Kaiser 1986). We compare the distribution of core-radii of distant, $z>0.4$ and nearby clusters. We find that the distribution of core radii at $z>0.4$ is very similar to that in nearby clusters; the average radius has changed at $z>0.4$ by a factor of only $0.9\pm0.1$. A stronger change is expected for hierarchical cluster formation in a flat universe (Kaiser 1986, Cen \& Ostriker 1994). We also note that the assumption of no evolution of cluster sizes has been essentially used in flux measurements and area calculations in several X-ray surveys (e.g.\ EMSS, Nichol et al. 1997), but is only verified here for the first time. \bigskip \centerline{\includegraphics[width=3.25in]{axdistr.ps}} \vskip -15pt \figcaption{Core-radius distribution for distant, $z>0.4$, clusters, derived from our survey (solid and dashed histogram for $q_0=0.5$ and $q_0=0$, respectively). The shaded histogram shows the core-radius distribution for nearby luminous clusters from Jones \& Forman (1998). The angular resolution limit of our survey (15\arcsec) corresponds to 120~kpc at the median redshift of distant clusters, well below the peak of the distribution. \label{fig:axdistr}} \medskip | 98 | 3 | astro-ph9803101_arXiv.txt |
9803 | astro-ph9803337_arXiv.txt | If the theoretical relationship between white dwarf mass and orbital period for wide-orbit binary radio pulsars is assumed to be correct, then the neutron star mass of PSR J2019+2425 is shown to be $\sim 1.20 M_{\odot}$. Hence the mass of the neutron star in this system prior to the mass transfer phase is expected to have been $< 1.1 M_{\odot}$. Alternatively this system descends from the accretion induced collapse (AIC) of a massive white dwarf.\\ We estimate the magnetic inclination angles of all the observed wide-orbit low-mass binary pulsars in the Galactic disk using the core-mass period relation and assuming that the spin axis of an accreting neutron star aligns with the orbital angular momentum vector in the recycling process of the pulsar. The large estimated magnetic inclination angle of PSR J2019+2425, in combination with its old age, gives for this system evidence against alignment of the magnetic field axis with the rotational spin axis. However, in the majority of the similar systems the distribution of magnetic inclination angles is concentrated toward low values (if the core-mass period relation is correct) and suggests that alignment has taken place. | 98 | 3 | astro-ph9803337_arXiv.txt |
||
9803 | astro-ph9803294_arXiv.txt | s{ Samples of high-redshift galaxies are easy to select in the millimetre/submillimetre (mm/submm) waveband using sensitive telescopes, because their flux density--redshift relations are expected to be flat, and so the selection function is almost redshift-independent at redshifts greater than 0.5. Source counts are expected to be very steep in the mm/submm waveband, and so the magnification bias due to gravitational lensing is expected to be very large, both for lensing by field galaxies and for lensing by clusters. Recent submm-wave observations of lensed images in clusters have constrained the submm-wave counts directly for the first time. In the next ten years our knowledge of galaxy evolution in this waveband will be greatly enhanced by the commissioning of sensitive new instruments and telescopes, including the CMBR imaging space mission {\em Planck Surveyor}. This paper highlights the important features of gravitational lensing in the submm waveband and discusses the excellent prospects for lens searches using these forthcoming facilities. } | \begin{enumerate} \item The surface density of distant galaxies in the submm waveband, and therefore the expected surface density of gravitational lenses and the effects of source confusion in future observations, is now known with reasonable accuracy. \item The {\em Planck Surveyor} survey and surveys using other forthcoming mm/submm-wave telescopes will produce catalogues of distant sources that will be of great interest for studies of galaxy evolution. Lensed galaxies and quasars will be detected with an efficiency of up to 10\% in these surveys, and a sample of order 1000 lenses could be compiled. \end{enumerate} | 98 | 3 | astro-ph9803294_arXiv.txt |
|
9803 | hep-ph9803418_arXiv.txt | The neutrino capture rate measured by the Russian-American Gallium Experiment is well below that predicted by solar models. To check the response of this experiment to low-energy neutrinos, a 517 kCi source of \nuc{51}{Cr} was produced by irradiating 512.7~g of 92.4\%-enriched \nuc{50}{Cr} in a high-flux fast neutron reactor. This source, which mainly emits monoenergetic 747-keV neutrinos, was placed at the center of a 13.1 tonne target of liquid gallium and the cross section for the production of \nuc{71}{Ge} by the inverse beta decay reaction $\mnuc{71}{Ga}(\nu_e,e^-)\mnuc{71}{Ge}$ was measured to be {[5.55 \+/-~0.60~(stat) \+/-~0.32~(syst)]} \E{-45} cm$^2$. The ratio of this cross section to the theoretical cross section of Bahcall for this reaction is 0.95 \+/- 0.12 (expt) $^{+0.035}_{-0.027}$ (theor) and to the cross section of Haxton is 0.87 \+/- 0.11 (expt) \+/- 0.09 (theor). This good agreement between prediction and observation implies that the overall experimental efficiency is correctly determined and provides considerable evidence for the reliability of the solar neutrino measurement. | Gallium experiments are uniquely able to measure the principal component of the solar neutrino spectrum. This is because the low threshold of 233 keV \cite{Audi and Wapstra 95} for inverse beta decay on the 40\% abundant isotope \nuc{71}{Ga} is well below the end point energy of the neutrinos from proton-proton fusion, which are predicted by standard solar models to be about 90\% of the total flux. The Russian-American Gallium Experiment (SAGE) has been measuring the capture rate of solar neutrinos with a target of gallium metal in the liquid state since January 1990. The measured capture rate \cite{Abdurashitov et al. 94,Gavrin 98} is 67~\+/-7~(stat)~$^{+5}_{-6}$ (syst) SNU\footnote{1 SNU corresponds to one neutrino capture per second in a target that contains 10$^{36}$ atoms of the neutrino absorbing isotope.}, a value that is well below solar model predictions of 137 $^{+8}_{-7}$ SNU \cite{Bahcall and Pinsonneault and Wasserburg 95} and 125~\+/- 5 SNU \cite{Turck-Chieze and Lopes 93}. In addition, the GALLEX Collaboration, which has been measuring the solar neutrino capture rate with an aqueous GaCl$_3$ target since 1991, observes a rate of 70~\+/- 7~$^{+4}_{-5}$ SNU \cite{Hampel et al. 96}. The other two operating solar neutrino experiments, the chlorine experiment \cite{Cleveland et al. 98} and the Kamiokande experiment \cite{Suzuki et al. 95}, have significantly higher-energy thresholds, and thus are not able to see the neutrinos from $pp$ fusion. When the results of these four solar neutrino experiments are considered together, a contradiction arises which cannot be accommodated by current solar models, but which can be explained if one assumes that neutrinos can transform from one species to another \cite{Bahcall 94,Berezinsky 94,Parke 95,Hata et al. 94,Castellani et al. 94,Bahcall et al. 95,Heeger and Robertson 96}. The gallium experiment, in common with other radiochemical solar neutrino experiments, relies on the ability to extract, purify, and count, all with well known efficiencies, a few atoms of a radioactive element that were produced by neutrino interactions inside many tonnes of the target material. In the case of 60 tonnes of Ga, this represents the removal of a few tens of atoms of \nuc{71}{Ge} from $5 \times 10^{29}$ atoms of Ga. To measure the efficiency of extraction, about 700 $\mu$g of stable Ge carrier is added to the Ga at the beginning of each exposure, but even after this addition, the separation factor of Ge from Ga is still 1 atom in 10$^{11}$. This impressively stringent requirement raises legitimate questions about how well the many efficiencies that are factored into the final result are known. It has been understood since the outset that a rigorous check of the entire operation of the detector (i.e., the chemical extraction efficiency, the counting efficiency, and the analysis technique) would be made if it is exposed to a known flux of low-energy neutrinos. In addition to verifying the operation of the detector, such a test also eliminates any significant concerns regarding the possibility that atoms of \nuc{71}{Ge} produced by inverse beta decay may be chemically bound to the gallium (so-called ``hot atom chemistry'') in a manner that yields a different extraction efficiency than that of the natural Ge carrier. In other words, it tests a fundamental assumption in radiochemical experiments that the extraction efficiency of atoms produced by neutrino interactions is the same as that of carrier atoms. This article describes such a test, in which a portion of the SAGE gallium target was exposed to a known flux of \nuc{51}{Cr} neutrinos and the production rate of \nuc{71}{Ge} was measured. Similar tests have also been made by GALLEX \cite{Hampel et al. 98}. Although a direct test with a well-characterized neutrino source lends significant credibility to the radiochemical technique, we note that numerous investigations have been undertaken during the SAGE experiment to ensure that the various efficiencies are as quoted \cite{Abdurashitov et al. 94}. The extraction efficiency has been determined by a variety of chemical and volumetric measurements that rely on the introduction and subsequent extraction of a known amount of the stable Ge carrier. A test was also carried out in which Ge carrier doped with a known number of \nuc{71}{Ge} atoms was added to 7 tonnes of Ga. Three standard extractions were performed, and it was demonstrated that the extraction efficiencies of the carrier and \nuc{71}{Ge} follow each other very closely. Another experiment was performed to specifically test the possibility that atomic excitations might tie up \nuc{71}{Ge} in a chemical form from which it would not be efficiently extracted. There is a concern that this might occur in liquid gallium because the metastable Ga$_2$ molecule exists with a binding energy of \about1.6 eV. In this experiment the radioactive isotopes \nuc{70}{Ga} and \nuc{72}{Ga} were produced in liquid gallium by neutron irradiation. These isotopes quickly beta decay to \nuc{70}{Ge} and \nuc{72}{Ge}. The Ge isotopes were extracted from the Ga using our standard procedure and their number was measured by mass spectrometry. The results were consistent with the number expected to be produced based on the known neutron flux and capture cross section, thus suggesting that chemical traps are not present. This experiment is not conclusive, however, because the maximal energy imparted to the \nuc{70}{Ge} nucleus following beta decay of \nuc{70}{Ga} is 32 eV, somewhat higher than the maximal energy of 20 eV received by the \nuc{71}{Ge} nucleus following capture of a 747-keV neutrino from \nuc{51}{Cr} decay (and considerably higher than the maximum nuclear recoil energy of 6.1 eV after capture of a 420-keV neutrino from proton-proton fusion). Further evidence that the extraction efficiency was well understood came from monitoring the initial removal from the Ga of cosmogenically produced \nuc{68}{Ge}. This nuclide was generated in the Ga as it resided on the surface exposed to cosmic rays. When the Ga was brought underground, the reduction in the \nuc{68}{Ge} content in the initial extractions was the same as for the Ge carrier. These numerous checks and auxiliary measurements have been a source of confidence in our methodology, yet it is clear that a test with an artificial neutrino source of known activity provides the most compelling validation of radiochemical procedures. This article is an elaboration of work that previously appeared in Ref.~\cite{Abdurashitov et al. 96}. The experimental changes since the previous Letter are some minor refinements in the selection of candidate \nuc{71}{Ge} events and in the treatment of systematic errors; recent cross section calculations are also included. The central experimental result given here is almost identical to what was reported earlier. | The primary motivation for the \nuc{51}{Cr} source experiment was to determine if there is any unexpected problem in either the chemistry of extraction or the counting of \nuc{71}{Ge}, i.e., to see if there is some unknown systematic error in one or both of the efficiency factors in $\epsilon$, the product of extraction and counting efficiencies. If some such systematic error were to exist, then the value of $\epsilon$ that we have used in the preceding will be in error by the factor $E$, defined as $E \equiv \epsilon_{\text{true}}/ \epsilon_{\text{measured}}$. Since the cross section is inversely proportional to $\epsilon$, this hypothetical error is equivalent to the cross section ratio, $E = \sigma_{\text{measured}}/\sigma_{\text{true}}$. An experimental value for $E$ can be set from our measured cross section, Eq.~(\ref{cross section result}), if one assumes that the true cross section is equivalent to the theoretically calculated cross section. Then $E \approx R \equiv \sigma_{\text{measured}}/\sigma_{\text{theoretical}}$. Neutrino capture cross sections averaged over the four neutrino lines of \nuc{51}{Cr} have been calculated by Bahcall \cite{Bahcall 97} and by Haxton \cite{Haxton 98}. Bahcall, assuming that the strength of the two excited states in \nuc{71}{Ge} that can be reached by \nuc{51}{Cr} neutrinos is accurately determined by forward-angle $(p,n)$ scattering, gives a result of 5.81 (1.0 $^{+0.036}_{-0.028})$ \E{-45} cm$^2$. The upper limit for the uncertainty was set by assuming that the excited state strength could be in error by as much as a factor of 2; minor contributions to the uncertainty arise from forbidden corrections, the \nuc{71}{Ge} lifetime, and the threshold energy. An independent consideration of the contribution of excited states has been made by Hata and Haxton \cite{Hata and Haxton 95} and very recently by Haxton \cite{Haxton 98}. They argue that, because of destructive interference between weak spin and strong spin-tensor amplitudes in \nuc{71}{Ge}, the strengths determined from $(p,n)$ reactions are, for some nuclear levels, poor guides to the true weak interaction strength. In particular, Haxton finds the weak interaction strength of the $(5/2)^-$ level in \nuc{71}{Ge} at an excitation energy of 175 keV to be much greater than the value that is measured by the $(p,n)$ scattering reaction, and calculates a total \nuc{51}{Cr} cross section of (6.39~\+/- 0.68) \E{-45} cm$^2$. This cross section was deduced from the measured $(p,n)$ cross sections for the two excited states, and uses a large-basis shell model calculation to correct for the presence of spin-tensor contributions. Since not all known theoretical uncertainties were included, the stated error here is a lower bound. Combining our statistical and systematic uncertainties for the cross section in quadrature into an experimental uncertainty, we can thus give estimates for $E$: \begin{eqnarray} E & \approx & R \equiv \frac{\sigma_{\text{measured}}} {\sigma_{\text{theoretical}}} \\ & = & \left\{ \begin{array}{l l} 0.95 \pm 0.12 \text{ (expt)}\ ^{+0.035}_{-0.027} \text{ (theor)} & \text{ (Bahcall)}, \\ 0.87 \pm 0.11 \text{ (expt)}\ \pm 0.09 \text{ (theor)} & \text{ (Haxton)}. \end{array} \right. \nonumber \end{eqnarray} \noindent With either of these theoretical cross sections, $R$ is consistent with unity, which implies that the total efficiency of the SAGE experiment to the neutrinos from \nuc{51}{Cr} is close to 100\%. The measurement reported here should not be interpreted as a direct calibration of the SAGE detector for solar neutrinos. This is because the \nuc{51}{Cr} neutrino spectrum differs from the solar spectrum, there is a 10\%--15\% uncertainty in the theoretical value for the \nuc{51}{Cr} cross section, and the total experimental efficiency for each solar neutrino measurement is known to a higher precision than the 12\% experimental uncertainty obtained with the \nuc{51}{Cr} source. As a result, the solar neutrino measurements reported by SAGE should not be scaled by the factor $E$. Rather, we consider the Cr experiment as a test of the experimental procedures, and conclude that it has demonstrated {\it with neutrinos} that there is no unknown systematic uncertainty at the 10\%--15\% level. The neutrino spectrum from \nuc{51}{Cr} is very similar to that of \nuc{7}{Be}, but at slightly lower energy. Since the response of \nuc{71}{Ga} to \nuc{7}{Be} neutrinos is governed by the same transitions that are involved in the \nuc{51}{Cr} source experiment, we can definitely claim that, if the interaction strength derived from the \nuc{51}{Cr} experiment is used in the analysis of the solar neutrino results, then the capture rate measured by SAGE includes the full contribution of neutrinos from \nuc{7}{Be}. This observation holds independent of the value of $E$ or of cross section uncertainties. This demonstration is of considerable importance because a large suppression of the \nuc{7}{Be} neutrino flux from the sun is one consequence of the combined analysis of the four operating solar neutrino experiments \cite{Bludman et al. 93,Akhmedov et al. 95}. GALLEX has completed two \nuc{51}{Cr} measurements whose combined result, using the cross section of Bahcall \cite{Bahcall 97}, can be expressed as $R$ = 0.93~\+/- 0.08 \cite{Hampel et al. 98}, where the uncertainty in the theoretical cross section has been neglected. Both SAGE and GALLEX, which employ very different chemistries, give similar results for the solar neutrino capture rate and have tested their efficiencies with neutrino source experiments. The solar neutrino capture rate measured in Ga is in striking disagreement with standard solar model predictions and there is considerable evidence that this disagreement is not an experimental artifact. | 98 | 3 | hep-ph9803418_arXiv.txt |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.