A. Galaxies, Quasars and Cosmology

N. Bahcall, in collaboration with S.P. Oh (graduate student), determined the peculiar velocity distribution function of clusters of galaxies using an accurate sample of cluster velocities (from Giovanelli and Haynes, 1996) based on Tully-Fisher distances of Sc galaxies. The observed velocity function does not exhibit a tail of high peculiar motion clusters, in contrast with previous samples with considerably larger velocity uncertainties. The current results indicate a low probability of of finding clusters with one-dimensional peculiar motions greater than . The root-mean-square cluster peculiar velocity is .

The observed cluster velocity distribution function was compared with expectations from different cosmological models. The absence of a high-velocity tail in the observed function is found to be most consistent with a low mass-density CDM model and is inconsistent at the level with CDM and HDM models. The rms one-dimensional cluster peculiar velocities in these models correspond, respectively, to 314, 516, and 632 km s (when convolved with the observational uncertainties). Comparison with the observed rms cluster velocity of 293 28 km s further supports the low-density CDM model.

N. Bahcall, R. Cen, L. Lubin (graduate student), and J.P. Ostriker compared observations of the baryon fraction and the velocity-temperature relation in clusters of galaxies with expectations from cosmological models using large-scale hydrodynamic simulations. Two cosmological models were investigated: Standard () and flat low-density ) CDM models, normalized to the COBE background fluctuations. The observed properties of clusters include the velocity dispersion versus temperature relation, the gas mass versus total mass relation, and the gas mass fraction versus velocity dispersion relation. The results show that while both cosmological models reproduce well the shape of these observed functions, only low-density CDM can reproduce the observed amplitudes. The models show that , as expected for approximate hydrostatic equilibrium with the cluster potential, and that the ratio of gas to total mass in clusters is approximately constant for both models. The amplitude of the relations, however, differs significantly between the two models. The low-density CDM model reproduces well the average observed relation for clusters of , while CDM yields a gas mass that is times too low . Both gas and total mass are measured within a fiducial radius of 1.5hMpc. The cluster gas mass fraction reflects approximately the baryon fraction in the models, . An model produces too few baryons in clusters compared with observations. Scaling the results as a function of , the authors find that a low-density CDM model, with , best reproduces the observed mean baryon fraction in clusters.

N. Bahcall, in collaboration with L. Lubin and V. Dorman (graduate students), used optical and X-ray mass determination of galaxies, groups and clusters of galaxies to investigate the amount and the location of dark matter. The results suggest that most of the dark matter may reside in very large halos around galaxies, typically extending to 200 kpc for bright galaxies. They show that the mass-to-light ratio of galaxy systems does not increase significantly with linear scale beyond the very large halos suggested for individual galaxies. Rather, the total mass of large scale systems such as groups and rich clusters of galaxies, even superclusters, can on average be accounted for by the total mass of their member galaxies, including their large halos (which may be stripped off in the dense cluster environment but still remain in the clusters) plus the mass of the hot intracluster gas. This conclusion also suggests that we may live in a low-density universe with .

N. Bahcall reviewed some of the unsolved problems in the study of large-scale structure of the universe and summarized goals for their resolution. Bahcall also reviewed the topics of clusters of galaxies, superclusters, and voids, and dark-matter in the universe, including both observations as well as cosmological implications.

R. Cen and J.P. Ostriker, in collaboration with H. Kang (Korea) and D. Ryu (Korea), present a new treatment of two popular models for the growth of structure, examining the X-ray emission from hot gas with allowance for spectral line emission from various atomic species, primarily ``metals''. The X-ray emission from the bright cluster sources is not significantly changed from prior work and shows the CDM model (LCDM) to be consistent but the standard, COBE normalized model (SCDM) to be inconsistent with existing observations -- even after allowance for the still considerable numerical modeling uncertainties. They find one important new result: radiation in the softer 0.5-1.0keV band is predominantly emitted by gas far from cluster centers (hence ``background''). This background emission dominates over the cluster emission below 1keV and observations of it should show clear spectral signatures indicating its origin. In particular the ``iron blend'' should be seen prominently in this spectral bin from cosmic background hot gas at high galactic latitudes and should show shadowing against the SMC indicating its extragalactic origin. Certain OVII lines also provide a signature of this gas which emits a spectrum characteristic of K gas. Recent ASCA observations of the X-ray background tentatively indicate the presence of component with the predicted spectral features.

R. Cen and J.P. Ostriker have computed a variety of cosmological models into the extreme nonlinear phase to enable comparisons with observations, including a current state-of-the-art treatment of hydrodynamical processes, heating and cooling. The results for such a suite of currently interesting models are summarized and compared. All models have a mean (z=0) temperature of - K, set essentially by photoheating processes. Most gas is in one of two components: either at the photoheating floor of K and primarily in low density regions or else shock-heated to - K and in regions of moderate overdensity (in caustics and near groups and clusters). It will be a major observational challenge to observationally detect this second, abundant component as it is neither an efficient radiator nor absorber. About 2% to 10% of the baryons cool and collapse into galaxies forming on caustics and migrating to clusters. About 1%-2% of baryons are in the very hot X-ray-emitting gas near cluster cores, in good agreement with observations. These correspondences between the simulations and the real world imply that there is some significant truth to the underlying standard scenarios for the growth of structure. The differences among model predictions may point out the path to the correct model. For COBE normalized models the most relevant differences concern the epoch of structure formation. In the open variants having , with or without a cosmological constant, structure formation on galactic scales is well advanced at redshift z=5, and reionization occurs early. But if observations require models for which most galaxy formation occurs more recently than z=2, then the flat models are to be preferred. The velocity dispersion on the Mpc scale also provides a strong discriminant with, as expected, the models giving a much higher (perhaps too high) value for that statistic.

R. Cen, J.R. Gott, and J.P. Ostriker studied the topology of large scale structure as a function of galaxy type using the genus statistic. In hydrodynamical cosmological CDM simulations, galaxies form on caustic surfaces (Zeldovich pancakes) and then slowly drain onto filaments and clusters. The earliest forming galaxies in the simulations (defined as ``ellipticals'') are thus seen at the present epoch preferentially in clusters (tending toward a meatball topology), while the latest forming galaxies (defined as ``spiral'') are seen currently in a spongelike topology. The topology is measured by the genus (= number of ``donut'' holes -- number of isolated regions) of the smoothed density-contour surfaces. The measured genus curve for all galaxies as a function of density obeys approximately the theoretical curve expected for random-phase initial conditions, but the early forming elliptical galaxies show a shift toward a meatball topology relative to the late forming spirals. Simulations using standard biasing schemes fail to show such an effect. Large observational samples separated by galaxy type could be used to test for this effect.

R. Cen, J.P. Ostriker and F.J. Summers, in collaboration with T. Padmanabhan (IUCAA, Pune), studied the nonlinear clustering of dark matter particles in an expanding universe using N-body simulations. One can gain some insight into this complex problem if simple relations between physical quantities in the linear and nonlinear regimes can be extracted from the results of N-body simulations. They investigate the relation between the mean relative pair velocities and the mean correlation function and other closely related issues in detail for the case of six different power spectra: power laws with spectral indexes n=-2,-1, cold dark matter (CDM), and hot dark matter models with density parameter ; CDM including a cosmological constant with , ; and an n=-1 model with . They find that: (i) Power law spectra lead to self-similar evolution in an universe. (ii) Stable clustering does not hold in an universe to the extent our simulations can ascertain. (iii) Stable clustering is a better approximation in the case of an universe in which structureformation freezes out at some low redshift. (iv) The relation between dimensionless pair velocity and the mean correlation function, , is only approximately independent of the shape of the power spectrum. At the nonlinear end, the asymptotic value of the dimensionless pair velocity decreases with increasing small scale power, because the stable clustering assumption is not universally true. (v) The relation between the evolved and the linear regime is also not universal but shows a weak spectrum dependence. Simple theoretical arguments for these conclusions are presented.

R. Cen and J.P. Ostriker, in collaboration with J. Wambsganss (Potsdam) described in detail a new method to trace light rays through an essentially three dimensional mass distribution up to high redshift. As an example, the method is applied to a standard cold dark matter universe. A variety of results are obtained, some of them statistical in nature, others from rather detailed case studies of individual ``lines of sight''. Among the former are the frequency of multiply imaged quasars, the distribution of separation of the multiple quasars, and the redshift distribution of lenses, all as a function of quasar redshift. Various effects are considered, ranging from very weak lensing up to highly magnified multiple images of high redshift objects. Applied to extended sources, i.e., galaxies, this ranges from slight deformations of the shapes, only measurable in a big ensemble, through tangentially aligned arclets up to giant luminous arcs. The weak coherent shear fields produced by lensing of large scale structure can be studied in directions that are devoid of large mass concentrations as well as the strong lensing around massive clusters of galaxies. Gravitational lensing directly measures mass density fluctuations along the line of sight to very distant objects. No assumptions need to be made concerning bias, the ratio of fluctuations in galaxy density to mass density. Hence lensing is a good tool to study the universe at medium and high redshifts. Cosmological models -- normalized to the universe at redshift zero -- differ considerably in their predictions for the mass distributions at these distance scales. Therefore lensing is a powerful tool to distinguish between various cosmological models. Our ultimate goal is to apply this method to a number of cosmogonic models in order to study their gravitational lensing effects and be able to eliminate some models whose properties are very different from the properties of the observed universe.

R. Cen, J.P. Ostriker, G. Xu (graduate student), and J. Wambsganss (Potsdam), examined the effects of weak gravitational lensing by large-scale structure on the determination of the cosmological deceleration parameter . They found that for true standard candles the lensing induced dispersions of 0.04 and 0.02 mag at redshift z=1 and z=0.5, respectively, in a COBE-normalized cold dark matter universe with km/s/Mpc and . It is shown that one would observe and (the error bars are limits) with standard candles with zero intrinsic dispersion at redshift z=1 and z=0.5, respectively, compared to the truth of =-0.40 in this case, i.e., a 10% error in will be made. A standard COBE normalized CDM model would produce three times as much variance and a mixed (hot and cold) dark matter model would lead to an intermediate result. One unique signature of this dispersion effect is its non-Gaussianity. Although the lensing-induced dispersion at lower redshift is still significantly smaller than the currently best observed (total) dispersion of 0.12 mag in a sample of type Ia supernovae, selected with the multicolor light curve shape method, it becomes significant at higher redshift. They show that there is an optimal redshift, in the range depending on the amplitude of the intrinsic dispersion of the standard candles, at which can be most accurately determined.

R. Cen and J.P. Ostriker, in collaboration with G.L. Bryan (NCSA), M.L. Norman (NCSA), and J.M. Stone (U.Md), described a hybrid scheme for cosmological simulations that incorporates a Lagrangian particle-mesh (PM) algorithm to follow the collisionless matter with the higher order accurate piecewise parabolic method (PPM) to solve the equations of gas dynamics. Both components interact through the gravitational potential, which requires the solution of Poisson's equation, here done by Fourier transforms. Due to the vast range of conditions that occur in cosmological flows (pressure difference of up to fourteen orders of magnitude), a number of additions and modifications to PPM were required to produce accurate results. These are described, as are a suite of cosmological tests.

R. Cen and R.A. Simcoe (undergraduate), performed a detailed analysis of the Lyman- clouds produced by cosmological hydrodynamic simulations of a spatially flat cold dark matter universe with a non-zero cosmological constant. They find a very wide variety of structures, ranging from roundish high density regions with , to filamentary and sheet-like structures with column densities below . The most common shape of the Ly clouds found in the simulation resembles a cigar squashed in the longitudinal direction. Furthermore, these Ly clouds range in size from several kiloparsecs to about a hundred kiloparsecs, indicating that if simple models with a single population of uniformly sized spheres (or other shapes) fit observations, this is only by coincidence. They showed that the method of inferring the sizes of Ly clouds using observations of double quasar sightlines is only meaningful (in terms of setting lower limits on cloud sizes) when the sightline separations are small ). Finally, they conjectured that high column density Ly clouds (cm) may be progenitors of faint blue galaxies at lower redshift, because the correlation length of these Ly clouds (extrapolated to lower redshift) resembles that of the observed faint blue galaxies, and their masses are close to those of starburst dwarf galaxies proposed by Babul & Rees.

R. Cen, J.P. Ostriker, J. Miralda-Escudé (IAS), and M. Rauch (Caltech), used an Eulerian hydrodynamic cosmological simulation to model the Ly forest in a spatially flat, COBE-normalized, cold dark matter model with . They found that the intergalactic, photoionized gas collapses into sheet-like and filamentary structures with HI having characteristics similar to the observed Ly forest. A typical filament is Mpc long with thickness kpc (in proper units), and baryonic mass . (In comparison the cell size is (2.5,9)hkpc in the two simulations.) The gas temperature is in the range K and increasing with time as structures with larger velocities collapse gravitationally. The predicted distributions of column densities, -parameters and equivalent widths of the Ly forest clouds agree reasonably with observations, and their evolution is consistent with the observed evolution, if the ionizing background has an approximately constant intensity between z=2 and z=4. A new method of identifying absorption lines as contiguous regions in the spectrum below a fixed flux threshold is suggested given that the Ly spectra arise from a continuous density field of neutral hydrogen rather than discrete clouds. They also predict the distribution of transmitted flux and its correlation along a spectrum and on parallel spectra, and the He II flux decrement as a function of redshift. A correlation length of kpc perpendicular to the line of sight is predicted for features in the Ly forest.

In order to reproduce the observed number of lines and average flux transmission, the baryon content of the clouds may need to be significantly higher than in previous models because of the predicted low densities and large volume-filling factors. If the background intensity is at least that predicted from the observed quasars, needs to be as high as , higher than expected by light element nucleosynthesis; the model also predicts that most of baryons at z > 2 are in Ly clouds, and that the rate at which the baryons move to more overdense regions is slow. A large fraction of the baryons which are not observed at present in galaxies might be intergalactic gas in the currently collapsing structures, with K.

R. Cen, using a large set of N-body simulations occupying a large volume in the four dimensional phase space (), showed that the abundance of rich clusters of galaxies can be described as a smooth analytic elementary function of one parameter, , which in turn depends on the four parameters in a very simple way. This relation enables us to compute the abundance of rich clusters of galaxies at any redshift for any cosmological model analytically, without resorting to expensive N-body calculations.

Two implications are worth stressing. First, it seems that a tilt of the spectrum from the Harrison-Zeldovich value of unity is required in order for CDM-like models to fit both COBE and galaxy cluster observations. Second, the evolution of rich clusters of galaxies will probably provide the single most strong discriminant of . Normalizing models to the present day rich cluster abundance, it is predicted that there should exist (0.004, 38, 4404) clusters with richness two and above at redshift two in three model universes with ()=(1.0, 0.0), (0.3,0.7), (0.3,0.0). ROSAT and future X-ray missions as well as large redshift surveys such as the Sloan Digital Sky Survey should provide a test.

N. Gnedin, together with with E. Bertschinger (MIT), worked on constructing a new self-gravitating hydrodynamic code. The project was motivated by the extensive study of the SLH cosmological hydrodynamic code that was developed in Gnedin's thesis. Gnedin and Bertschinger showed that the Moving Mesh Gravity solver, previously used in the SLH code, had generic errors that could negate the results of a simulation. Some of those errors but not all, were identified and cured. The gravity solver in the SLH code was then replaced with the well tested P3M solver. While incorporating the P3M solver in LH code, it was found that, in order for a self-gravitating hydrodynamic code to be strictly energy conserving, a special ``Consistency Condition'' ought to be satisfied; a new SLH-P3M code was used to demonstrate the effect of including/neglecting the Consistency Condition and also pointed out that most of existing cosmological hydrodynamic codes satisfied that condition.

J.P. Ostriker and N. Gnedin (MIT) further improved the SLH code by including new physical effects that had not been included into numerical simulations before, namely: self-shielding of the intergalactic gas from the radiation background, time-dependent ionization evolution of the intergalactic plasma, detailed non-equilibrium chemistry of molecular hydrogen, and approximate corrections for the finite resolution of a simulation. All these pieces of physics are required in order to simulate the reionization of the pregalactic gas and formation of Lyman- systems. The work is still in progress: most of the new physics is now incorporated in the code and tested. Large state-of-the-art simulations are planned. The simulations will include 4 million particles and will achieve a dynamical range of , which is an unprecedented resolution for a hydrodynamic simulation. It is planned to include spatially distributed sources of ionizing radiation, to complete the treatment of radiative transfer.

Gnedin, with J.P. Ostriker and J. Miralda-Escudé (IAS) initiated a project to carry out simple physical modeling of Lyman-alpha systems with the ultimate goal to understand all major physical effects that play roles in formation and evolution of Lyman-alpha systems. The project is currently in progress.

M. Richmond continued to investigate the properties of supernovae, in concert with colleagues at the University of California, Berkeley. Two automatic telescopes at Berkeley's Leuschner Observatory were used to measure precisely the optical light curve of the unusual SN 1994I in M51. This event had a peculiar spectrum, which showed no evidence for hydrogen and little for helium. Its brightness rose very quickly to a peak, then faded equally rapidly, suggesting that its envelope contained little mass; this, in turn, suggests that its progenitor may have been stripped of its outer layers by a companion, or by a very strong stellar wind.

The group also searched through archival images from the Hubble Space Telescope to find high-resolution pictures of the sites of historical supernovae. Of ten candidate sites, most interesting was near the center of the galaxy M83, home of SN 1968L. Multicolor photometry of several star clusters near the location of SN 1968L showed that their stars must be young, less than 7 million years old. If the supernova's progenitor was born at the same time as the clusters, models of stellar evolution predict its mass to have been greater than 25 solar masses, larger than expected for the progenitor of a ``classical'' Type II supernova.

M. Strauss continued his work on observations of the large-scale distribution of galaxies, and statistical and theoretical analyses thereof. In collaboration with B. Santiago and O. Lahav (Cambridge), M. Davis (U.C. Berkeley), A. Dressler (Carnegie), and J. Huchra (Harvard), a redshift survey of the brightest 8600 galaxies in the sky at high Galactic latitudes (|b| > 20) was completed. This is the first deep redshift survey of optically selected galaxies performed over most of the celestial sphere. Techniques were developed for deriving the galaxy density field from these data correctly accounting for Galactic extinction and the different selection of each of the three galaxy catalogs making up the survey. The luminosity and diameter functions of galaxies are derived; although these quantities are biased by magnitude errors in the catalog, the density field is surprisingly insensitive to magnitude errors.

In collaboration with T. Crawford (Colorado), J. Marr (Union), and B. Partridge (Haverford), Strauss carried out a VLA study at 6 and 20 cm of a sample of 40 ultraluminous IRAS galaxies. The radio morphologies of these objects were found to vary widely, from very compact unresolved sources, to resolved disks, to jet-like linear sources. Nevertheless, the strong correlation observed between the far-infared and radio luminosities of these sources at lower luminosities appears to continue up through the very highest luminosities, arguing that they share a common energy source, namely star formation.

In collaboration with A. Szomoru (Groningen), J. van Gorkom (Columbia) and M. Gregg (LLNL), Strauss carried out a VLA HI study of galaxies in the Bootes Void and in more normal environments. The HI properties of these galaxies (especially HI mass and number of close companions) were found to be surprisingly insensitive to the large-scale (30 Mpc) environment, arguing that galaxy properties are much more determined by their environs on scales of roughly 1 Mpc.

J. Kepner (graduate student), F. Summers, and M. Strauss, derived an extension of the Cosmic Virial Theorem of Peebles which relates the small-scale velocity dispersion of galaxies to their correlations. They show that a similar relation holds for subsets of particles with a common density, which motivates them to suggest a statistic based on redshift surveys that can separate out the density dependence of the velocity dispersion. This statistic may be a useful discriminant between models when the Sloan Digital Sky Survey data become available.

Strauss completed a review of redshift surveys, with special emphasis on developments of the last few years, to be published in the proceedings of a winter school held in Jerusalem in January.

G.R. Knapp and M.P. Rupen (NRAO) completed a survey of elliptical galaxies in the CO(2-1) line using the Caltech Submillimeter Observatory (CSO). Dense cold molecular gas has now been detected in several tens of elliptical galaxies. The overall detection rate is 45%, and the molecular gas correlates well with interstellar dust, seen via its emission at wavelengths m. The molecular gas content of elliptical galaxies is completely uncorrelated with their luminosity or color. The dense molecular gas appears to be confined to the inner regions of the galaxies (within 1 or 2 kpc) in most cases. CO absorption is seen against a flat spectrum radio source in four galaxies, and the velocities of the narrow absorption components suggest infall to the galactic centers.

Knapp and Rupen, together with M. Fich (Waterloo), D.A. Harper (Chicago) and C.G. Wynn-Williams (Hawaii) have begun a project to acquire broad- band images between 4 and 200 m of a large sample of early type galaxies using ISO, to study the distribution of starlight, circumstellar dust and interstellar dust. First results have been obtained from quick-look ISOCAM and ISOPHOT images at 4.5, 6.75, 15 and 21.1 m of the S0/E galaxy NGC 3998, which has an HI polar ring and a bright semi-stellar nucleus. Emission from the cool bulge stars is seen at the shorter wavelengths, while the longer wavelength observations show a compact source close to the nucleus of the galaxy which is likely to be several thousand of dust of temperature 100 - 200 K, perhaps heated by the radiation from the AGN.

G. Jungman (Syracuse), M. Kamionkowski (Columbia), A. Kosowsky (Harvard), and D.N. Spergel showed that the angular power spectrum of the cosmic microwave background (CMB) contains information on virtually all cosmological parameters of interest, including the geometry of the Universe , the baryon density, the Hubble constant (h), the cosmological constant (), the number of light neutrinos, the ionization history, and the amplitudes and spectral indices of the primordial scalar and tensor perturbation spectra. They reviewed the imprint of each parameter on the CMB. Assuming only that the primordial perturbations were adiabatic, they used a covariance-matrix approach to estimate the precision with which these parameters can be determined by a CMB temperature map as a function of the fraction of sky mapped, the level of pixel noise, and the angular resolution. For example, with no prior information about any of the cosmological parameters, a full-sky CMB map with angular resolution and a noise level of 15 K per pixel can determine , h, and with standard errors of 0.1 or better, and provide determinations of other parameters which are inaccessible with traditional observations. Smaller beam sizes or prior information on some of the other parameters from other observations improve the sensitivity. The dependence on the underlying cosmological model was discussed.

D. Spergel, N. Cornish (Case Western Reserve University) and G. Starkman (CWRU) proposed that we live in a finite negatively closed universe. They showed that this suggestion can help reconcile observations that suggest that the universe is open with the predictions of inflation. They have also shown how future microwave background observations could be used to test this hypothesis.

S. Malhotra, D.N. Spergel and J.E. Rhoads used the near infrared fluxes of local galaxies derived from Cosmic Background Explorer (COBE)/ Diffuse Infrared Background Experiment (DIRBE) J(1.25 m) K (2.2 m) & L (3.5 m) band maps and published Cepheid distances to construct Tully-Fisher diagrams for the nearby galaxies. The measured dispersions in these luminosity-linewidth diagrams are remarkably small: = 0.09 magnitudes, = 0.13 magnitudes, and = 0.20 magnitudes. These dispersions include contributions from both the intrinsic Tully-Fisher relation scatter and the errors in estimated galaxy distances, fluxes, inclination angles, extinction corrections, and circular speeds. For the J and K bands, Monte Carlo simulations give a 95% confidence interval upper limit on the true scatter in the Tully-Fisher diagram of and . The Milky Way's luminosity was determined and the Milky Way placed in the Tully-Fisher diagram by fitting a bar plus exponential disk model to the all-sky DIRBE maps. For ``standard'' values of its size and circular speed (Sun-Galactic center distance = 8.5kpc and =220km/s), the Milky Way lies within 1.5 of the TF relations.

Malhotra, Spergel and Rhoads used the TF relation and the Cepheid distances to nearby bright galaxies to constrain and : - log(/ 8.5kpc) +1.63log(/ 220km/s) = 0.08 0.03. Alternatively, if standard values are assigned to the parameters of the Galaxy and the Cepheid zero-point is ignored, the Tully-Fisher relation can be used to determine the Hubble Constant directly: 12 km/s/Mpc.

They have also tested the Tully-Fisher relation at longer wavelengths, where the emission is dominated by dust. No evidence was found for a Tully Fisher relation at wavelengths beyond 10m. The tight correlation seen in the L band suggests that stellar emission dominates over the 3.3 m PAH emission.

R. Kulsrud, R. Cen, J.P. Ostriker and D. Ryu (Korea) proposed a new origin for galactic magnetic fields. They demonstrated that strong magnetic fields are produced from a zero initial magnetic field during the pregalactic era, when galaxies are first forming. The development of the magnetic fields proceeds in three phases. In the first phase, weak magnetic fields are created by the Biermann battery mechanism, acting in shocked parts of the intergalactic medium where caustics form and intersect. In the second phase, these weak magnetic fields are amplified to strong magnetic fields by the Kolmogoroff turbulence endemic to gravitational structure formation of galaxies. During this second phase, the magnetic fields reach saturation with the turbulent power, but they are coherent only on the scale of the smallest eddy. In the third phase, the magnetic field strength increases to equipartition with the turbulent energy, and the coherence length of the magnetic fields increases to the scale of the largest turbulent eddy, comparable to the scale of the entire galaxy. The resulting magnetic field represents a galactic magnetic field of primordial origin. No further dynamo action is necessary, after the galaxy forms, to explain the origin of magnetic fields. However, the magnetic field may be altered by dynamo action once the galaxy and the galactic disk have formed.

The first phase was demonstrated by direct numerical simulation in which the magnetic field equation is added on to the normal equations for structure formation. It was shown that the vorticity and the cyclotron frequency during this phase should be equal everywhere up to a factor involving the fractional ionization. This remarkable result is confirmed by the simulations

The second phase could not be followed by the numerical simulation because of the large numerical resistivity and viscosity. Its behavior was derived by employing a standard analytic theory for the generation of magnetic energy by turbulence.

The investigation of the third phase is being carried out by a complicated numerical turbulence plasma calculation (the Direct Interaction Approximation). This calculation is being done by B. Chandran (graduate student). It is shown that given a steady input of kinetic energy at large scales, the magnetic energy builds up to equipartition with the kinetic energy on all scales. Thus, the magnetic field resulting from phase three appears to be coherent on the largest scale, the scale of the entire galaxy. Qualitative physical arguments seem to bear this out.

Future calculations will investigate the buildup of magnetic energy when the input of kinetic energy is not steady in time. This numerical calculation is also important from the point of view of basic plasma physics. It demonstrates steady state power spectra for MHD turbulence, and explores technical points such as correlation times in MHD turbulence and the effects of kinetic energy sources.

A. Ulmer (graduate student) showed that the two-point correlation function, , of the Lyman- forest is found to be large, , confidence level, on the scale of 250-500 km/s for a sample of absorbers (0 < z < 1.3) assembled from HST Key Project Observations. This correlation function is stronger than at high redshift (z > 1.7) where for velocities > 250 km/s.

By comparing neutrino fluxes and central temperatures calculated from 1000 detailed numerical solar models, J. N. Bahcall (I.A.S.) and A. Ulmer derived improved scaling laws which show how each of the neutrino fluxes depends upon the central temperature (flux ); they also estimated uncertainties for the scaling exponents. With the aid of a one-zone model of the sun, Bahcall and Ulmer derived analytical expressions for the temperature-exponents of the neutrino fluxes. For the most important neutrino fluxes, the exponents calculated analytically agreed to 20% or better with the exponents extracted from the detailed numerical models. The one-zone model provides a physical understanding of the temperature dependence of the neutrino fluxes. For the pp neutrino flux, the one-zone model explains the (initially-surprising) dependence of the flux upon a negative power of the temperature and suggests a new functional dependence. This new function makes explicit the strong anti-correlation between the Be and pp neutrino fluxes. The one-zone model also predicts successfully the average correlation between other neutrino fluxes, but cannot predict the appreciable scatter in a versus correlation diagram.

E.L. Turner, W.N. Colley (graduate student) and J.A. Tyson (Bell Labs) obtained a unique reconstruction of the image of a high-redshift galaxy responsible for multiple long arcs in the z = 0.4 cluster 0024+16 by inverse lensing calculations. Deep B and I band imaging with the Hubble Space Telescope allowed high resolution of the arcs due to strong gravitational lensing of the background source. Each of the five strongly lensed images of the source yielded the same reconstructed source image, exhibiting a beaded, ringlike morphology. The U luminosity of the ring alone is equivalent to that of a normal bright galaxy, and it is tempting to conclude that this is a galaxy in formation.

Turner and Y. Wang (Fermilab) note that interferometric gravitational wave detectors may someday measure the frequency sweep of a binary neutron star inspiral (characterized by its chirp mass) to high accuracy. The observed chirp mass is the intrinsic chirp mass of the binary source multiplied by (1 + z), where z is the source redshift. Assuming a non-zero cosmological constant, the expected redshift distribution of observed events for an advanced LIGO style detector was computed. This redshift distribution has a robust and sizable dependence on the cosmological constant.

Turner, D.J. Eisenstein and A. Loeb (CfA) propose a new method to measure the mass of large-scale filaments found in galaxy redshift surveys. The method is based on the fact that the mass per unit length of isothermal filaments depends only on the transverse velocity dispersion. Filaments that lie transverse to the line of sight may therefore have their masses measured from their thickness in redshift space. Tests of the method on filaments found in N-body simulations show that it is accurate to about 35%, and a preliminary application of the technique to a selected region of the Perseus-Pisces supercluster gives a mass-to-light ratio of 450h in solar units to within a factor of two. This method allows mass-to-light ratio determinations on mass scales up to 10 times larger than that of individual galaxy clusters and could thereby yield new information on the large scale behavior of dark matter.

Turner, A. Stebbins and Y. Wang (Fermilab) have investigated the gravitational lensing of gravitational waves from merging neutron star binaries. They find that the distribution of observed event redshifts (or, equivalently, of observed chirp masses) will have a sharp cut-off in the absence of lensing effects for an advanced LIGO style gravitational wave detector. However, a low amplitude tail extending to higher redshifts (chirp masses) is expected due to gravitational microlensing events. An advanced system might see a few such events per year if compact objects comprised close to 10% of the critical cosmological density.