arXiv subjects
Publications and source records attributed to Lars Hernquist.
We consider the field equations of a static magnetic field including one-loop QED corrections, and calculate the corrections to the field of a magnetic dipole. PACS: 12.20.Ds, 97.60.Jd, 97.60.Gb
We use an analytic form for the Heisenberg-Euler Lagrangian to calculate the birefringent and dichroic properties of the vacuum for arbitrarily strong wrenchless fields. PACS : 12.20.Ds, 42.25.Lc 97.60.Jd, 98.70.Rz
We calculate the number of damped Lyman-alpha absorbers expected in various popular cosmological models as a function of redshift and compare our predictions with observed abundances. The Press-Schechter formalism is used to obtain the distribution of halos with circular velocity in different cosmologies, and we calibrate the relation between circular velocity and absorption cross-section using detailed gas dynamical simulations of a ``standard'' cold dark matter (CDM) model. Because of this calibration, our approach makes more realistic assumptions about the absorption properties of collapsed objects than previous, analytic calculations of the damped Lyman-alpha abundance. CDM models with Omega_0=1, H_0=50, baryon density Omega_b=0.05, and scale-invariant primeval fluctuations reproduce the observed incidence and redshift evolution of damped Lyman-alpha absorption to within observational uncertainty, for both COBE normalization (sigma_8=1.2) and a lower normalization (sigma_8=0.7) that better matches the observed cluster abundance at z=0. A tilted (n=0.8, sigma_8=0.7) CDM model tends to underproduce absorption, especially at z=4. With COBE normalization, a CDM model with Omega_0=0.4, Omega_{Lambda}=0.6 gives an acceptable fit to the observed absorption; an open CDM model is marginally acceptable if Omega_0 is at least 0.4 and strongly inconsistent with the z=4 data if Omega_0=0.3. Mixed dark matter models tend not to produce sufficient absorption, being roughly comparable to tilted CDM models if Omega_{nu} = 0.2 and failing drastically if Omega_{nu} = 0.3.
Any viable cosmological model must produce enough structure at early epochs to explain the amount of gas associated with high-redshift damped Ly$α$ systems. We study the evolution of damped Ly$α$ systems at redshifts $z\ge 2$ in cold dark matter (CDM) and cold+hot dark matter (CDM+HDM) models using both N-body and hydrodynamic simulations. Our approach incorporates the effects of gas dynamics, and we find that all earlier estimates which assumed that all the baryons in dark matter halos would contribute to damped Ly$α$ absorption have overestimated the column density distribution $f(N)$ and the fraction of neutral dense gas $Ω_g$ in damped Ly$α$ systems. The differences are driven by ionization of hydrogen in the outskirts of galactic halos and by gaseous dissipation near the halo centers, and they tend to exacerbate the problem of late galaxy formation in CDM+HDM models. We only include systems up to the highest observed column density $N\sim 10^{21.8}$ cm$^{-2}$ in the estimation of $Ω_g$ for a fair comparison with data. If the observed $f(N)$ and $Ω_g$ inferred from a small number of confirmed and candidate absorbers are robust, the amount of gas in damped Ly$α$ systems at high redshifts in the $Ω_ν=0.2$ CDM+HDM model falls well below the observations.
In gravitational instability models, \lya absorption arises from a continuous fluctuating medium, so that spectra provide a non-linear one-dimensional ``map'' of the underlying density field. We characterise this continuous absorption using statistical measures applied to the distribution of absorbed flux. We describe two simple members of a family of statistics which we apply to simulated spectra in order to show their sensitivity as probes of cosmological parameters (H$_{0}$, $Ω$, the initial power spectrum of matter fluctuations) and the physical state of the IGM. We make use of SPH simulation results to test the flux statistics, as well as presenting a preliminary application to Keck HIRES data.
We describe PTreeSPH, a gravity treecode combined with an SPH hydrodynamics code designed for massively parallel supercomputers having distributed memory. Our computational algorithm is based on the popular TreeSPH code of Hernquist & Katz (1989). PTreeSPH utilizes a domain decomposition procedure and a synchronous hypercube communication paradigm to build self-contained subvolumes of the simulation on each processor at every timestep. Computations then proceed in a manner analogous to a serial code. We use the Message Passing Interface (MPI) communications package, making our code easily portable to a variety of parallel systems. PTreeSPH uses individual smoothing lengths and timesteps, with a communication algorithm designed to minimize exchange of information while still providing all information required to accurately perform SPH computations. We have additionally incorporated cosmology, periodic boundary conditions with forces calculated using a quadrupole Ewald summation method, and radiative cooling and heating from a parameterized ionizing background following Katz, Weinberg & Hernquist (1996). The addition of other physical processes, such as star formation, is straightforward. A cosmological simulation from z=49 to z=2 with 64^3 gas particles and 64^3 dark matter particles requires ~6000 node-hours on a Cray T3D, with a communications overhead of ~10% and is load balanced to a ~90% level. When used on the new Cray T3E, this code will be capable of performing cosmological hydrodynamical simulations down to z=0 with ~2x10^6 particles, or to z=2 with ~10^7 particles, in a reasonable amount of time. Even larger simulations will be practical in situations where the matter is not highly clustered or when periodic boundaries are not required.
We study metal line absorption of CIV, CII, SiIV, and NV at redshifts z=3.5 to z=2 within the framework of a cosmological model for the Lyman alpha forest, comparing the results of numerical simulations to recent observations by Songaila & Cowie (1996, SC). In agreement with Rauch, Haehnelt & Steinmetz (1996), we find that the observed mean value of the CIV/HI ratio around z=3 is reproduced if a uniform metallicity of [C/H] = -2.5 is assumed in our model, but that the observed scatter in this ratio is larger than predicted, implying a scatter in the metallicity of the absorbing systems of roughly an order of magnitude. We further argue that absorbers with relatively low column densities (log(N_HI) < 15) likely have a mean metallicity [C/H] less than -2.5, a result which is basically independent of the model considered, and similar to what is predicted in the simulations by Gnedin & Ostriker (1997). Our model predicts no significant evolution in the mean values of metal line column densities between z=3.5 and z=2. Comparison of the predicted and observed numbers of SiIV and NV systems suggests that the photoionizing background radiation field around z=3 is somewhat softer than that proposed by Haardt & Madau (1996). Our model does not account for the increase in the SiIV/CIV ratio at z=3.2 found by SC. While SC suggested that the increase could be explained by a softening of the radiation field due to HeII absorption at z > 3, such a modification does not raise the mean value of SiIV/CIV in our simulation because it shifts numerous weak SiIV systems to just above the detection limit, thus keeping the mean column density of observable SiIV systems low.
We derive analytic lower bounds on the cosmic baryon density by requiring that the high-redshift IGM contain enough neutral hydrogen to produce the observed \lya absorption in quasar spectra. The key theoretical assumption is that absorbing structures are no more extended in redshift space than in real space. This assumption might not hold if \lya clouds are highly overdense and thermally broadened, but it is likely to hold in the gravitational instability picture for the \lya forest suggested by cosmological simulations, independently of the details of the cosmological model. The other ingredients in these bounds are an estimate of the UV background from quasars, a temperature $T~10^4K$ for the "warm" photoionized IGM that produces most of the \lya absorption, a value of the Hubble constant, and observational estimates of the mean \lya flux decrement $\Dbar$ or, for a more restrictive bound, the distribution function $P(τ)$ of \lya optical depths. With plausible parameter values, the mean decrement bound implies a baryon density parameter $\Omb \ga 0.0125/h^2$. With conservative values, the bound weakens to $\Omb \ga 0.005/h^2$, but the required clustering of the IGM is then incompatible with other properties of quasar spectra. A recent observational determination of $P(τ)$ implies $\Omb \ga 0.0125/h^2$ even for a conservative estimate of the UV background, and $\Omb \ga 0.018/h^2$ for a more reasonable estimate. These bounds are consistent with recent low estimates of the primordial deuterium-to-hydrogen ratio $\dtoh$, which imply $\Omb \approx 0.025/h^2$, but the $P(τ)$ bound can only be reconciled with high $\dtoh$ estimates by abandoning standard big bang nucleosynthesis or the gravitational instability picture for the origin of the \lya forest. (Shortened abstract.)
We use smoothed-particle hydrodynamics to examine differences between direct collisions of single stars and binary star mergers in their roles as possible blue straggler star formation mechanisms. We find in all cases that core helium in the progenitor stars is largely retained in the core of the remnant, almost independent of the type of interaction or the central concentration of the progenitor stars. We have also modelled the subsequent evolution of the hydrostatic remnants, including mass loss and energy input from the hydrodynamical interaction. The combination of the hydrodynamical and hydrostatic models enables us to predict that little mixing will occur during the merger of two globular cluster stars of equal mass. In contrast to the results of Proctor Sills, Bailyn, & Demarque (1995), we find that neither completely mixed nor unmixed models can match the absolute colors of observed blue stragglers in NGC 6397 at all luminosity levels. We also find that the color distribution is probably the crucial test for explanations of BSS formation - if stellar collisions or mergers are the correct mechanisms, a large fraction of the lifetime of the straggler must be spent away from the main sequence. This constraint appears to rule out the possibility of completely mixed models. For NGC 6397, unmixed models predict blue straggler lifetimes ranging from about 0.1 to 4 Gyr, while completely mixed models predict a range from about 0.6 to 4 Gyr.
We use an automated Voigt-profile fitting procedure to extract statistical properties of the Ly$α$ forest in a numerical simulation of an $Ω=1$, cold dark matter (CDM) universe. Our analysis method is similar to that used in most observational studies of the forest, and we compare the simulations to recently published results derived from Keck HIRES spectra. With the Voigt-profile decomposition analysis, the simulation reproduces the large number of weak lines ($N_{\rm HI}\la 10^{13}\cdunits$) found in the HIRES spectra. The column density distribution evolves significantly between $z=3$ and $z=2$, with the number of lines at fixed column density dropping by a factor $\sim 1.6$ in the range where line blending is not severe. At $z=3$, the $b$-parameter distribution has a median of $35 \kms$ and a dispersion of $20 \kms$, in reasonable agreement with the observed values. The comparison between our new analysis and recent data strengthens earlier claims that the \lya forest arises naturally in hierarchical structure formation as photoionized gas falls into dark matter potential wells. However, there are two statistically signficant discrepancies between the simulated forest and the HIRES results: the model produces too many lines at $z=3$ by a factor $\sim 1.5-2$, and it produces more narrow lines ($b<20 \kms$) than are seen in the data. The first result is sensitive to our adopted normalization of the mean \lya optical depth, and the second is sensitive to our assumption that helium reionization has not significantly raised gas temperatures at $z=3$. It is therefore too early to say whether these discrepancies indicate a fundamental problem with the high-redshift structure of the $Ω=1$ CDM model or reflect errors of detail in our modeling of the gas distribution or the observational procedure.
Lyman limit and damped Lyman-alpha absorption systems probe the distribution of collapsed, cold gas at high redshift. Numerical simulations that incorporate gravity and gas dynamics can predict the abundance of such absorbers in cosmological models. We develop a semi-analytical method to correct the numerical predictions for the contribution of unresolved low mass halos, and we apply this method to the Katz et al. (1996) simulation of the standard cold dark matter model ($Ω=1$, $h=0.5$, $Ω_b=0.05$, $σ_8=0.7$). Using this simulation and higher resolution simulations of individual low mass systems, we determine the relation between a halo's circular velocity $v_c$ and its cross section for producing Lyman limit or damped absorption. We combine this relation with the Press-Schechter formula for the abundance of halos to compute the number of absorbers per unit redshift. The resolution correction increases the predicted abundances by about a factor of two at z=2, 3, and 4, bringing the predicted number of damped absorbers into quite good agreement with observations. Roughly half of the systems reside in halos with circular velocities $v_c>100\kms$ and half in halos with $35\kms 150\kms$ typically harbor two or more systems capable of producing damped absorption. Even with the resolution correction, the predicted abundance of Lyman limit systems is a factor of three below observational estimates, signifying either a failure of standard CDM or a failure of these simulations to resolve the systems responsible for most Lyman limit absorption. By comparing simulations with and without star formation, we find that depletion of the gas supply by star formation affects absorption line statistics at $z>=2$ only for column densities exceeding $N_{HI}=10^{22} cm^{-2}$.
We use an automated Voigt-profile fitting procedure to extract statistical properties of the Ly$α$ forest in a numerical simulation of an $Ω=1$, cold dark matter (CDM) universe. Our analysis method is similar to that used in most observational studies of the forest, and we compare the simulations to recently published results derived from Keck HIRES spectra. With the Voigt-profile decomposition analysis, the simulation reproduces the large number of weak lines ($N_{\rm HI}\la 10^{14} cm^{-2}$) found in the HIRES spectra. At $z=3$, the $b$-parameter distribution has a median of $35\kms$ and a dispersion of $20\kms$, in reasonable agreement with the observed values. The comparison between our new analysis and recent data strengthens earlier claims that the Ly$alpha$ forest arises naturally in hierarchical structure formation as photoionized gas falls into dark matter potential wells.
We calculate the reaction cross-sections for the fusion of hydrogen and deuterium in strong magnetic fields as are believed to exist in the atmospheres of neutron stars. We find that in the presence of a strong magnetic field ($B \gsim 10^{12}$G), the reaction rates are many orders of magnitude higher than in the unmagnetized case. The fusion of both protons and deuterons are important over a neutron star's lifetime for ultrastrong magnetic fields ($B \sim 10^{16}$G). The enhancement may have dramatic effects on thermonuclear runaways and bursts on the surfaces of neutron stars.
We derive an analytic form for the Heisenberg-Euler Lagrangian in the limit where the component of the electric field parallel to the magnetic field is small. We expand these analytic functions to all orders in the field strength ($F_{μν}F^{μν}$) in the limits of weak and strong fields, and use these functions to estimate the pair-production rate in arbitrarily strong electric fields and the photon-splitting rate in arbitrarily strong magnetic fields.
Using cosmological simulations that incorporate gas dynamics and gravitational forces, we investigate the influence of photoionization by a UV radiation background on the formation of galaxies. In our highest resolution simulations, we find that photoionization has essentially no effect on the baryonic mass function of galaxies at $z=2$, down to our resolution limit of $5e9 M_\sun$. We do, however, find a strong interplay between the mass resolution of a simulation and the microphysics included in the computation of heating and cooling rates. At low resolution, a photoionizing background can appear to suppress the formation of even relatively massive galaxies. However, when the same initial conditions are evolved with a factor of eight better mass resolution, this effect disappears. Our results demonstrate the need for care in interpreting the results of cosmological simulations that incorporate hydrodynamics and radiation physics. For example, we conclude that a simulation with limited resolution may yield more realistic results if it ignores some relevant physical processes, such as photoionization. At higher resolution, the simulated population of massive galaxies is insensitive to the treatment of photoionization and star formation, but it does depend significantly on the amplitude of the initial density fluctuations. By $z=2$, an $Ω=1$ cold dark matter model normalized to produce the observed masses of present-day clusters has already formed galaxies with baryon masses exceeding $1e11 M_\sun$.
The role that minor mergers have played in the formation and structure of the Milky Way is still an open question, about which there is much debate. We use numerical simulations to explore the evolution of debris from a tidally disrupted satellite, with the aim of developing a method that can be used to identify and quantify signatures of accretion in a survey of halo stars. For a Milky Way with a spherical halo, we find that debris from minor mergers can remain aligned along great circles throughout the lifetime of the Galaxy. We exploit this result to develop the method of Great Circle Cell Counts (GC3), which we test by applying it to artificially constructed halo distributions. Our results suggest that if as few as 1\% of the stars in a halo survey are accreted from the disruption of a single subsystem smaller than the Large Magellanic Cloud, GC3 can recover the great circle associated with this debris. The dispersion in GC3 can also be used to detect the presence of structure characteristic of accretion in distributions containing a much smaller percentage of material accreted from any single satellite.
Using numerical simulation, we study the development of gaseous inflows and triggering of starburst activity in mergers of comparable-mass disk galaxies. In all encounters studied, the galaxies experience strong gaseous inflows and moderate to intense starburst activity. We find that galaxy structure plays a dominant role in regulating activity. The gaseous inflows are strongest when galaxies with dense central bulges are in the final stages of merging, while inflows in bulgeless galaxies are weaker and occur earlier in the interaction. Orbital geometry plays only a relatively modest role in the onset of collisionally-induced activity. Through an analysis of the torques acting on the gas, we show that these inflows are generally driven by gravitational torques from the host galaxy (rather than the companion), and that dense bulges act to stabilize galaxies against bar modes and inflow until the galaxies merge, at which point rapidly varying gravitational torques drive strong dissipation and inflow of gas in the merging pair. The strongest inflows (and associated starburst activity) develop in co-planar encounters, while the activity in inclined mergers is somewhat less intense and occurs slightly later during the merger. The starbursts which develop in mergers of galaxies with central bulges represent an increase in the star formation rate of two orders of magnitude over that in isolated galaxies. We find that the gaseous and stellar morphology and star-forming properties of these systems provide a good match to those of observed ultraluminous infrared galaxies. Our results imply that the internal structure of the merging galaxies, rather than orbital geometry, may be the key factor in producing ultraluminous infrared galaxies.
Cosmological simulations with gas dynamics suggest that the Lyman-alpha forest is produced mainly by "small scale structure" --- filaments and sheets that are the high redshift analog of today's galaxy superclusters. There is no sharp distinction between Lyman-alpha clouds and "Gunn-Peterson" absorption produced by the fluctuating IGM -- the Lyman-alpha forest {\it is} the Gunn-Peterson effect. Lyman limit and damped Lyman-alpha absorption arises in the radiatively cooled gas of forming galaxies. At $z~2-3$, most of the gas is in the photoionized, diffuse medium associated with the Lyman-alpha forest, but most of the {\it neutral} gas is in damped Lyman-alpha systems. We discuss generic evolution of cosmic gas in a hierarchical scenario of structure formation, with particular attention to the prospects for detecting 21cm emission from high redshift HI. A scaling argument based on the present-day cluster mass function suggests that objects with $M_{HI} >~ 5e11 h^{-1} \msun$ should be extremely rare at $z~3$, so detections with existing instruments will be difficult. An instrument like the proposed Square Kilometer Array could detect individual damped Lyman-alpha systems at high redshift, making it possible to map structure in the high redshift universe in much the same way that today's galaxy redshift surveys map the local large scale structure.