We use hydrodynamic cosmological simulations in a (600 Mpc)^3 volume to study the observability of baryon acoustic oscillations (BAO) in the intergalactic medium as probed by Lyman alpha forest (LAF) absorption. The large scale separation between the wavelength of the BAO mode (~150 Mpc) and the size of LAF absorbers (~100 kpc) makes this a numerically challenging problem. We report on several 2048^3 simulations of the LAF using the ENZO code. We adopt WMAP5 concordance cosmological parameters and power spectrum including BAO perturbations. 5000 synthetic HI absorption line spectra are generated randomly piercing the box face. We calculate the cross-correlation function between widely separated pairs. We detect the BAO signal at z=3 where theory predicts to moderate statistical significance.

We give a comprehensive statistical description of the Lyα absorption from the intergalactic medium in a hydrodynamic simulation at redshifts 0.1–1.6, the range of redshifts covered by spectra of quasi-stellar objects obtained with the Hubble Space Telescope (HST). We use the enzo code to make a simulation in a 10243 cube with side length 76.8 Mpc comoving using 75-kpc cells, for a Hubble constant of 71 km s−1 Mpc−1. The best prior work, by Davé et al., used a smoothed particle hydrodynamics simulation in a 15.6-Mpc box with an effective resolution of 245 kpc at the cosmic mean density and slightly different cosmological parameters. We use a popular cosmological model and astrophysical parameters that describe the ultraviolet background that photoionizes and heats the gas. At redshift z= 2 this simulation is different from data. Tytler et al. found that the simulated spectra at z= 2 have too little power on large scales, Lyα lines are too wide, there is a lack of high column density lines and there is a lack of pixels with low flux. Here we present statistics at z < 1.6, including the flux distribution, the mean flux, the effective opacity and the power and correlation of the flux. We also give statistics of the Lyα lines including the linewidth distribution, the column density distribution, the number of lines per unit equivalent width and redshift and the correlation between the linewidth and column density. We find that the mean amount of absorption in the simulated spectra changes smoothly with redshift as DA(z) = 0.0102(1 +z)2.252. Both the trend and absolute values are close to measurements of HST spectra by Kirkman et al. The column density distribution is the same as data at redshift 1.6, but at lower redshifts the simulation has fewer lines than data, perhaps because of issues making measurements in low signal-to-noise ratio spectra. The lines in data have smaller widths than in the simulation but again the data are not very reliable. We argue that these differences are probably not caused by lack of numerical resolution. Although these possible differences are larger than those that we saw at z= 2, overall, the simulation gives a good description of HST spectra at 0.1 < z < 1.6 given that we did not adjust either the parameters that we input to specify the simulations or the output from the simulation.

We explore the effects of size of the box that we use for simulations of the intergalactic medium (IGM) at redshift 2. We examine simulations from the hydrodynamic code enzo that differ only in box size. We study the cold dark matter (CDM) distribution and many statistics of the Lyα forest absorption from the IGM. Larger boxes have fewer pixels with significant absorption (flux <0.96), more pixels in longer stretches with little or no absorption, and they have wider Lyα lines. The larger boxes differ only because they include power from longer wavelength modes. These modes result in higher peak densities, higher velocities and hotter gas. Small simulations are too cold compared to larger ones. When we deliberately increase the heat we put into the IGM, we can approximate the Lyα forest in a simulation of twice the size. When we double the box size, the difference of most statistics from their value in our largest 76.8 Mpc box is reduced by approximately a factor of 2. When we enlarge the box from 38.4 to 76.8 Mpc, the mean Lyα absorption decreases 0.5 per cent, the frequency with which we encounter different common CDM densities changes by 2 per cent, typical Lyα linewidths, the frequency of flux values and the power spectrum of the flux all change by 4–7 per cent, and the column density distribution changes by up to 15 per cent. A 76.8 Mpc box is large enough to give errors of 1017 cm−2. Decreasing the cell size from 75 to 18.75 kpc makes the difference larger at log NH i < 14 cm−2, and does not help with the higher columns. The Lyα lines in the simulated spectra from a very large box with 18.75 kpc cells would be too wide by 2.6 km s−1. The simulated Lyα forest has 20 per cent too little power on small scales and 50 per cent too little large scales. A much larger box would increase the large-scale power only a few per cent. We confirm the Kim et al. (2007) and Bolton et al. (2008) finding that the simulated spectra also have different flux distributions than data. It is hard to see how our optically thin simulations using popular cosmological and astrophysical parameters can match the Lyα forest data at z= 2. Adding radiation transfer effects, especially shelf-shielding will reduce the temperature at high overdensities, possibly improving the match to linewidths, and it will help match the number of high column density lines. We could also decrease the linewidths with a softer ionizing spectrum, or by using σ8 > 0.9, which has the additional benefit of increasing the large-scale power.

We consider a PDE system comprising compressible hydrodynamics, flux-limited diffusion radiation transport and chemical ionization kinetics in a cosmologically-expanding universe. Under an operator-split framework, the cosmological hydrodynamics equations are solved through the piecewise parabolic method, as implemented in the Enzo community hydrodynamics code. The remainder of the model, including radiation transport, chemical ionization kinetics, and gas energy feedback, form a stiff coupled PDE system, which we solve using a fully-implicit inexact Newton approach, and which forms the crux of this paper. The inner linear Newton systems are solved using a Schur complement formulation, and employ a multigrid-preconditioned conjugate gradient solver for the inner Schur systems. We describe this approach and provide results on a suite of test problems, demonstrating its accuracy, robustness, and scalability to very large problems.

We give a comprehensive statistical description of the Ly alpha absorption from the intergalactic medium in a hydrodynamic simulation at redshifts 0.1-1.6, the range of redshifts covered by spectra of quasi-stellar objects obtained with the Hubble Space Telescope (HST). We use the ENZO code to make a simulation in a 1024(3) cube with side length 76.8Mpc comoving using 75-kpc cells, for a Hubble constant of 71 km s(-1) Mpc(-1). The best prior work, by Dave et al., used a smoothed particle hydrodynamics simulation in a 15.6-Mpc box with an effective resolution of 245 kpc at the cosmic mean density and slightly different cosmological parameters. We use a popular cosmological model and astrophysical parameters that describe the ultraviolet background that photoionizes and heats the gas. At redshift z = 2 this simulation is different from data. Tytler et al. found that the simulated spectra at z = 2 have too little power on large scales, Ly alpha lines are too wide, there is a lack of high column density lines and there is a lack of pixels with low flux. Here we present statistics at z < 1.6, including the flux distribution, the mean flux, the effective opacity and the power and correlation of the flux. We also give statistics of the Ly alpha lines including the linewidth distribution, the column density distribution, the number of lines per unit equivalent width and redshift and the correlation between the linewidth and column density. We find that the mean amount of absorption in the simulated spectra changes smoothly with redshift as DA(z) = 0.0102(1 + z)(2.252). Both the trend and absolute values are close to measurements of HST spectra by Kirkman et al. The column density distribution is the same as data at redshift 1.6, but at lower redshifts the simulation has fewer lines than data, perhaps because of issues making measurements in low signal-to-noise ratio spectra. The lines in data have smaller widths than in the simulation but again the data are not very reliable. We argue that these differences are probably not caused by lack of numerical resolution. Although these possible differences are larger than those that we saw at z = 2, overall, the simulation gives a good description of HST spectra at 0.1 < z < 1.6 given that we did not adjust either the parameters that we input to specify the simulations or the output from the simulation.

We present the first large sample of metal absorption systems in pairs of QSOs with sightlines separated by about 1 Mpc at z = 2. We found 690 absorption systems in the spectra of 310 QSOs in 170 pairings. Most systems show C IV or Mg II absorption.When we see absorption in one QSO, the probability that we see absorption in the paired QSO, within about 500 km s(-1), is at least similar to 50 per cent at <100 kpc, declining rapidly to similar to 8 per cent at 200-400 kpc, similar to 0.8 per cent by 1-2 Mpc proper distance. Although we may occasionally see an individual absorbing halo in two sightlines, the absorber-absorber correlation is primarily a probe of the distribution of metals around galaxies and Mpc scale galaxy clustering. QSO absorption lines give redshifts errors of similar to 23 km s(-1), almost 10 times smaller than the error for galaxy spectra at these redshifts, hence we can measure clustering on small scales, around 0.5 Mpc proper, with a small sample. The distribution of 23 absorber-absorber coincidences separated by <2.5 Mpc at z similar to 2 is consistent with an origin in galaxies with a normal correlation function, normal systematic infalling velocities and low random pair-wise velocity differences, more consistent with blue than with red galaxies. Absorption in gas flowing out from galaxies with a mean velocity of 250 km s(-1) would produce more redshift elongation than we see. The fast winds detected by Adelberger et al. in the same ions account for less than 1/3 of the absorption systems we see. Such winds may be confined to the ultraviolet (UV) luminous star-forming regions of Lyman break galaxies. If most galaxies have winds, they cannot extend to 40 kpc with large velocities, while continuing to make UV absorption that we can detect. This suggests that most metals seen in the intergalactic medium at z = 2 arrived long before.We see an excess of C IV absorbers, with an a posteriori probability of 0.0003, when a line of sight passes a foreground QSO. We see 16 absorbers where we expect 5.8 at 0-600 km s(-1), on the front side of the partner QSO. At these velocities, we see an excess absorber in similar to 6 per cent of sightlines that pass within 0.1-2.5 Mpc of a QSO, but in <2 per cent of cases when we look directly at a QSO. These transverse associated absorbers are not the normal line-of-sight associated absorbers that have a broader velocity distribution. Excluding the sightlines to us, the 3D distribution of 59 absorbers around 313 QSOs is approximately isotropic, except for the 1.5-2 sigma tendency for the excess C IV absorbers to be on the front side of the QSOs. Our QSO redshifts may be too large by similar to 300 km s(-1), or there might be a real asymmetry coming from a hypothetical anisotropy in the QSO UV emission, or from isotropic UV emission that lasted less than similar to 1 Myr, possibilities suggested by the excess H I behind these QSOs. The velocity dispersion of the excess absorbers near the QSOs is small, similar to 250 km s(-1), suggesting that both these absorbers and the QSOs are in the blue sequence of galaxies. The probability of seeing absorption when a sightline passes a QSO rises only slowly as the impact parameter drops from 2.5 to 0.1 Mpc, perhaps because the UV radiation from QSOs destroys many nearby absorbers.

We use high quality echelle spectra of 24 QSOs to provide a calibrated measurement of the total amount of Lyα forest absorption (DA) over the redshift range 2.2 < z < 3.2. Our measurement of DA excludes absorption from metal lines or the Lyα lines of Lyman limit systems and damped Lyα systems. We use artificial spectra with realistic flux calibration errors to show that we are able to place continuum levels that are accurate to better than 1%. When we combine our results with our previous results between 1.6 < z < 2.2, we find that the redshift evolution of DA is well described over 1.6 < z < 3.2 as A(1 + z) , where A = 0.0062 and γ = 2.75. We detect no significant deviations from a smooth power law evolution over the redshift range studied. We find less H I absorption than expected at z = 3, implying that the UV background is about 40% higher than expected. Our data appears to be consistent with an H I ionization rate of Γ ∼ 1.4 × 10 s.

We present the first large sample of absorption systems in paired QSOs consisting of 691 absorption systems in the spectra of 310 QSOs including 170 pairings. All these absorption systems have metal lines, usually C IV or Mg II. We see 17 cases of absorption in one line-of-sight within 200 km/s (1 Mpc) of absorption in the paired line-of-sight with the probability at least approx 50 at 100kpc, declining rapidly to 23 0.5Mpc scales and see a hint of the "fingers of God" redshift-space distortion. The distribution matches absorbers arising in galaxies at z=2 with a normal correlation function and systematic infall velocities but unusually low random pair-wise velocity differences. Absorption in gas flowing out from galaxies at a mean velocity of 250 km/s would produce vastly more elongation than we see. The UV absorption from fast winds that Adelberger et al. 2005 see in spectra of LBGs is not representative of the absorption that we see. Either the winds are confined to LBGs, or they can not extend to 40 kpc with large velocities, while continuing to make UV absorption we see, implying most metals were in place in the IGM long before z=2. Separately, when we examine the absorption seen when a sight line passes a second QSO, we see 19 absorbers within 400 km/s of the partner QSO. The probability of seeing absorption is approximately constant for impact parameters 0.1 - 1.5 Mpc. Perhaps we do not see a rapid rise in the probability at small impact parameters because the UV from QSOs destroys some absorbers near to the QSOs. The 3D distribution of 64 absorbers around 313 QSOs is to first order isotropic, with just a hint of the anisotropy expected if the QSO UV emission is beamed, or alternatively QSOs might emit UV isotropically but for a surprisingly short time of only 0.3Myr.

We study the ionization and thermal evolution of the intergalactic medium during the epoch of \heii reionization by means of radiation hydrodynamical cosmological simulations. We post-process baryonic density fields from a standard optically-thin IGM simulation with a homogeneous galaxy-dominated UV background (UVB) which reionizes \hi and \hei at z=6.5 but does not have any contribution to the ionization of \heii. Quasars with luminosities proportional to the mass of the host halos are then introduced as point sources throughout the 100 Mpc simulation volume consistent with the Pei luminosity function. We evolve the spatial distribution of the \heii ionizing radiation field using a time-implicit variable tensor Eddington factor radiative transfer scheme. Simultaneously, we also solve for the local ionization of \heii to \heii and the associated photoheating of the gas. We find that the percolation of the \heiii regions is essentially complete by z=2.5. When comparing to a self-consistent optically thin simulation we find that in optically thick calculation the gas temperature is higher by a factor of approximately 1.7 at the mean gas density level. We use 300 random lines of sight to compute at $\bar{z} = 2.5 \pm 0.1$ a mean \heii \lya line transmission of $\bar{F} = 0.304 \pm 0.002$. We compare the broadening width of the \hi and \heii \lya lines to the results from the self-consistent optically thin simulation and find a shift by approximately 1.25 km/s of the b-parameter distribution. Estimating the relative broadening width between the two forests shows that the \heii median b-parameter is about 0.8 times the median \hi broadening width. This implies that the \heii absorbers are physically extended consistent with conclusions from observed lines of sight.

We use a ΛCDM hydrodynamic cosmological simulation in a 9.6 comoving Mpc box to examine the evolution of the intergalactic medium transmissivity from z = 2.5 up to the epoch of reionization (z ≳ 6.5). Ionization is due to a uniform ultraviolet background from an evolving mix of stellar and QSO populations. We synthesize noiseless H I Lyα absorption spectra by casting lines of sight through our continuously evolving box and analyze the statistical properties of the transmitted flux. We find a smooth evolution of the effective optical depth of the form τeff ∝ 4.16±0.02 up to reionization. During reionization, the mean transmission and variance sharply deviate from an extrapolated smooth evolution. This introduces large margins of error that would require an unobtainable number of lines of sight to minimize. Subsampling in "high" and "low" transmission lines of sight results in either inferring the continuation of a smooth profile or underestimating the global transmissivity during reionization, respectively. Nonetheless, the end of an opacity phase transition of the intergalactic medium correlates well with the redshift when both the mean and variance sharply deviate from a smooth profile. The lower bound estimates of the flux variance in this work also suggest that it is a more sensitive tool than the mean transmission in mapping reionization. In addition to optical transmission, we derive the "dark gap" length distribution, which is in agreement with observations. We also find a positive correlation between the mean gap optical depth and the gap length.

We use high-quality echelle spectra of 24 quasi-stellar objects to provide a calibrated measurement of the total amount of Ly alpha forest absorption (DA) over the redshift range 2.2 < z < 3.2. Our measurement of DA excludes absorption from metal lines or the Ly alpha lines of Lyman-limit systems and damped Ly alpha systems. We use artificial spectra with realistic flux calibration errors to show that we are able to place continuum levels that are accurate to better than 1 per cent. When we combine our results with our previous results between 1.6 < z < 2.2, we find that the redshift evolution of DA is well described over f1.6 < z < 3.2 as A (1 +z)(gamma), where A = 0.0062 and gamma = 2.75. We detect no significant deviations from a smooth power-law evolution over the redshift range studied. We find less H i absorption than expected at z = 3, implying that the ultraviolet background is similar to 40 per cent higher than expected. Our data appears to be consistent with an H i ionization rate of Gamma similar to 1.4 x 10(-12) s(-1).

We present 40 fully hydrodynamical numerical simulations of the intergalactic gas that gives rise to the Ly alpha forest. The simulation code, input and output files are available at http://www.cosmos.ucsd.edu/similar to gso/index.html. For each simulation we predict the observable properties of the HI absorption in quasar or quasi-stellar object (QSO) spectra. We then find the sets of cosmological and astrophysical parameters which result in spectra whose properties match that of the QSO spectra. We present our results as scaling relationships between input and output parameters. The input parameters include the main cosmological parameters Omega(b), Omega(m), Omega(Lambda), H-0 and sigma(8); and two astrophysical parameters gamma(912) and X-228. The parameter. 912 controls the rate of ionization of H I, He I and He II and is equivalent to the intensity of the ultraviolet background (UVB). The second parameter X-228 controls the rate of heating from the photoionization of He II and can be related to the shape of the UVB at lambda < 228 angstrom. We show how these input parameters, especially s8,. 912 and X-228, affect the output parameters that we measure in simulated spectra. These parameters are the mean flux (F) over bar F, a measure of the most common ( as defined more precisely in Section 5.1) Ly alpha linewidth (b-value) b(sigma), and the one-dimensional power spectrum of the flux on scales from 0.01 to 0.1 s km(-1). We compare the simulation output with data from Kim et al. and Tytler et al., and we give a new measurement of the flux power from HIRES and UVES spectra for the low-density intergalactic medium (IGM) alone at z = 1.95.We find that simulations with a wide variety of sigma(8)-values, from at least 0.8 to 1.1, can fit the small-scale flux power and b-values when we adjust X-228 to compensate for the s 8 change. We can also use. 912 to adjust the HI ionization rate to match the mean flux simultaneously. When we examine only the mean flux, b-values and small-scale flux power we cannot readily break the strong degeneracy between s 8 and X-228.We can break the degeneracy using large-scale flux power or other data to fix s8. When we pick a specific sigma(8)-value the simulations give the value of X-228 that we need to match the observed small-scale flux power and b-values. We can then also find the gamma(912) required to match the mean flux for that combination of s 8 and X-228. We derive scaling relations that give the output parameter values expected for a variety of input parameters. We predict the linewidth parameter b(sigma) with an error of 1.4 per cent and the mean amount of HI absorption to 2 per cent, equivalent to a 0.27 per cent error on (F) over bar at z = 1.95. These errors are four times smaller than those on the best current measurement. We can readily calculate the sets of input parameters that give outputs that match the data. For sigma(8) = 0.9, with Omega(b) = 0.044, Omega(m) = 0.27, Omega(Lambda) = 0.73, h = 0.71 and n = 1.0, we find X-228 = 1.26 and gamma(912) = 1.00, equivalent to Gamma(912) = 1.33 x 10(-12) ionizations per H I atom per second. If we run an optically thin simulation with these parameters in a box size of 76.8 Mpc comoving and with a cell size of 18.75 kpc comoving, we expect that the simulated spectra will match Lya forest data at z = 1.95. The rates predicted by Madau, Haardt & Rees correspond to gamma(912) = 1 and X-228 = 1. Our results for gamma(912) match while the larger X-228 is reasonable to correct for the opacity that is missing from the optically thin simulations. For a smaller value of sigma(8) the structures are generally more extended and we need a smaller X-228 corresponding to a cooler IGM, as found by Bryan & Machacek ( their fig. 7). We also need a larger gamma(912) to stop the neutral fraction from increasing at the lower temperatures.

We identify a concordant model for the intergalactic medium (IGM) at redshift z = 1.9 that uses popular values for cosmological and astrophysical parameters and accounts for all baryons with an uncertainty of 5%. The amount of absorption by H I in the IGM provides the best evidence on the physical conditions in the IGM, especially the combination of the mean gas density, the density fluctuations, the intensity of the ionizing flux, and the level of ionization. We have measured the amount of absorption, known as the flux decrement (DA), in the Lyα forest at redshift 1.9. We used spectra of 77 QSOs that we obtained with 250 km s-1 resolution from the Kast spectrograph on the Lick observatory 3 m telescope. We fitted unabsorbed continua to these spectra using b-splines. We also fitted equivalent continua to 77 artificial spectra that we made to match the real spectra in most obvious ways: redshift, resolution, signal-to-noise ratio (S/N), emission lines and absorption lines. The typical relative error in our continuum fits to the artificial spectra is 3.5%. Averaged over all 77 QSOs, the mean level is within 1%-2% of the correct value, except at S/N < 6, where we systematically placed the continuum too high. We then adjusted the continua on the real spectra to remove this bias as a function of S/N and a second smaller bias. Absorption from all lines in the Lyα forest at z = 1.9 removes DA(z = 1.9) = 15.1% ± 0.7% of the flux at rest-frame wavelengths 1070 < λr < 1170 Å. This is the first measurement using many QSOs at this z, and the first calibrated measurement at any redshift. Using similar methods on 1225 < λr < 1500 Å, we find metal lines absorb an average 1.6% the flux, increasing slightly as the rest-frame wavelength λr decreases because more types of spectral lines contribute and there is more C IV at lower redshifts. We estimate that the metal lines absorb 2.3% ± 0.5% of the flux in the Lyα forest at z = 1.9. The absorption from Lyα alone then has DA = 12.8% ± 0.9%. The Lyα lines of Lyman limit systems with H I column densities log N > 17.2 cm-2 are responsible for a DA = 1.0% ± 0.4% at z = 1.9. These lines arise in higher density regions than the bulk of the IGM Lyα absorption, and hence they are harder to simulate in the huge boxes required to represent the large-scale variations in the IGM. If we subtract these lines, for comparison with simulations of the lower density bulk of the IGM, we are left with DA = 11.8% ± 1.0%. The mean DA in segments of individual spectra with Δz = 0.1, or 153 Mpc comoving at z = 1.9, has a large dispersion, σ = 6.1% ± 0.3% including Lyman limit systems (LLSs) and metal lines, and σ(Δz = 0.1) = 3.9% for the Lyα from the lower density IGM alone, excluding LLSs and metal lines. This is consistent with the usual description of large-scale structure and accounts for the large variations from QSO to QSO. Although the absorption at z = 1.9 is mostly from the lower density IGM, the Lyα of LLSs and the metal lines are both major contributors to the variation in the mean flux on 153 Mpc scales at z = 1.9, and they make the flux field significantly different from a random Gaussian field with an enhanced probability of a large amount of absorption. We find that a hydrodynamic simulation on a 10243 grid in a 75.7 Mpc box reproduces the observed DA from the low-density IGM alone when we use popular parameters values H0 = 71 km s-1 Mpc-1, Ωb = 0.044, Ωm = 0.27, ΩΛ = 0.73, σ8 = 0.9, and an ultraviolet background (UVB) that has an ionization rate per H I atom of Γ912 = (1.44 ± 0.11) × 10-12 s-1. This is 1.08 ± 0.08 times the prediction by Madau et al. with 61% from QSOs and 39% from stars. Our measurement of the DA gives a new joint constraint on these parameters, and the DA is very sensitive to each parameter. Given fixed values for all other parameters, and assuming the simulation has insignificant errors, the error of our DA measurement gives an error on H0 of 10%, ΩΛ of 6%, Ωb of 5%, and σ8 of 4%, comparable to the best measurements by other methods.

We examine the dynamical evolution and statistical properties of the supernova ejecta of massive primordial stars in a cosmological framework to determine whether this first population of stars could have enriched the universe to the levels and dispersions seen by the most recent observations of the Lyalpha forest. We evolve a Lambda cold dark matter model in a 1 Mpc3 volume to a redshift of z = 15 and add "bubbles" of metal corresponding to the supernova ejecta of the first generation of massive stars in all dark matter halos with masses greater than 5 x 10(5) M-.. These initial conditions are then evolved to z = 3, and the distribution and levels of metals are compared to observations. In the absence of further star formation, the primordial metal is initially contained in halos and filaments. Photoevaporation of metal-enriched gas due to the metagalactic ultraviolet background radiation at the epoch of reionization (z similar to 6) causes a sharp increase of the metal volume filling factor. At z = 3, similar to 2.5% of the simulation volume (approximate to20% of the total gas mass) is filled with gas enriched above a metallicity of 10(-4) Z(.), and less than 0.6% of the volume is enriched above a metallicity of 10(-3) Z(.). This suggests that even with the most optimistic prescription for placement of primordial supernovae and the amount of metals produced by each supernova, this population of stars cannot entirely be responsible for the enrichment of the Lyalpha forest to the levels and dispersions seen by current observations unless we have severely underestimated the duration of the Population III epoch. However, comparison to observations using carbon as a tracer of metals shows that Population III supernovae can be significant contributors to the very low overdensity Lyalpha forest.

We identify a concordant model for the intergalactic medium (IGM) at redshift z=1.9 that uses popular values for cosmological and astrophysical parameters and accounts for all baryons with an uncertainty of 6%. We have measured the amount of absorption, DA, in the Ly-alpha forest at redshift 1.9 in spectra of 77 QSO from the Kast spectrograph. We calibrated the continuum fits with realistic artificial spectra, and we found that averaged over all 77 QSOs the mean continuum level is within 1-2% of the correct value. Absorption from all lines in the Ly-alpha forest at z=1.9 removes DA=15.1 +/- 0.7% of the flux between 1070 and 1170 (rest) Angstroms. This is the first measurement using many QSOs at this z, and the first calibrated measurement at any redshift. Metal lines absorb 2.3 +/- 0.5%, and LLS absorb 1.0 +/- 0.4% leaving 11.8 +/- 1.0% from the lower density bulk of the IGM. Averaging over Delta z=0.1 or 154 Mpc, the dispersion is 6.1 +/- 0.3% including LLS and metal lines, or 3.9 (+0.5, -0.7)% for the lower density IGM alone, consistent with the usual description of large scale structure. LLS and metal lines are major contributors to the variation in the mean flux, and they make the flux field significantly non-Gaussian. We find that a hydrodynamic simulation on a 1024 cubed grid in a 75.7 Mpc box reproduces the observed DA from the low density IGM with parameters values H_o=71 km/s/Mpc, Omega_Lambda=0.73, Omega_m=0.27, Omega_b=0.044, sigma_8=0.9 and a UV background that has an ionization rate that is 1.08 +/- 0.08 times the prediction by Madau, Haardt & Rees (1999).

We identify a concordant model for the intergalactic medium (IGM) at redshift z = 1.9 that uses popular values for cosmological and astrophysical parameters and accounts for all baryons with an uncertainty of 6%. The amount of absorption by H I in the IGM provides the best evidence on the physical conditions in the IGM, especially the combination of the mean gas density, the density fluctuations, the intensity of the ionizing flux, and the level of ionization. We have measured the amount of absorption, known as the flux decrement, DA, in the Lyα forest at redshift 1.9. We used spectra of 77 QSO that we obtained with 250 km s resolution from the Kast spectrograph on the Lick observatory 3m telescope. We fit the unabsorbed continua to these spectra using b-splines. We also fit equivalent continua to 77 artificial spectra that we made to match the real spectra in all obvious ways: redshift, resolution, S/N, emission lines and absorption lines. The typical relative error in our continuum fits to the artificial spectra is 3.5%. Averaged over all 77 QSOs the mean level is within 1–2% of the correct value, except at S/N < 6 where we systematically placed the continuum too high. We then adjusted the continua on the real spectra to remove this bias as a function of S/N and a second smaller bias. Absorption from all lines in the Lyα forest at z = 1.9 removes DA(z=1.9) = 15.1± 0.7% of the flux at rest frame wavelengths 1070 < λr < 1170 Å. This is the first measurement using many QSOs at this z, and the first calibrated measurement at any redshift. Using similar methods on 1225 < λr < 1500 Å we find metal lines absorb an average 1.6% the flux, increasing slightly as the rest frame wavelength λr decreases because Based on data obtained with the Kast spectrograph on the Lick Observatory 3-m Shane telescope. E-mail: tytler at ucsd.edu