Search

We have developed a new fully automated Artificial Intelligence (AI)-based method for deriving optimal models of complex absorption systems. The AI structure is built around VPFIT, a well-developed and extensively tested nonlinear least-squares code. The new method forms a sophisticated parallelized system, eliminating human decision-making and hence bias. Here, we describe the workings of such a system and apply it to synthetic spectra, in doing so establishing recommended methodologies for future analyses of Very Large Telescope (VLT) and Extremely Large Telescope (ELT) data. One important result is that modelling line broadening for high-redshift absorption components should include both thermal and turbulent components. Failing to do so means it is easy to derive the wrong model and hence incorrect parameter estimates. One topical application of our method concerns searches for spatial or temporal variations in fundamental constants. This subject is one of the key science drivers for the European Southern Observatory's ESPRESSO spectrograph on the VLT and for the HIRES spectrograph on the ELT. The quality of new data demands completely objective and reproducible methods. The Monte Carlo aspects of the new method described here reveal that model non-uniqueness can be significant, indicating that it is unrealistic to expect to derive an unambiguous estimate of the fine structure constant α from one or a very small number of measurements. No matter how optimal the modelling method, it is a fundamental requirement to use a large sample of measurements to meaningfully constrain temporal or spatial α variation. [ABSTRACT FROM AUTHOR]

We present a new, uniform analysis of the H  i transmitted flux (F) and H  i column density (⁠|$N_{\mathrm{H\,{\small I}}}$|⁠) distribution in the low-density IGM as a function of redshift z for 0 <  z  < 3.6 using 55 HST /COS FUV (Δ z  = 7.2 at z  < 0.5), five HST /STIS + COS NUV (Δ z  = 1.3 at z  ∼ 1) and 24 VLT/UVES, and Keck/HIRES (Δ z  = 11.6 at 1.7 <  z  < 3.6) AGN spectra. We performed a consistent, uniform Voigt profile analysis to combine spectra taken with different instruments, to reduce systematics and to remove metal-line contamination. We confirm previously known conclusions on firmer quantitative grounds in particular by improving the measurements at z  ∼ 1. Two flux statistics at 0 <  F  < 1, the mean H  i flux and the flux probability distribution function (PDF), show that considerable evolution occurs from z  = 3.6 to z  = 1.5, after which it slows down to become effectively stable for z  < 0.5. However, there are large sightline variations. For the H  i column density distribution function (CDDF, f  ∝  |$N_{\rm H\,{\small I}}^{-\beta }$|⁠) at |$\log (N_{\mathrm{H\,{\small I}}}/1\, {\mathrm{cm}^{-2}})$|  ∈ [13.5, 16.0], β increases as z decreases from β = 1.60 at z  ∼ 3.4 to β = 1.82 at z ∼ 0.1. The CDDF shape at lower redshifts can be reproduced by a small amount of clockwise rotation of a higher- z CDDF with a slightly larger CDDF normalization. The absorption line number per z (d n /d z) shows a similar evolutionary break at z ∼ 1.5 as seen in the flux statistics. High- |$N_{\mathrm{H\,{\small I}}}$| absorbers evolve more rapidly than low- |$N_{\mathrm{H\,{\small I}}}$| absorbers to decrease in number or cross-section with time. The individual d n /d z shows a large scatter at a given z. The scatter increases towards lower z , possibly caused by a stronger clustering at lower z. [ABSTRACT FROM AUTHOR]

For around 100 yr, hydrogen spectral modelling has been based on Voigt profile fitting. The semiclassical Voigt profile is based on a two-level atom approximation. Whilst the Voigt profile is excellent for many circumstances, the accuracy is insufficient for very high column density damped Lyman-α absorption systems. We have adapted the quantum-mechanical Kramers–Heisenberg model to include thermal broadening, producing a new profile, the KHT profile. Interactions involving multiple discrete atomic levels and continuum terms, not accounted for in the Voigt model, generate asymmetries in the Lyman line wings. If not modelled, this can lead to significant systematics in parameter estimation when modelling real data. There are important ramifications in particular for measurements of the primordial deuterium abundance. However, the KHT model is complicated. We therefore present a simplified formulation based on Taylor series expansions and look-up tables, quantifying the impact of the approximations made. The KHT profile has been implemented within the widely used vpfit code. [ABSTRACT FROM AUTHOR]

The primordial deuterium abundance probes fundamental physics during the big bang nucleosynthesis and can be used to infer cosmological parameters. Observationally, the abundance can be measured using absorbing clouds along the lines of sight to distant quasars. Observations of the quasar PKS1937–101 contain two absorbers for which the deuterium abundance has previously been determined. Here, we focus on the higher redshift one at zabs = 3.572. We present new observations with significantly increased signal-to-noise ratio that enable a far more precise and robust measurement of the deuterium to hydrogen column density ratio, resulting in DI/H I = 2.62 ± 0.05 × 10-5. This particular measurement is of interest because it is amongst the most precise assessments to date and it has been derived from the second lowest column-density absorber [N(H I) = 17.9 cm-2] that has so-far been utilized for deuterium abundance measurements. The majority of existing high-precision measurements were obtained from considerably higher column density systems [i.e. N(H I) > 19.4 cm-2]. This bodes well for future observations as low column density systems are more common. [ABSTRACT FROM AUTHOR]

At low densities, the standard ionization history of the intergalactic medium (IGM) predicts a decreasing temperature of the IGM with decreasing density once hydrogen (and helium) reionization is complete. Heating the high-redshift, low-density IGM above the temperature expected from photoheating is difficult, and previous claims of high/rising temperatures in lowdensity regions of the Universe based on the probability density function (PDF) of the opacity in Ly α forest data at 2 < z < 4 have been met with considerable scepticism, particularly since they appear to be in tension with other constraints on the temperature-density relation (TDR). We utilize here an ultrahigh signal-to-noise spectrum of the Quasi-stellar object HE0940-1050 and a novel technique to study the low opacity part of the PDF. We show that there is indeed evidence (at 90 per cent confidence level) that a significant volume fraction of the underdense regions at z ~ 3 has temperatures as high or higher than those at densities comparable to the mean and above. We further demonstrate that this conclusion is nevertheless consistent with measurements of a slope of the TDR in overdense regions that imply a decreasing temperature with decreasing density, as expected if photoheating of ionized hydrogen is the dominant heating process.We briefly discuss implications of our findings for the need to invoke either spatial temperature fluctuations, as expected during helium reionization, or additional processes that heat a significant volume fraction of the low-density IGM. [ABSTRACT FROM AUTHOR]

In this work, we investigate the abundance and distribution of metals in the intergalactic medium (IGM) at 〈z〉 ≃ 2.8 through the analysis of an ultra-high signal-to-noise ratio UVES spectrum of the quasar HE0940-1050. In the CIV forest, our deep spectrum is sensitive at 3σ to lines with column density down to log NCIV ≃ 11.4 and in 60 per cent of the considered redshift range down to ≃11.1. In our sample, all H I lines with log NHI ≥ 14.8 show an associated C IV absorption. In the range 14.0 ≤ log NHI < 14.8, 43 per cent of H I lines has an associated C IV absorption. At log NHI < 14.0, the detection rates drop to <10 per cent, possibly due to our sensitivity limits and not to an actual variation of the gas abundance properties. In the range log NHI ≥ 14, we observe a fraction of H I lines with detected C IV a factor of 2 larger than the fraction of H I lines lying in the circumgalactic medium (CGM) of relatively bright Lyman-break galaxies hosted by dark matter haloes with 〈M〉 ~ 1012 M☉. The comparison of our results with the output of a grid of photoionization models and of two cosmological simulations implies that the volume filling factor of the IGM gas enriched to a metallicity logZ/Z☉≳-3log⁡Z/Z☉≳-3 should be of the order of ~10-13 per cent. In conclusion, our results favour a scenario in which metals are found also outside the CGM of bright star-forming galaxies, possibly due to pollution by lower mass objects and/or to an early enrichment by the first sources. [ABSTRACT FROM AUTHOR]

From the Voigt profile fitting analysis of 183 intervening CIVCIV systems at 1.7 < z < 3.3 in 23 high-quality UVES/VLT and HIRES/Keck QSO spectra, we find that a majority of CIVCIV systems (~75 per cent) display a well-characterized scaling relation between integrated column densities of HIHI and CIVCIV with a negligible redshift evolution, when column densities of all the HIHI and CIVCIV components are integrated within a given ±150?km?s-1 range centred at the CIVCIV flux minimum. The integrated CIVCIV column density NCIV,sys increases with NHI,sys at logNHI,sys∈[14,16] and logNCIV,sys∈[11.8,14.0], then becomes almost independent of NHI,sys at logNHI,sys=16, with a large scatter: at logNHI,sys∈[14,22], logNCIV,sys=[C1/logNHI,sys+C2]+C3, with C1 = -1.90 ± 0.55, C2 = -14.11 ± 0.19 and C3 = 14.76 ± 0.17, respectively. The steep (flat) part is dominated by SiIVSiIV-free (SiIVSiIV-enriched) CIVCIV systems. Extrapolating the NHI,sys-NCIV,sys relation implies that most absorbers with logNHI≤14 are virtually CIVCIV-free. The NHI,sys-NCIV,sys relation does not hold for individual components, clumps or the integrated velocity range less than ±100?km?s-1. This is expected if the line-of-sight extent of CIVCIV is smaller than HIHI and NCIV,sys decreases more rapidly than NHI,sys at the larger impact parameter, regardless of the location of the HIHI+CIVCIV gas in intergalactic medium filaments or in intervening galactic haloes. [ABSTRACT FROM AUTHOR]

We present the Voigt profile analysis of 132 intervening CIV+CIII components associated with optically thinHI absorbers at 2.1= 1.0 ± 0.3 with a negligible redshift evolution. For 54 CIV components tied (aligned) with HI at logNHI [12.2, 16.0] and logNCIV [11.8, 13.8], the gas temperature Tb estimated from absorption line widths is well approximated to a Gaussian peaking at log Tb ~ 4.4 ± 0.3 for log Tb [3.5, 5.5], with a negligible non-thermal contribution. For 32 of 54 tied HI+CIV pairs, also tied with CIII at logNCIII [11.7, 13.8], we ran both photoionization equilibrium (PIE) and non-PIE (using a fixed temperature Tb) CLOUDY models for the Haardt-Madau QSOs+galaxies 2012 UV background. We find evidence of bimodality in observed and derived physical properties. High-metallicity branch absorbers have a carbon abundance [C/H]temp =-1.0, a line-of-sight length Ltemp = 20 kpc and a total (neutral and ionized) hydrogen volume density log nH,temp [-4.5, -3.3] and log Tb [3.9, 4.5]. Low-metallicity branch absorbers have [C/H]temp = -1.0, Ltemp [20, 480] kpc and log nH,temp [-5.2, -4.3] and log Tb ∼ 4.5. High-metallicity branch absorbers seem to be originated from extended discs, inner haloes or outflowing gas of intervening galaxies, while low-metallicity absorbers are produced by galactic haloes or the surrounding intergalactic medium filament. [ABSTRACT FROM AUTHOR]

We have investigated the distribution and evolution of ~3100 intergalactic neutral hydrogen (HI) absorbers with H i column densities log  NHI = [12.75,17.0] at 1.9 < z < 3.2, using 18 high resolution, high signal-to-noise quasar spectra obtained from the ESO VLT/UVES archive. We used two sets of Voigt profile fitting analysis, one including all the available high-order Lyman lines to obtain reliable HI column densities of saturated lines, and another using only the Lyα transition. There is no significant difference between the Lyα-only fit and the high-order Lyman fit results. Combining our Lyα-only fit results at 1.7 < z < 3.6 with high-quality literature data, the mean number density at 0 < z < 4 is not well described by a single power law and strongly suggests that its evolution slows down at z ≤ 1.5 at the high and low column density ranges. We also divided our entire HI absorbers at 1.9 < z < 3.2 into two samples, the unenriched forest and the CIV-enriched forest, depending on whether HI lines are associated with CIV at log NCIV ≥ 12.2 within a given velocity range. The entire HI column density distribution function (CDDF) can be described as the combination of these two well-characterised populations which overlap at log NHI ~ 15. At log NHI ≤ 15, the unenriched forest dominates, showing a similar power-law distribution to the entire forest. The CIV-enriched forest dominates at log NHI ≥ 15, with its distribution function as ∝NHI~-1.45. However, it starts to flatten out at lower NHI, since the enriched forest fraction decreases with decreasing NHI. The deviation from the power law at log NHI = [14,17] shown in the CDDF for the entire HI sample is a result of combining two different HI populations with a different CDDF shape. The total HI mass density relative to the critical density is ΩHI ~ 1.6 × 10-6 h-1, where the enriched forest accounts for ~40% of ΩHI. [ABSTRACT FROM AUTHOR]

We have investigated the distribution and evolution of ~3100 intergalactic neutral hydrogen (HI) absorbers with HI column densities log NHI = [12.75, 17.0] at 1.9 < z < 3.2, using 18 high resolution, high signal-to-noise quasar spectra obtained from the ESO VLT/UVES archive. We used two sets of Voigt profile fitting analysis, one including all the available high-order Lyman lines to obtain reliable HI column densities of saturated lines, and another using only the Lyα transition. There is no significant difference between the Lyα-only fit and the high-order Lyman fit results. Combining our Lyα-only fit results at 1.7 < z < 3.6 with high-quality literature data, the mean number density at 0 < z < 4 is not well described by a single power law and strongly suggests that its evolution slows down at z ⩽ 1.5 at the high and low column density ranges.We also divided our entire HI absorbers at 1.9 < z < 3.2 into two samples, the unenriched forest and the CIV-enriched forest, depending on whether HI lines are associated with CIV at log NCIV ⩾ 12.2 within a given velocity range. The entire HI column density distribution function (CDDF) can be described as the combination of these two well-characterised populations which overlap at log NHI ~ 15. At log NHI ⩽ 15, the unenriched forest dominates, showing a similar power-law distribution to the entire forest. The CIV-enriched forest dominates at log NHI ⩾ 15, with its distribution function as ∝N~-1.45 HI . However, it starts to flatten out at lower NHI, since the enriched forest fraction decreases with decreasing NHI. The deviation from the power law at log NHI = [14, 17] shown in the CDDF for the entire Hi sample is a result of combining two different Hi populations with a different CDDF shape. The total Hi mass density relative to the critical density is ΩHI ~ 1.6 × 10-6 h-1, where the enriched forest accounts for ~40% of ΩHI. [ABSTRACT FROM AUTHOR]

ABSTRACT By comparing the widths of absorption lines from O i, Si ii and Fe ii in the redshift zabs= 2.076 single-component damped Lyα system in the spectrum of Q2206−199 we establish that these absorption lines arise in warm neutral medium (WNM) gas at ∼12 000 ± 3000 K. This is consistent with thermal equilibrium model estimates of ∼8000 K for the WNM in galaxies, but not with the presence of a significant cold component. It is also consistent with, but not required by, the absence of C ii fine structure absorption in this system. Some possible implications concerning abundance estimates in narrow-line WNM absorbers are discussed. [ABSTRACT FROM AUTHOR]

Molecular transitions recently discovered at redshift towards the bright background quasar are analysed to limit cosmological variation in the proton-to-electron mass ratio, . Observed with the Keck telescope, the optical echelle spectrum has the highest resolving power and largest number (86) of H2 transitions in such analyses so far. Also, (seven) HD transitions are used for the first time to constrain μ-variation. These factors, and an analysis employing the fewest possible free parameters, strongly constrain μ's relative deviation from the current laboratory value: , indicating an insignificantly larger μ in the absorber. This is the first Keck result to complement recent null constraints from three systems at observed with the Very Large Telescope. The main possible systematic errors stem from wavelength calibration uncertainties. In particular, distortions in the wavelength solution on echelle order scales are estimated to contribute approximately half the total systematic error component, but our estimate is model dependent and may therefore under or overestimate the real effect, if present. To assist future μ-variation analyses of this kind, and other astrophysical studies of H2 in general, we provide a compilation of the most precise laboratory wavelengths and calculated parameters important for absorption-line work with H2 transitions redwards of the hydrogen Lyman limit. [ABSTRACT FROM AUTHOR]

We compare the improved measurement of the Lyα forest flux probability distribution at 1.7 < z < 3.2 presented by Kim et al. to a large set of hydrodynamical simulations of the Lyα forest with different cosmological parameters and thermal histories. The simulations are in good agreement with the observational data if the temperature–density relation for the low-density intergalactic medium (IGM), , is either close to isothermal or inverted (γ < 1). Our results suggest that the voids in the IGM may be significantly hotter and the thermal state of the low-density IGM may be substantially more complex than is usually assumed at these redshifts. We discuss radiative transfer effects which alter the spectral shape of ionizing radiation during the epoch of He ii reionization as a possible physical mechanism for achieving an inverted temperature–density relation at z≃ 3. [ABSTRACT FROM AUTHOR]

We have performed an extensive Voigt profile analysis of the neutral hydrogen (H i) and metal absorption present in a sample of 18 high-resolution, high signal-to-noise ratio quasi-stellar object (QSO) spectra observed with the Very Large Telescope Ultraviolet and Visual Echelle Spectrograph. We use this analysis to separate the metal contribution from the H i absorption and present an improved measurement of the flux probability distribution function (PDF) due to H i absorption alone at , 2.52 and 2.94. The flux PDF is sensitive to the continuum fit in the normalized flux range 0.8 < F < 1.0 and to metal absorption at 0.2 < F < 0.8. Our new measurements of the flux PDF due to H i absorption alone are systematically lower at 0.2 < F < 0.8 by up to 30 per cent compared to the widely used measurement of McDonald et al., based on a significantly smaller sample of Keck High Resolution Echelle Spectrometer data. This discrepancy is probably due to a combination of our improved removal of the metal absorption and cosmic variance, since variations in the flux PDF between different lines-of-sight are large. The H i effective optical depth τeffH i at 1.7 < z < 4 is best fit with a single power law, , in good agreement with previous measurements from comparable data. As also found previously, the effect of noise on the flux distribution is not significant in high-resolution, high signal-to-noise ratio data. [ABSTRACT FROM AUTHOR]

We report on a further analysis of the ratio of deuterium to hydrogen (D/H) using Hubble Space Telescope ( HST) spectra of the Lyman limit system towards the quasi-stellar object (QSO) PG1718+481. Initial analyses of this absorber found it gave a high D/H value, 1.8– , inconsistent with several higher redshift measurements. It is thus important to critically examine this measurement. By analysing the velocity widths of the D i, H i and metal lines present in this system, Kirkman et al. report that the additional absorption in the blue wing of the Lyα line cannot be D i, with a confidence level of 98 per cent. Here we present a more detailed analysis, taking into account possible wavelength shifts between the three sets of HST spectra used in the analysis. We find that the constraints on this system are not as strong as those claimed by Kirkman et al. The discrepancy between the parameters of the blue wing absorption and the parameters expected for D i is marginally worse than 1σ. Tytler et al. commented on the first analysis of Webb et al., reporting the presence of a contaminating lower redshift Lyman limit system, with at , which biases the N(H i) estimate for the main system. Here we show that this absorber actually has and does not impact on the estimate of N(H i) in the system of interest at . The purpose of the present paper is to highlight important aspects of the analysis which were not explored in previous studies, and hence to help refine the methods used in future analyses of D/H in quasar spectra. [ABSTRACT FROM AUTHOR]

Abstract Combining a new, increased data set of eight quasi-stellar objects (QSOs) covering the Lyα forest at redshifts 1.5 > z < 3.6 from VLT/UVES observations with previously published results, we have investigated the properties of the Lyα forest at 1.5 < z <4 . With the six QSOs covering the Lyα forest at 1.5 < z < 2.5 , we have extended previous studies in this redshift range. In particular, we have concentrated on the evolution of the line number density and the clustering of the Lyα forest at z≤ 2.5 , where the Lyα forest starts to show some inhomogeneity from sightline to sightline. We have fitted Voigt profiles to the Lyα absorption lines as in previous studies, and have, for two QSOs with z[sub em]∼ 2.4 , fitted Lyα and higher order of Lyman lines down to 3050 Å simultaneously. This latter approach has been taken in order to study the Lyβ forest at z∼ 2.2 and the higher H i column density Lyα forest in the Lyβ forest region. For a given N[sub H I] range, the Lyα forest at 1.5 < z < 4 shows the monotonic evolution, which is governed mainly by the Hubble expansion at this redshift range. In general, the Lyα forest line number density (dn/dz) is best approximated with dn/dz= 6.1(1 +z)[sup 2.47 ± 0.18] for the H i column density N[sub H I]= 10[sup 13.64-17] cm[sup -2] at 1.5 < z < 4 . When the results at 0 < z < 1.5 from Hubble Space Telescope (HST) observations are combined, the slow-down in the number density evolution occurs at z < 1.5 . For higher column density clouds at N[sub Hi] > 10[sup 14] cm[sup -2] , there is a variation in the line number density from sightline to sightline at z < 2.5 . This variation is stronger for higher column density systems, probably due to more gravitationally evolved structures at lower z. The mean H i opacity is at 1.5 < z < 4 . HST observations show evidence for slower evolution of at z < 1 . For N[sub Hi]= 10[sup 12.5–15] cm[sup -2] , the differential column density distribution function, f(N[sub Hi]) , can be best fitted by f(N[sub HI]∝N[sub Hi][sup -β] with β≈ 1.5 for 1.5 < z < 4 . When combined with HST observations, the exponent β increases as z decreases at 0 < z < 4 for N[sub Hi]= 10[sup 13–17] cm[sup -2] . The correlation strength of the step optical depth correlation function shows the strong evolution from = 3.3 to = 2.1 , although there is a large scatter along different sightlines. The analyses of the Lyβ forest at z∼ 2.2 are, in general, in good agreement with those of the Lyα forest. [ABSTRACT FROM AUTHOR]

ABSTRACTWe report on high-resolution observations (R=35 000) of the Hubble Deep Field South QSO J2233-606 obtained with the University College London Echelle Spectrograph at the Anglo-Australian Telescope. We present spectral data for the wavelength region 3530 < λ <4390 Å. The data have a mean signal-to-noise ratio in the continuum of approximately 15 per 0.05-Å resolution element at λ > 3700 Å. Voigt profiles are fitted to determine the redshift and column densities of the observed features. Eight heavy-element intervening systems are identified, as well as an associated system showing evidence for incomplete coverage of the background source. [ABSTRACT FROM AUTHOR]

Spectroscopic observations of distant quasars have resulted in the detection of molecular hydrogen in intervening damped Lyman α absorption clouds (DLAs). We use observations compiled from different experimental groups to show that the molecular hydrogen abundance exhibits a dramatic increase over a cosmological time period corresponding to 13 to 24 per cent of the age of the Universe. We also tentatively show that the heavy-element abundances in the same gas clouds exhibit a faster and more well-defined cosmological evolution compared with the general DLA population over the same time baseline. We argue that this latter point is unsurprising, because the general DLA population arises in a wide variety of galaxy types and environments, and thus a spans broad range of ISM gas-phases and abundances at the same cosmic time. DLAs exhibiting H2 absorption may therefore circumvent this problem, efficiently identifying a narrower class of objects, and provide a more sensitive probe of cosmological chemical evolution. [ABSTRACT FROM AUTHOR]