arXiv subjects

Jason X. Prochaska

Publications and source records attributed to Jason X. Prochaska.

At least 37 records · Page 2Linked to original sources

arXiv2011

The First Observations of Low Redshift Damped Lyman-alpha Systems with the Cosmic Origins Spectrograph

We report on the first Cosmic Origins Spectrograph (COS) observations of damped and sub-damped Lyman-alpha (DLA) systems discovered in a new survey of the gaseous halos of low-redshift galaxies. From observations of 37 sightlines, we have discovered three DLAs and four sub-DLAs. We measure the neutral gas density Omega(HI), and redshift density dN/dz, of DLA and sub-DLA systems at z<0.35. We find dN/dz=0.25 and Omega(HI)=1.4x10^-3 for DLAs, and dN/dz=0.08 with Omega(HI)=4.2x10^-5 for sub-DLAs over a redshift path delta z=11.9. To demonstrate the scientific potential of such systems, we present a detailed analysis of the DLA at z=0.1140 in the spectrum of SDSS J1009+0713. Profile fits to the absorption lines determine log N(H I)=20.68pm0.10 with a metallicity determined from the undepleted element Sulfur of [S/H]=-0.62pm0.18. The abundance pattern of this DLA is similar to that of higher z DLAs, showing mild depletion of the refractory elements Fe and Ti with [S/Fe]=+0.24pm0.22 and [S/Ti]=+0.28pm0.15. Nitrogen is underabundant in this system with [N/H]=-1.40pm0.14, placing this DLA below the plateau of the [N/alpha] measurements in the local Universe at similar metallicities. This DLA has a simple kinematic structure with only two components required to fit the profiles and a kinematic width of 52 km/s. Imaging of the QSO field with WFC3 reveals a spiral galaxy at very small impact parameter to the QSO and several galaxies within 10". Followup spectra with LRIS reveal that none of the nearby galaxies are at the redshift of the DLA. The spiral galaxy is identified as the host galaxy of the QSO based on the near perfect alignment of the nucleus and disk of the galaxy as well as spectra of an H II region showing emission lines at the QSO redshift. A small feature appears 0.70" from the nucleus of the QSO after PSF subtraction, providing another candidate for the host galaxy of the DLA. (abb)

astro-ph.CO↗

arXiv2010

A Mature Dusty Star-forming Galaxy Hosting GRB080607 at z=3.036

We report the discovery of the host galaxy of dark burst GRB080607 at z_GRB=3.036. GRB080607 is a unique case of a highly extinguished (A_V~3 mag) afterglow that was yet sufficiently bright for high-quality absorption-line spectroscopy. The host galaxy is clearly resolved in deep HST WF3/IR F160W images and well detected in the Spitzer IRAC 3.5 micron and 4.5 micron channels, while displaying little/no fluxes in deep optical images from Keck and Magellan. The extremely red optical-infrared colors are consistent with the large extinction seen in the afterglow light, suggesting that the large amount of dust and gas surface mass density seen along the afterglow sightline is not merely local but likely reflects the global dust content across the entire host galaxy. Adopting the dust properties and metallicity of the host ISM derived from studies of early-time afterglow light and absorption-line spectroscopy, we perform a stellar population synthesis analysis of the observed spectral energy distribution to constrain the intrinsic luminosity and stellar population of this dark burst host. The host galaxy is best described by an exponentially declining star formation rate of e-folding time tau=2 Gyr and an age of ~2 Gyr. We also derive an extinction corrected star formation rate of SFR 125 h^{-2} M_sun/yr and a total stellar mass of M_* ~ 4x10^11 h^{-2} M_sun. Our study provides an example of massive, dusty star-forming galaxies contributing to the GRB host galaxy population, supporting the notion that long-duration GRBs trace the bulk of cosmic star formation.

astro-ph.CO↗

arXiv2009

Dust depletion, chemical uniformity and environment of CaII H&K quasar absorbers

CaII 3934,3969 absorbers, which are likely to be a subset of damped Lyman alpha systems, are the most dusty quasar absorbers known, with an order of magnitude more extinction in E(B-V) than other absorption systems. There is also evidence that CaII absorbers trace galaxies with more ongoing star-formation than the average quasar absorber. Despite this, relatively little is known in detail about these unusual absorption systems. Here we present the first high resolution spectroscopic study of 19 CaII quasar absorbers, in the range 0.6<= z_abs<=1.2, with W3934>=0.2A. Their general depletion patterns are similar to measurements in the warm halo phase of the Milky Way and Magellanic Clouds ISM. Dust depletions and alpha-enrichments profiles of sub-samples of 7 and 3 absorbers, respectively, are measured using a combination of Voigt profile fitting and apparent optical depth techniques. Deviations in [Cr/Zn]~0.3+-0.1dex and in [Si/Fe]>~0.8+-0.1dex are detected across the profile of one absorber, which we attribute to differential dust depletion. The remaining absorbers have <0.3dex (3sigma limit) variation in [Cr/Zn], much like the general DLA population, though the dustiest CaII absorbers remain relatively unprobed in our sample. A limit on electron densities in CaII absorbers, n_e<0.1cm^-3, is derived using the ratio of neutral and singly ionised species, assuming a MW-like radiation field. These electron densities may imply hydrogen densities sufficient for the presence of molecular hydrogen in the absorbers. The CaII absorber sample comprises a wide range of velocity widths, v_90=50-470km/s, and velocity structures, thus a range of physical models for their origin, from simple discs to galactic outflows and mergers, would be required to explain the observations.

astro-ph↗

arXiv2008

An 84 microGauss Magnetic Field in a Galaxy at Redshift z=0.692

The magnetic field pervading our Galaxy is a crucial constituent of the interstellar medium: it mediates the dynamics of interstellar clouds, the energy density of cosmic rays, and the formation of stars. The field associated with ionized interstellar gas has been determined through observations of pulsars in our Galaxy. Radio-frequency measurements of pulse dispersion and the rotation of the plane of linear polarization, i.e., Faraday rotation, yield an average value B ~ 3 microGauss. The possible detection of Faraday rotation of linearly polarized photons emitted by high-redshift quasars suggests similar magnetic fields are present in foreground galaxies with redshifts z > 1. As Faraday rotation alone, however, determines neither the magnitude nor the redshift of the magnetic field, the strength of galactic magnetic fields at redshifts z > 0 remains uncertain. Here we report a measurement of a magnetic field of B ~ 84 microGauss in a galaxy at z =0.692, using the same Zeeman-splitting technique that revealed an average value of B = 6 microGauss in the neutral interstellar gas of our Galaxy. This is unexpected, as the leading theory of magnetic field generation, the mean-field dynamo model, predicts large-scale magnetic fields to be weaker in the past rather than stronger.

astro-ph↗

arXiv2008

Ubiquitous outflows in DEEP2 spectra of star-forming galaxies at z=1.4

Galactic winds are a prime suspect for the metal enrichment of the intergalactic medium and may have a strong influence on the chemical evolution of galaxies and the nature of QSO absorption line systems. We use a sample of 1406 galaxy spectra at z~1.4 from the DEEP2 redshift survey to show that blueshifted Mg II 2796, 2803 A absorption is ubiquitous in starforming galaxies at this epoch. This is the first detection of frequent outflowing galactic winds at z~1. The presence and depth of absorption are independent of AGN spectral signatures or galaxy morphology; major mergers are not a prerequisite for driving a galactic wind from massive galaxies. Outflows are found in coadded spectra of galaxies spanning a range of 30x in stellar mass and 10x in star formation rate (SFR), calibrated from K-band and from MIPS IR fluxes. The outflows have column densities of order N_H ~ 10^20 cm^-2 and characteristic velocities of ~ 300-500 km/sec, with absorption seen out to 1000 km/sec in the most massive, highest SFR galaxies. The velocities suggest that the outflowing gas can escape into the IGM and that massive galaxies can produce cosmologically and chemically significant outflows. Both the Mg II equivalent width and the outflow velocity are larger for galaxies of higher stellar mass and SFR, with V_wind ~ SFR^0.3, similar to the scaling in low redshift IR-luminous galaxies. The high frequency of outflows in the star-forming galaxy population at z~1 indicates that galactic winds occur in the progenitors of massive spirals as well as those of ellipticals. The increase of outflow velocity with mass and SFR constrains theoretical models of galaxy evolution that include feedback from galactic winds, and may favor momentum-driven models for the wind physics.

astro-ph↗

arXiv2008

Baryons: What, When and Where?

We review the current state of empirical knowledge of the total budget of baryonic matter in the Universe as observed since the epoch of reionization. Our summary examines on three milestone redshifts since the reionization of H in the IGM, z = 3, 1, and 0, with emphasis on the endpoints. We review the observational techniques used to discover and characterize the phases of baryons. In the spirit of the meeting, the level is aimed at a diverse and non-expert audience and additional attention is given to describe how space missions expected to launch within the next decade will impact this scientific field.

astro-ph↗

arXiv2008

A Survey for NV Absorption at z~z_GRB in GRB Afterglow Spectra: Clues to Gas Near the Progenitor Star

We survey NV absorption in the afterglow spectra of long-duration gamma-ray bursts (GRBs) with the intent to study highly ionized gas in the galaxies hosting these events. We identify a high incidence (6/7) of spectra exhibiting NV gas with z~z_GRB and the majority show large column densities NV > 10^14 cm^-2. With one exception, the observed line-profiles are kinematically `cold', i.e. they are narrow and have small velocity offset (Dv < 20 km/s) from absorption lines associated with neutral gas. In addition, the NV absorption has similar velocity as the UV-pumped fine-structure lines indicating these high ions are located within ~1kpc of the GRB afterglow. These characteristics are unlike those for NV gas detected in the halo/disk of the Milky Way or along sightlines through high z damped Lya systems but resemble the narrow absorption line systems associated with quasars and some high z starbursts. We demonstrate that GRB afterglows photoionize nitrogen to NV at r~10pc. This process can produce NV absorption with characteristics resembling the majority of our sample and and we argue it is the principal mechanism for NV along GRB sightlines. Therefore, the observations provide a snapshot of the physical conditions at this distance. In this scenario, the observations imply the progenitor's stellar wind is confined to r<10pc which suggests the GRB progenitors occur within dense (n > 10^3 cm^-3) environments, typical of molecular clouds. The observations, therefore, primarily constrain the physical conditions -- metallicity, density, velocity fields -- of the gas within the (former) molecular cloud region surrounding the GRB.

astro-ph↗

arXiv2008

Resolving The ISM Surrounding GRBs with Afterglow Spectroscopy

We review current research related to spectroscopy of gamma-ray burst (GRB) afterglows with particular emphasis on the interstellar medium (ISM) of the galaxies hosting these high redshift events. These studies reveal the physical conditions of star-forming galaxies and yield clues to the nature of the GRB progenitor. We offer a pedagogical review of the experimental design and review current results. The majority of sightlines are characterized by large HI column densities, negligible molecular fraction, the ubiquitous detection of UV pumped fine-structure transitions, and metallicities ranging from 1/100 to nearly solar abundance.

astro-ph↗

arXiv2008

The Rapidly Flaring Afterglow of the Very Bright and Energetic GRB 070125

We report on multi-wavelength observations, ranging from the X-ray to radio wave bands, of the IPN-localized gamma-ray burst GRB 070125. Spectroscopic observations reveal the presence of absorption lines due to O I, Si II, and C IV, implying a likely redshift of z = 1.547. The well-sampled light curves, in particular from 0.5 to 4 days after the burst, suggest a jet break at 3.7 days, corresponding to a jet opening angle of ~7.0 degrees, and implying an intrinsic GRB energy in the 1 - 10,000 keV band of around E = (6.3 - 6.9)x 10^(51) erg (based on the fluences measured by the gamma-ray detectors of the IPN network). GRB 070125 is among the brightest afterglows observed to date. The spectral energy distribution implies a host extinction of Av < 0.9 mag. Two rebrightening episodes are observed, one with excellent time coverage, showing an increase in flux of 56% in ~8000 seconds. The evolution of the afterglow light curve is achromatic at all times. Late-time observations of the afterglow do not show evidence for emission from an underlying host galaxy or supernova. Any host galaxy would be subluminous, consistent with current GRB host-galaxy samples. Evidence for strong Mg II absorption features is not found, which is perhaps surprising in view of the relatively high redshift of this burst and the high likelihood for such features along GRB-selected lines of sight.

astro-ph↗

arXiv2008

The Molecular Hydrogen Deficit in Gamma-Ray Burst Afterglows

Recent analysis of five gamma-ray burst (GRB) afterglow spectra reveal the absence of molecular hydrogen absorption lines, a surprising result in light of their large neutral hydrogen column densities and the detection of H$_2$ in similar, more local star-forming regions like 30 Doradus in the Large Magellanic Cloud (LMC). Observational evidence further indicates that the bulk of the neutral hydrogen column in these sight lines lies 100 pc beyond the progenitor and that H$_2$ was absent prior to the burst, suggesting that direct flux from the star, FUV background fields, or both suppressed its formation. We present one-dimensional radiation hydrodynamical models of GRB host galaxy environments, including self-consistent radiative transfer of both ionizing and Lyman-Werner photons, nine-species primordial chemistry with dust formation of H$_2$, and dust extinction of UV photons. We find that a single GRB progenitor is sufficient to ionize neutral hydrogen to distances of 50 - 100 pc but that a galactic Lyman-Werner background is required to dissociate the molecular hydrogen in the ambient ISM. Intensities of 0.1 - 100 times the Galactic mean are necessary to destroy H$_2$ in the cloud, depending on its density and metallicity. The minimum radii at which neutral hydrogen will be found in afterglow spectra is insensitive to the mass of the progenitor or the initial mass function (IMF) of its cluster, if present.

astro-ph↗

arXiv2008

Can gravitational infall energy lead to the observed velocity dispersion in DLAs?

The median observed velocity width v_90 of low-ionization species in damped Ly-alpha systems is close to 90 km/s, with approximately 10% of all systems showing v_90 > 210 km/s at z=3. We show that a relative shortage of such high-velocity neutral gas absorbers in state-of-the-art galaxy formation models is a fundamental problem, present both in grid-based and particle-based numerical simulations. Using a series of numerical simulations of varying resolution and box size to cover a wide range of halo masses, we demonstrate that energy from gravitational infall alone is insufficient to produce the velocity dispersion observed in damped Ly-alpha systems, nor does this dispersion arise from an implementation of star formation and feedback in our highest resolution (~ 45 pc) models, if we do not put any galactic winds into our models by hand. We argue that these numerical experiments highlight the need to separate dynamics of different components of the multiphase interstellar medium at z=3.

astro-ph↗

arXiv2008

Bimodality in Damped Lyman alpha Systems

We report evidence for a bimodality in damped Ly systems (DLAs). Using [C II] 158 mu cooling rates, lc, we find a distribution with peaks at lc=10^-27.4 and 10^-26.6 ergs s^-1 H^-1 separated by a trough at lc^crit ~= lc < 10^-27.0 ergs s^-1 H^-1. We divide the sample into low cool DLAs with lc < lc^crit and high cool DLAs with lc > lc^crit and find the Kolmogorv-Smirnov probabilities that velocity width, metallicity, dust-to-gas ratio, and Si II equivalent width in the two subsamples are drawn from the same parent population are small. All these quantities are significantly larger in the high cool population, while the H I column densities are indistinguishable in the two populations. We find that heating by X-ray and FUV background radiation is insufficient to balance the cooling rates of either population. Rather, the DLA gas is heated by local radiation fields. The rare appearance of faint, extended objects in the Hubble Ultra Deep Field rules out in situ star formation as the dominant star-formation mode for the high cool population, but is compatible with in situ star formation as the dominant mode for the low cool population. Star formation in the high cool DLAs likely arises in Lyman Break galaxies. We investigate whether these properties of DLAs are analogous to the bimodal properties of nearby galaxies. Using Si II equivalent width as a mass indicator, we construct bivariate distributions of metallicity, lc, and areal SFR versus the mass indicators. Tentative evidence is found for correlations and parallel sequences, which suggest similarities between DLAs and nearby galaxies. We suggest that the transition-mass model provides a plausible scenario for the bimodality we have found. As a result, the bimodality in current galaxies may have originated in DLAs.

astro-ph↗

arXiv2007

Characterizing the Low-Redshift Intergalactic Medium towards PKS1302-102

We present a detailed analysis of the intergalactic metal-line absorption systems in the archival HST/STIS and FUSE ultraviolet spectra of the low-redshift quasar PKS1302-102 (z_QSO = 0.2784). We supplement the archive data with CLOUDY ionization models and a survey of galaxies in the quasar field. There are 15 strong Lya absorbers with column densities logN_HI > 14. Of these, six are associated with at least CIII 977 absorption (logN(C^++) > 13); this implies a redshift density dN_CIII/dz = 36+13/-9 (68% confidence limits) for the five detections with rest equivalent width W_r > 50 mA. Two systems show OVI 1031,1037 absorption in addition to CIII (logN(O^+5) > 14). One is a partial Lyman limit system (logN_HI = 17) with associated CIII, OVI, and SiIII 1206 absorption. There are three tentative OVI systems that do not have CIII detected. For one OVI doublet with both lines detected at 3 sigma with W_r > 50 mA, dN_OVI/dz = 7+9/-4. We also search for OVI doublets without Lya absorption but identify none. From CLOUDY modeling, these metal-line systems have metallicities spanning the range -4 < [M/H] < -0.3. The two OVI systems with associated CIII absorption cannot be single-phase, collisionally-ionized media based on the relative abundances of the metals and kinematic arguments. From the galaxy survey, we discover that the absorption systems are in a diverse set of galactic environments. Each metal-line system has at least one galaxy within 500 km/s and 600 h^-1 kpc with L > 0.1 L_*.

astro-ph↗

arXiv2007

The color excess of quasars with intervening DLA systems- Analysis of the SDSS data release five

We analyzed the spectroscopic and photometric database of the 5th data release of the Sloan Digital Sky Survey (SDSS) to search for evidence of the quasar reddening produced by dust embedded in intervening damped Ly alpha (DLA) systems. From a list of 5164 quasars in the interval of emission redshift 2.25 </= z_e </= 3.5 and SDSS spectra with signal-to-noise ratio SNR >/= 4, we built up an "absorption sample" of 248 QSOs with a single DLA system in the interval of absorption redshift 2.2 < z_a </= 3.5 and a "pool" of 1959 control QSOs without DLA systems or strong metal systems. For each QSO of the absorption sample we extracted from the pool a subset of control QSOs that are closest in redshift and magnitude. The mean color of this subset was used as a zero point to measure the "deviation from the mean color" of individual DLA-QSOs, Delta_i. The colors were measured using "BEST" ugriz SDSS imaging data. The mean color excess of the absorption sample, , was estimated by averaging the individual color deviations Delta_i. We find = 27 +/- 9 x 10**(-3) mag and = 54 +/- 12 x 10**(-3) mag. These values are representative of the reddening of DLA systems at z_a ~ 2.7 in SDSS QSOs with limiting magnitude r =/~ 20.2. The detection of the mean reddening is confirmed by several statistical tests. Analysis of the results suggests an origin of the reddening in dust embedded in the DLA systems, with an SMC-type extinction curve. By converting the reddening into rest-frame extinction, we derive a mean dust-to-gas ratio ~ 2 to 4 x 10**(-23) mag cm^2. This value is ~ -1.25 dex lower than the mean dust-to-gas ratio of the Milky Way, in line with the lower level of metallicity in the present DLA sample.

astro-ph↗

arXiv2007

On the Incidence of C IV Absorbers Along the Sightlines to Gamma-Ray Bursts

We report on the statistics of strong (W_r > 0.15 A) C IV absorbers at z=1.5-3.5 toward high-redshift gamma-ray bursts (GRBs). In contrast with a recent survey for strong Mg II absorption systems at z < 2, we find that the number of C IV absorbers per unit redshift dN/dz does not show a significant deviation from previous surveys using quasi-stellar objects (QSOs) as background sources. We find that the number density of C IV toward GRBs is dN/dz(z~1.5)= 2.2 +2.8/-1.4, dN/dz(z~2.5)= 2.3 +1.8/-1.1 and dN/dz(z~3.5)= 1.1 +2.6/-0.9. These numbers are consistent with previous C IV surveys using QSO spectra. Binning the entire dataset, we set a 95% c.l. upper limit to the excess of C IV absorbers along GRB sightlines at twice the incidence observed along QSO sightlines. Furthermore, the distribution of equivalent widths of the GRB and QSO samples are consistent with being drawn from the same parent population. Although the results for Mg II and C IV absorbers along GRB sightlines appear to contradict one another, we note that the surveys are nearly disjoint: the C IV survey corresponds to higher redshift and more highly ionized gas than the Mg II survey. Nevertheless, analysis on larger statistical samples may constrain properties of the galaxies hosting these metals (e.g. mass, dust content) and/or the coherence-length of the gas giving rise to the metal-line absorption.

astro-ph↗

arXiv2007

A New Constraint on the Escape Fraction in Distant Galaxies Using Gamma-ray Burst Afterglow Spectroscopy

We describe a new method to measure the escape fraction fesc of ionizing radiation from distant star-forming galaxies using the afterglow spectra of long-duration gamma-ray bursts (GRBs). Optical spectra of GRB afterglows allow us to evaluate the optical depth of the host ISM, according to the neutral hydrogen column density N(HI) observed along the sightlines toward the star-forming regions where the GRBs are found. Different from previous effort in searching for faint, transmitted Lyman continuum photons, our method is not subject to background subtraction uncertainties and does not require prior knowledge of either the spectral shape of the host galaxy population or the IGM Lya forest absorption along these GRB sightlines. Because most GRBs occur in sub-L_* galaxies, our study also offers the first constraint on fesc for distant low-mass galaxies that dominate the cosmic luminosity density. We have compiled a sample of 27 GRBs at redshift z>2 for which the underlying N(HI) in the host ISM are known. These GRBs together offer a statistical sampling of the integrated optical depth to ionizing photons along random sightlines from star-forming regions in the host galaxies, and allow us to estimate the mean escape fraction averaged over different viewing angles. We find =0.02\pm 0.02 and place a 95% c.l. upper limit <= 0.075 for these hosts. We discuss possible biases of our approach and implications of the result. Finally, we propose to extend this technique for measuring at z~0.2 using spectra of core-collapse supernovae.

astro-ph↗

arXiv2007

The Chemical Evolution Of High Z Galaxies From The Relative Abundances Of N, Si, S, And Fe In Damped Lyman Alpha Systems

Abundances of N, Si, S, and Fe for 45 damped Lyman alpha systems (DLAs) have been compiled and detailed one-zone chemical evolution models have been constructed for 30 of them. Assuming continuous star formation, we found that final abundances in each object can be modelled by adjusting only two parameters, i.e. its time-averaged star formation efficiency and evolutionary age, with ranges in our sample of 0.01-1.5 Gyr^-1 and 0.18-2.0 Gyr, respectively. In addition, average star formation efficiency and evolutionary age appear to be anticorrelated for the sample, suggesting that the star formation efficiency in a typical DLA decreases with age. At the same time, N/Si in DLAs is directly linked to an object's age. There is an apparent bimodality in the distribution of N/Si values which could be the result of a statistical accident or an effect produced by a truncated or flattened IMF. We find that the mean and small dispersion of Si/Fe values is related to the generally young ages of DLAs, wherein not all Fe has yet been expelled by Type Ia supernovae. Finally, the large scatter and generally lower values of N/Si of DLAs with respect to blue compact galaxies, despite their partially overlapping metallicities, indicate that DLAs are generally younger than the latter.

astro-ph↗

arXiv2007

The Galactic Deuterium Abundance and Dust Depletion: Insights From an Expanded Ti/H Sample

The primordial abundance of deuterium (D/H) yields a measure of the density of baryons in the universe and is an important complement to determinations from cosmic microwave background (CMB) experiments. Indeed, the current small sample of high redshift D/H measurements from quasar absorption line studies are in excellent agreement with CMB-derived values. Conversely, absorption line measurements of the Galactic D/H ratio in almost 50 stellar sightlines show a puzzlingly large scatter outside the Local Bubble which is difficult to explain simply by astration from the primordial value. Here, we investigate the dust depletion scenario by studying the correlation between D/H and the abundance of titanium, one of the most refractory elements readily observed in the ISM. With a sample 3 times larger than previous work, we confirm a correlation between Ti/H and D/H at the 97% confidence level. However, the magnitude of this dependence is difficult to reconcile with a simple model of dust depletion for two reasons. First, contrary to what is expected from local depletion rates, the gradient of the highly refractory Ti is much shallower than that observed for Fe and Si. Second, we do not observe the established tight, steep correlation between [Ti/H] and the mean volume density of hydrogen. Therefore, whilst dust remains a plausible explanation for the local D/H variations, the abundances of at least some of the refractory elements do not provide unanimous support for this scenario. We also argue that the correlations of [Si/H], [Fe/H], and [Ti/H] with D/H are inconsistent with a simple infall model of low metallicity gas with approximately solar abundances as the dominant cause for D variations.

astro-ph↗