J. Michael Shull

THE COSMIC ORIGINS SPECTROGRAPH

The Cosmic Origins Spectrograph (COS) is a moderate-resolution spectrograph with unprecedented sensitivity that was installed into the Hubble Space Telescope (HST) in 2009 May, during HST Servicing Mission 4 (STS-125). We present the design philosophy and summarize the key characteristics of the instrument that will be of interest to potential observers. For faint targets, with flux F ? 1.0 ? 10?14?erg?cm?2?s?1 ??1, COS can achieve comparable signal to noise (when compared to Space Telescope Imaging Spectrograph echelle modes) in 1%-2% of the observing time. This has led to a significant increase in the total data volume and data quality available to the community. For example, in the first 20 months of science operation (2009 September-2011 June) the cumulative redshift pathlength of extragalactic sight lines sampled by COS is nine times than sampled at moderate resolution in 19 previous years of Hubble observations. COS programs have observed 214 distinct lines of sight suitable for study of the intergalactic medium as of 2011 June. COS has measured, for the first time with high reliability, broad Ly? absorbers and Ne VIII in the intergalactic medium, and observed the He II reionization epoch along multiple sightlines. COS has detected the first CO emission and absorption in the UV spectra of low-mass circumstellar disks at the epoch of giant planet formation, and detected multiple ionization states of metals in extra-solar planetary atmospheres. In the coming years, COS will continue its census of intergalactic gas, probe galactic and cosmic structure, and explore physics in our solar system and Galaxy.

The Low-z Intergalactic Medium. III. H I and Metal Absorbers at z < 0.4

We conduct an ultraviolet (HST and FUSE) spectroscopic survey of H I (Lyman lines) and seven metal ions (O VI, N V, C IV, C III, Si IV, Si III, Fe III) in the low-redshift IGM at z < 0.4. We analyzed 650 Lyα absorbers over redshift path length Δ z = 5.27, detecting numerous absorbers: 83 O VI systems, 39 C III, 53 Si III, 24 C IV, 24 N V, and so on. In the low-z IGM, we have accounted for ~40% of the baryons: 30% in the photoionized Lyα forest and 10% in the (T = 105–106) WHIM traced by O VI. Statistical metallicities are consistent with the canonical value of 10% solar, with considerable scatter. Improved statistics for weak absorbers allows us to estimate ΩWHIM/Ωb = 0.073 ± 0.008 down to log NO VI = 13.4 and 0.086 ± 0.008 down to log NO VI = 13.0. The O VI absorber line frequency, d/dz = 40−8+14, down to 10 mA equivalent width suggests a 250-300 kpc extent of metals around dwarf galaxies. Many absorbers appear to contain multiphase gas, with both collisional ionization and photoionization determining the ionization state. N V absorption is well correlated with O VI, and both ions show similarly steep power-law indices d/dz ∝ N−β with βO VI ≈ βN V ≈ 2 while βH I. We conclude that O VI and N V are reliable tracers of the portion of the WHIM at T ≈ 105–106 K. C IV may be present in both collisional and photoionized phases; NC IV correlates poorly with both NH I and NO VI and βH I < βC IV < βO VI. The ions C III, Si III, and Si IV are well correlated with H I and show patterns typical of photoionization. Adjacent ion stages of the same element (C III/IV and Si III/IV) provide useful constraints on the photoionization parameter, log U ≈ − 1.5 ± 0.5.

What Is the Total Deuterium Abundance in the Local Galactic Disk

Analyses of spectra obtained with the Far Ultraviolet Spectroscopic Explorer (FUSE) satellite, together with spectra from the Copernicus and interstellar medium absorption profile spectrograph (IMAPS) instruments, reveal an unexplained, very wide range in the observed deuterium/hydrogen (D/H) ratios for interstellar gas in the Galactic disk beyond the Local Bubble. We argue that spatial variations in the depletion of deuterium onto dust grains can explain these local variations in the observed gas-phase D/H ratios. We present a variable deuterium depletion model that naturally explains the constant measured values of D/H inside the Local Bubble, the wide range of gas-phase D/H ratios observed in the intermediate regime [log N(H ) = 19.2-20.7], and the low gas-phase D/H ratios observed at larger hydrogen column densities. We consider empirical tests of the deuterium depletion hypothesis: (1) correlations of gas-phase D/H ratios with depletions of the refractory metals iron and silicon, and (2) correlation with the H2 rotational temperature. Both of these tests are consistent with deuterium depletion from the gas phase in cold, not recently shocked regions of the ISM, and high gas-phase D/H ratios in gas that has been shocked or otherwise heated recently. We argue that the most representative value for the total (gas plus dust) D/H ratio within 1 kpc of the Sun is ≥23.1 ± 2.4(1 σ) parts per million (ppm). This ratio constrains Galactic chemical evolution models to have a very small deuterium astration factor, the ratio of primordial to total (D/H) ratio in the local region of the Galactic disk, which we estimate to be fd ≤ 1.19(1 σ) or ≤1.12 ± 0.14(1 σ) depending on the adopted light-element nuclear reaction rates.

The Fate of the First Galaxies. II. Effects of Radiative Feedback

We use three-dimensional cosmological simulations with radiative transfer to study the formation and evolution of the first galaxies in a ΛCDM cosmology. The simulations include continuum radiative transfer using the optically thin variable Eddington tensor (OTVET) approximation and line radiative transfer in the H2 Lyman-Werner bands of the UV background radiation. Chemical and thermal processes are treated in detail, particularly the ones relevant for H2 formation and destruction. We find that the first luminous objects (small-halo objects) are characterized by bursting star formation (SF) that is self-regulated by a feedback process acting on cosmological instead of galactic scales. The global SF history is regulated by the mean number of ionizing photons that escape from each source, UVfesc. It is almost independent of the assumed SF efficiency parameter, *, and the intensity of the dissociating background. The main feedback process that regulates the SF is the reformation of H2 in front of H II regions and inside relic H II regions. The H II regions remain confined inside filaments, maximizing the production of H2 in overdense regions through cyclic destruction/reformation of H2. If UVfesc > 10-7/*, the SF is self-regulated, photoevaporation of small-halo objects dominates the metal pollution of the low-density intergalactic medium, and the mass of produced metals depends only on fesc. If UVfesc 10-7/*, positive feedback dominates, and small-halo objects constitute the bulk of the mass in stars and metals until at least redshift z ~ 10. Small-halo objects cannot reionize the universe because the feedback mechanism confines the H II regions inside the large-scale structure filaments. In contrast to massive objects (large halos), which can reionize voids, small-halo objects partially ionize only the dense filaments while leaving the voids mostly neutral.

Zero-Metallicity Stars and the Effects of the First Stars on Reionization

We present stellar structure and atmosphere models of metal-free stars and examine them from a cosmological point of view. Metal-free stars exhibit high effective temperatures and small sizes relative to metal-enriched stars of equal mass. These unique physical characteristics enhance the ionizing photon production by metal-free stars, particularly in the He ii (hnu>/=4 ryd) continuum. The star formation rate of metal-free stars necessary to reionize the hydrogen in the universe by z=5 is consistent with the inferred star formation rate at that epoch. However, the hard stellar spectra are inconsistent with the observations of He ii opacity in the intergalactic medium (IGM) at z approximately 3, indicating that the period of metal-free star formation ended before that epoch. We examine the effects of these stars on the ionization balance of the IGM, the radiative feedback of the first luminous objects, and the extragalactic radiation field. We comment on the prospects for detecting metal-free stellar populations with the lambda1640 and lambda4686 recombination lines of He ii.

A Composite Extreme-Ultraviolet QSO Spectrum from FUSE

The Far Ultraviolet Spectroscopic Explorer (FUSE) has surveyed a large sample (>100) of active galactic nuclei (AGNs) in the low-redshift universe (z 0.33 formed from archival Hubble Space Telescope (HST) spectra, α = -1.76 ± 0.12. We identify several prominent emission lines in the FUSE composite and find that the high-ionization O VI and Ne VIII emission lines are enhanced relative to the HST composite. Power-law continuum fits to the individual FUSE AGN spectra reveal a correlation between EUV spectral slope and AGN luminosity in the FUSE and FUSE+HST samples, in the sense that lower luminosity AGNs show harder spectral slopes. We find an anticorrelation between the hardness of the EUV spectral slope and AGN black hole mass, using estimates of this quantity found in the literature. We interpret these results in the context of the well-known anticorrelation between AGN luminosity and emission-line strength, the Baldwin effect, given that the median luminosity of the FUSE AGN sample is an order of magnitude lower than that of the HST sample.The Far Ultraviolet Spectroscopic Explorer (FUSE) has surveyed a large sample (>100) of active galactic nuclei in the low-redshift universe (z<1). Its response at short wavelengths makes it possible to measure directly the far ultraviolet spectral properties of quasistellar objects (QSOs) and Seyfert 1 galaxies at z<0.3. Using archival FUSE spectra, we form a composite extreme ultraviolet (EUV) spectrum of QSOs at z<0.67. After consideration of many possible sources of systematic error in our analysis, we find that the spectral slope of the FUSE composite spectrum, \alpha= -0.56^+0.38_-0.28 for F_\nu \propto \nu^\alpha, is significantly harder than the EUV (\lambda \lesssim 1200 A) portion of the composite spectrum of QSOs with z>0.33 formed from archival Hubble Space Telescope spectra, \alpha=-1.76 \pm 0.12. We identify several prominent emission lines in the \fuse composite and find that the high-ionization O VI and Ne VIII emission lines are enhanced relative to the HST composite. Power law continuum fits to the individual FUSE AGN spectra reveal a correlation between EUV spectral slope and AGN luminosity in the FUSE and FUSE + HST samples in the sense that lower luminosity AGNs show harder spectral slopes. We find an anticorrelation between the hardness of the EUV spectral slope and AGN black hole mass, using estimates of this quantity found in the literature. We interpret these results in the context of the well-known anticorrelation between AGN luminosity and emission line strength, the Baldwin effect, given that the median luminosity of the FUSE AGN sample is an order of magnitude lower than that of the HST sample.

Feedback from Galaxy Formation: Production and Photodissociation of Primordial H2

We use one-dimensional radiative transfer simulations to study the evolution of H2 gas-phase (H- catalyzed) formation and photodissociation regions in the primordial universe. We find a new positive feedback mechanism capable of producing shells of H2 in the intergalactic medium (IGM), which are optically thick in some Lyman-Werner bands. While these shells exist, this feedback effect is important in reducing the H2 dissociating background flux and the size of photodissociation spheres around each luminous object. The maximum background opacity of the IGM in the H2 Lyman-Werner bands is ? ? 1-2 for a relic molecular fraction x = 2 ? 10-6, about 6 times greater than that found by Haiman, Abel, & Rees. Therefore, the relic molecular hydrogen can decrease the photodissociation rate by about an order of magnitude. The problem is relevant to the formation of small primordial galaxies with masses MDM 108 M? that rely on molecular hydrogen cooling to collapse. Alternatively, the universe may have remained dark for several hundred million years after the birth of the first stars, until galaxies with virial temperature Tvir 104 K formed.We use one-dimensional radiative transfer simulations to study the evolution of H_2 gas-phase (H^- catalyzed) formation and photo-dissociation regions in the primordial universe. We find a new positive feedback mechanism capable of producing shells of H_2 in the intergalactic medium, which are optically thick in some Lyman-Werner bands. While these shells exist, this feedback effect is important in reducing the H_2 dissociating background flux and the size of photo-dissociation spheres around each luminous object. The maximum background opacity of the IGM in the H_2 Lyman-Werner bands is \tau_{H_2} ~ 1-2 for a relic molecular fraction x_{H_2}=2 x 10^{-6}, about 6 times greater than found by Haiman, Abel & Rees. Therefore, the relic molecular hydrogen can decrease the photo-dissociation rate by about an order of magnitude. The problem is relevant to the formation of small primordial galaxies with masses M_{DM} 10^4 K formed.

Heating and Ionization of the Intergalactic Medium by an Early X-Ray Background

Observational studies indicate that the intergalactic medium (IGM) is highly ionized up to redshifts just over 6. A number of models have been developed to describe the process of reionization and the effects of the ionizing photons from the first luminous objects. In this paper we study the impact of an X-ray background, such as high-energy photons from early quasars, on the temperature and ionization of the IGM prior to reionization, before the fully ionized bubbles associated with individual sources have overlapped. X-rays have large mean free paths relative to EUV photons, and their photoelectrons can have significant effects on the thermal and ionization balance. We find that hydrogen ionization is dominated by the X-ray photoionization of neutral helium and the resulting secondary electrons. Thus, the IGM may have been warm and weakly ionized prior to full reionization. We examine several related consequences, including the filtering of the baryonic Jeans mass scale, signatures in the cosmic microwave background, and the H--catalyzed production of molecular hydrogen.

Critical metallicity and fine-structure emission of primordial gas enriched by the first stars

The influence of the first stars on the formation of second-generation objects at high redshift may be determined, in part, by their metal enrichment of surrounding gas. At a critical metallicity, Zcrit, primordial gas cools more efficiently by fine-structure lines of [C II] (157.74 μm), [O I] (63.18 μm, 145.5 μm), [Si II] (34.8 μm), and [Fe II] (25.99 μm, 35.35 μm) than by H I or H2 emission. This cooling may alter the process of fragmentation into smaller units. We study the time-dependent cooling of primordial gas enriched by heavy elements from early massive stars, particularly O, Si, and Fe. We define Zcrit as the point when the total cooling rate by metals plus H2 equals the adiabatic compressional heating. We explore two metallicity scenarios: (1) a single metallicity for all heavy elements and (2) individual metallicities (ZC, ZO, ZSi, and ZFe) from theoretical supernova yields. For dense gas (n ≥ 103 cm-3) with metals in relative solar abundances, fragmentation occurs at Zcrit ≈ 10-3.5 Z☉. However, for lower density gas (n = 1-100 cm-3), particularly in halos enriched in Si, O, and Fe, we find Zcrit ≈ 0.1%-1% Z☉. The critical metallicity approaches a minimum value at high n set by efficient LTE cooling, with thermalized level populations of fine-structure states and H2 rotational states (J = 2 and J = 3). Primordial clouds of 108 M☉ at 200 K are detectable in redshifted fine-structure lines, with far-infrared fluxes between 10-22 and 10-21 W m-2. For metallicities ZO ≈ 10-3 and molecular fractions f ≈ 10-3, the fine-structure emission lines of [O I], [Si II], and [Fe II] could be 102-103 times stronger than the H2 rotational lines at 28.22 μm (J = 2 → 0) and 17.03 μm (J = 3 → 1).

The Galaxy Environment of O VI Absorption Systems

We combine a FUSE sample of O VI absorbers (z < 0.15) with a database of 1.07 million galaxy redshifts to explore the relationship between absorbers and galaxy environments. All 37 absorbers with N ≥ 1013.2 cm-2 lie within 800 h kpc of the nearest galaxy, with no compelling evidence for O VI absorbers in voids. The O VI absorbers often appear to be associated with environments of individual galaxies. Gas with 10% ± 5% solar metallicity (O VI and C III) has a median spread in distance of 350-500 h kpc around L galaxies and 200-270 h kpc around 0.1L galaxies (ranges reflect uncertain metallicities of gas undetected in Lyα absorption). In order to match the O VI line frequency, (d/dz) ≈ 20 for N ≥ 1013.2 cm-2, galaxies with L ≤ 0.1L must contribute to the cross section. Lyα absorbers with N ≥ 1013.2 cm-2 cover ~50% of the surface area of typical galaxy filaments. Two-thirds of these show O VI and/or C III absorption, corresponding to a 33%-50% covering factor at 0.1 Z☉ and suggesting that metals are spread to a maximum distance of 800 h kpc, within typical galaxy supercluster filaments. Approximately 50% of the O VI absorbers have associated Lyα line pairs with separations (Δv)Lyα = 50-200 km s-1. These pairs could represent shocks at the speeds necessary to create copious O VI, located within 100 h kpc of the nearest galaxy and accounting for much of the two-point correlation function of low-z Lyα forest absorbers.