Rajib Ganguly

On the Fraction of Quasars with Outflows

Outflows from active galactic nuclei (AGNs) seem to be common and are thought to be important from a variety of perspectives: as an agent of chemical enhancement of the interstellar and intergalactic media, as an agent of angular momentum removal from the accreting central engine, and as an agent limiting star formation in starbursting systems by blowing out gas and dust from the host galaxy. To understand these processes, we must determine what fraction of AGNs feature outflows and understand what forms they take. We examine recent surveys of quasar absorption lines, reviewing the best means to determine if systems are intrinsic and result from outflowing material, and the limitations of approaches taken to date. The surveys reveal that, while the fraction of specific forms of outflows depends on AGN properties, the overall fraction displaying outflows is fairly constant, approximately 60%, over many orders of magnitude in luminosity. We emphasize some issues concerning classification of outflows driven by data type rather than necessarily the physical nature of outflows and illustrate how understanding outflows probably requires a more comprehensive approach than has usually been taken in the past.

Outflows and the Physical Properties of Quasars

We have investigated a sample of 5088 quasars from the Sloan Digital Sky Survey Second Data Release in order to determine how the frequency and properties of broad absorptions lines (BALs) depend on black hole mass, bolometric luminosity, Eddington fraction (L/LEdd), and spectral slope. We focus only on high-ionization BALs and find a number of significant results. While quasars accreting near the Eddington limit are more likely to show BALs than lower L/LEdd systems, BALs are present in quasars accreting at only a few percent Eddington. We find a stronger effect with bolometric luminosity, such that the most luminous quasars are more likely to show BALs. There is an additional effect, previously known, that BAL quasars are redder on average than unabsorbed quasars. The strongest effects involving the quasar physical properties and BAL properties are related to terminal outflow velocity. Maximum observed outflow velocities increase with both the bolometric luminosity and the blueness of the spectral slope, suggesting that the ultraviolet luminosity to a great extent determines the acceleration. These results support the idea of outflow acceleration via ultraviolet line scattering. Subject headings: quasars: general — quasars: absorption lines — quasars: fundamental parameters

On the Origin of Intrinsic Narrow Absorption Lines in z ≲ 1 QSOs

We present an exhaustive statistical analysis of the associated (Δvabs -0.5], and mediocre C IV FWHM (6000 km s-1) do not have detectable associated NALs, down to Wr(C ) = 0.35 A. We also find that broad absorption line (BAL) QSOs have an enhanced probability of hosting detectable NAL gas. In addition, we find that the velocity distribution of associated NALs is peaked around the emission redshifts rather than the systemic redshifts of the QSOs. Finally, we find only one strong NAL [Wr(C ) 1.5 A] in our low-redshift sample. A comparison with previous higher redshift surveys reveals evolution in the number of strong NAL systems with redshift. We interpret these results in the context of an accretion disk model. We propose that NAL gas hugs the streamlines of the faster, denser, low-latitude wind, which has been associated with BALs. In the framework of this scenario, we can explain the observational clues as resulting from differences in orientation and wind properties, the latter presumably associated with the QSO radio properties.

A Census of Intrinsic Narrow Absorption Lines in the Spectra of Quasars at z = 2-4*

We use Keck HIRES spectra of 37 optically bright quasars at z = 2-4 to study narrow absorption lines that are intrinsic to the quasars (intrinsic NALs, produced in gas that is physically associated with the quasar central engine). We identify 150 NAL systems, which contain 124 C IV, 12 N V, and 50 Si IV doublets, of which 18 are associated systems (within 5000 km s-1 of the quasar redshift). We use partial coverage analysis to separate intrinsic NALs from NALs produced in cosmologically intervening structures. We find 39 candidate intrinsic systems (28 reliable determinations and 11 that are possibly intrinsic). We estimate that 10%-17% of C IV systems at blueshifts of 5000-70,000 km s-1 relative to quasars are intrinsic. At least 32% of quasars contain one or more intrinsic C IV NALs. Considering N V and Si IV doublets showing partial coverage as well, at least 50% of quasars host intrinsic NALs. This result constrains the solid angle subtended by the absorbers to the background source(s). We identify two families of intrinsic NAL systems, those with strong N V absorption and those with negligible absorption in N V but with partial coverage in the C IV doublet. We discuss the idea that these two families represent different regions or conditions in accretion disk winds. Of the 26 intrinsic C IV NAL systems, 13 have detectable low-ionization absorption lines at similar velocities, suggesting that these are two-phase structures in the wind rather than absorbers in the host galaxy. We also compare possible models for quasar outflows, including radiatively accelerated disk-driven winds, magnetocentrifugally accelerated winds, and pressure-driven winds, and we discuss ways of distinguishing between these models observationally.

Variability of Narrow, Associated Absorption Lines in Moderate- and Low-Redshift Quasars

We present the results of a search for variability in the equivalent widths (EWs) of narrow, associated (jv j� 5000 km s � 1 ) absorption lines found in the UV spectra of z � 1:5 quasars. The goal of this search was to use variability as a means of identifying absorption lines arising in gas that is intrinsic to the quasar central engine. We have compared archival Hubble Space Telescope (HST) Faint Object Spectrograph (FOS) spectra of quasars with recent spectra obtained as part of our own snapshot survey of the same objects with the Space Telescope Imaging Spectrograph (STIS). The intervals between observations are 4-10 yr. We primarily focused on the C iv absorption lines, although we also studied other lines available in the same spectra (e.g., Ly� ,N v ,O vi). Our main result is that 4 out of 15 quasars, or 4 out of 19 associated absorption systems, contained variable narrow absorption lines, which are indicative of intrinsic absorption. We conclude that a minimum of 21% of the associated absorption-line systems are variable. Because not all systems will have necessarily varied, this is a lower limit on this fraction and is consistent with previous estimates based on variability, partial coverage analysis, or statistical arguments. If we interpret the upper limits on the variability timescale as upper limits on the recombination time of the absorber, we constrain the density of the absorber to be ne >3000 cm � 3 and its distance from the ionizing source to be R P100 pc. Moreover, we are now able to pick out specific intrinsic absorption-line systems to be followed up with high-dispersion spectroscopy in order to constrain the geometry, location, and physical conditions of the absorber. We briefly illustrate how follow-up studies can yield such constraints by means of a simulation.

High-Resolution Absorption Spectroscopy of Multiphase, High-Metallicity Gas Associated with the Luminous Quasar HE 0226?4110*

We present FUSE and HST STIS observations of the absorption-line system near the emission redshift of the radio-quiet, X-ray-bright quasar HE 0226-4110 (z = 0.495, V = 15.2). The spectra cover the rest-frame wavelength range 610-1150 A, and we detect a wide range of ionization species, including four adjacent stages of oxygen: O III-VI, which reveal a striking change in gas kinematics with ionization. Examinaton of the O VI λλ1031, 1037 doublet profiles reveals no evidence for partial coverage or unresolved saturated structure. O III is only detected in a narrow feature that is also traced by the H I and C III lines, suggesting that they arise in the same gas. Absorption at the same velocity is also present in other species (N IV, O IV-VI, S IV, and possibly Ne VIII ), but the kinematics differ from the O III, implying production in separate gaseous phases. The combination of H I, O III , and C III information yields an estimate of both the photoionization parameter and the metallicity of the O III-bearing gas: [O/H] = +0.12, log U = -2.29. We discuss two possible locations for the gas in this associated absorption-line system: the narrow emission line region of the quasar, and the halo of the quasar host galaxy. An additional narrow (and thus photoionized) component that is only detected in O VI appears 58 km s-1 redward of the O III-bearing gas with -0.35 log U 0.02. Additional structure is detected in the associated absorber in the form of two broad components that only appear in moderate- to high-ionization species.

Probing the Size of Low-Redshift Lyα Absorbers

The 3C 273 and RX J1230.8+0115 sight lines probe the outskirts of the Virgo Cluster at physical separations between the sight lines of 200 and 500 h kpc. We present an analysis of available Hubble Space Telescope STIS echelle and Far-Ultraviolet Spectroscopic Explorer (FUSE) UV spectroscopy of RX J1230.8 + 0115, in which we detect five Ly? absorbers at Virgo distances. One of these absorbers is a blend of two strong metal-line absorbers at a recession velocity coincident with the highest neutral hydrogen column density absorber in the 3C 273 sight line, ~350 h kpc away. The consistency of the metal-line column density ratios in the RX J1230.8+0115 sight line allows us to determine the ionization mechanism (photoionization) for these absorbers. While the low signal-to-noise ratio of the FUSE spectrum limits our ability to model the neutral hydrogen column density of these absorbers precisely, we are able to constrain them to be in the range 1016-1017 cm-2. The properties of these absorbers are similar to those obtained for the nearby 3C 273 absorber studied by Tripp and collaborators. However, the inferred line-of-sight size for the 3C 273 absorber is only 70 pc, much smaller than those inferred in RX J1230.8+0115, which are 10-30 h kpc. The small sizes of all three absorbers are at odds with the 350 h kpc minimum transverse size implied by an application of the standard QSO line-pair analysis. On the basis of absorber associations between these two sight lines we conclude that a large-scale structure filament produces a correlated, not contiguous, gaseous structure in this region of the Virgo Supercluster. These data may indicate that we are detecting overdensities in the large-scale structure filaments in this region. Alternatively, the presence of a galaxy 71 h kpc from a 3C 273 absorber may indicate that we have probed outflowing, starburst-driven shells of gas associated with nearby galaxies.

Highly Ionized Gas in the Galactic Halo and the High-Velocity Clouds toward PG 1116+215

We have obtained high-resolution Far Ultraviolet Spectroscopic Explorer (FUSE) and Hubble Space Telescope (HST) Space Telescope Imaging Spectrograph (STIS) echelle observations of the quasar PG 1116 + 215 (zem = 0.1765, l = 22336, b = +6821). The semicontinuous coverage of the ultraviolet spectrum over the wavelength range 916-2800 ? provides detections of Galactic and high-velocity cloud (HVC) absorption over a wide range of ionization species: H I, C II-IV, N I-II, O I, O VI, Mg II, Si II-IV, P II, S II, and Fe II over the velocity range -100 km s-1 < vLSR < +200 km s-1. The high dispersion of these spectra (6.5-20 km s-1) reveals that low-ionization species consist of five discrete components: three at low and intermediate velocities (vLSR ? -44, -7, +56 km s-1) and two at high velocities (vLSR ? +100, +184 km s-1). Over the same velocity range, the higher ionization species (C III-IV, O VI, Si IV)?those with ionization potentials larger than 40 eV?show continuous absorption with column density peaks at vLSR ? 10 km s-1, the expected velocity of halo gas corotating with the Galactic disk, and vLSR ? +184 km s-1, the velocity of the higher velocity HVC. The velocity coincidence of both low- and high-ionization species in the vLSR ? +184 km s-1 HVC gas suggests that they arise in a common structure, though not necessarily in the same gaseous phase. The absorption structure in the high-ionization gas, which extends to very low velocities, suggests a scenario in which a moderately dense cloud of gas is streaming away from the Galaxy through a hot external medium (either the Galactic halo or corona) that is stripping gas from this cloud. The cloud core produces the observed neutral atoms and low-ionization species. The stripped material is the likely source of the high-ionization species. Among the host of collisionally ionized nonequilibrium models, we find that shock ionization and conductive interfaces can account for the column density ratios of high-ionization species. The nominal metallicity of the neutral gas using the O I and H I column densities is [O/H] ~ -0.66, with a substantial uncertainty caused by the saturation of the H I Lyman series in the FUSE band. The ionization of the cloud core is likely dominated by photons, and assuming the source of ionizing photons is the extragalactic UV background, we estimate the cloud has a density of 10-2.7 cm-3 with a thermal pressure p/k ? 24 cm-3 K. If photons escaping the Galactic disk are also included (i.e., if the cloud lies closer than the outer halo), the density and thermal pressure could be higher by as much as 2 dex. In either case, the relative abundances of O, Si, and Fe in the cloud core are readily explained without departures from the solar pattern. We compare the column density ratios of the HVCs toward the PG 1116+215 to other isolated HVCs as well as Complex C. Magellanic Stream gas (either a diffuse extension of the leading arm or gas stripped from a prior passage) is a possible origin for this gas and is consistent with the location of the high-velocity gas on the sky, as well as its high positive velocity, the ionization, and metallicity.

An Intrinsic Absorption Complex toward RX J1230.8+0115: Geometry and Photoionization Conditions

We present Hubble Space Telescope/Space Telescope Imaging Spectrograph and Far Ultraviolet Spectroscopic Explorer spectra of the quasar RX J1230.8+0115 (V = 14.4,z = 0.117). In addition to Galactic, Virgo, and intervening absorption, this quasar is host to a remarkable intrinsic absorption complex. Four narrow absorption line systems, strong in C IV, N V, and O VI, lie within 5000 km s-1 of the QSO redshift. Three of the systems appear to be line locked, two in N V and two in O VI, with the common system residing in between the other two (in velocity). All three systems show signs of an intrinsic origin—smooth windlike profiles, high ionization, and partial coverage of the central engine. The fourth system, which appears at the systemic redshift of the QSO, may originate from host galaxy or intervening gas. Photoionization analyses imply column densities in the range 19.1 < log N(H) < 21 and ionization parameters in the range -1.3 < log U < 0.3. Revisiting the issue of line locking, we discuss a possible model in the context of the accretion disk/wind scenario and point out several issues that remain for future simulations and observations.

Variable Ultraviolet Absorption in the Spectrum of MR 2251–178*

We present an ultraviolet spectrum of MR 2251-178 taken with Hubble Space Telescope (HST)/Space Telescope Imaging Spectrograph (STIS) with the G230L grating. The observation is part of a snapshot program of quasars (QSOs) that systematically searches for intrinsic absorption lines through variability. The sample consists of all QSOs observed with the HST/Faint Object Spectrograph (FOS) that showed associated narrow absorption lines. The FOS spectrum of MR 2251-178, taken in 1996, showed an associated C IV doublet with a λ1548 equivalent width of 1.09 ± 0.09 A. It is not detected in the STIS spectrum taken 4 years later, down to a 3 σ threshold of 0.19 A. In addition to its other accolades, these observations make MR 2251-178 the very first QSO at low redshift in which the associated absorption is shown to be truly intrinsic. We discuss the implications of this and suggest courses for future study.