arXiv subjects

Nahum Arav

Publications and source records attributed to Nahum Arav.

89 records · Page 5Linked to original sources

arXiv2005

X-ray/UV Observing Campaign on the Mrk 279 AGN Outflow: A Global Fitting Analysis of the UV Absorption

We present an analysis of the intrinsic UV absorption in the Seyfert 1 galaxy Mrk 279 based on simultaneous long observations with the Hubble Space Telescope (41 ks) and the Far Ultraviolet Spectroscopic Explorer (91 ks). To extract the line-of-sight covering factors and ionic column densities, we separately fit two groups of absorption lines: the Lyman series and the CNO lithium-like doublets. For the CNO doublets we assume that all three ions share the same covering factors. The fitting method applied here overcomes some limitations of the traditional method using individual doublet pairs; it allows for the treatment of more complex, physically realistic scenarios for the absorption-emission geometry and eliminates systematic errors that we show are introduced by spectral noise. We derive velocity-dependent solutions based on two models of geometrical covering -- a single covering factor for all background emission sources, and separate covering factors for the continuum and emission lines. Although both models give good statistical fits to the observed absorption, we favor the model with two covering factors because: (a) the best-fit covering factors for both emission sources are similar for the independent Lyman series and CNO doublet fits; (b) the fits are consistent with full coverage of the continuum source and partial coverage of the emission lines by the absorbers, as expected from the relative sizes of the nuclear emission components; and (c) it provides a natural explanation for variability in the Ly$α$ absorption detected in an earlier epoch. We also explore physical and geometrical constraints on the outflow from these results.

astro-ph↗

arXiv2004

X-ray/UV Campaign on the Mrk 279 AGN Outflow: Constraining Inhomogeneous Absorber Models

We investigate the applicability of inhomogeneous absorber models in the formation of AGN outflow absorption-troughs. The models we explore are limited to monotonic gradients of absorbing column densities in front of a finite emission source. Our main finding is that simple power-law and gaussian distributions are hard pressed to fit the Mrk 279 high-quality UV outflow data. An acceptable fit for the O VI troughs can only be obtained by assuming unrealistic optical depth values (upward of 100). The strongest constraints arise from the attempt to fit the Lyman series troughs. In this case it is evident that even allowing for complete freedom of both the power-law exponent and the optical depth as a function of velocity cannot yield an acceptable fit. In contrast, partial covering models do yield good fits for the Lyman series troughs. We conclude that monotonic inhomogeneous absorber models that do not include a sharp edge in the optical depth distribution across the source are not an adequate physical model to explain the trough formation mechanism for the outflow observed in Mrk 279.

astro-ph↗

arXiv2004

Intrinsic Absorption in the Spectrum of Mrk 279: Simultaneous Chandra, FUSE, and STIS Observations

We present a study of the intrinsic X-ray and far-ultraviolet absorption in the Seyfert 1.5 galaxy Markarian 279 using simultaneous observations from the Chandra X-ray Observatory, the Space Telescope Imaging Spectrograph aboard the Hubble Space Telescope, and the Far Ultraviolet Spectroscopic Explorer (FUSE). We also present FUSE observations made at three additional epochs. We detect the Fe K-alpha emission line in the Chandra spectrum, and its flux is consistent with the low X-ray continuum flux level of Mrk 279 at the time of the observation. Due to low signal-to-noise ratios in the Chandra spectrum, no O VII or O VIII absorption features are observable in the Chandra data, but the UV spectra reveal strong and complex absorption from HI and high-ionization species such as O VI, N V, and C IV, as well as from low-ionization species such as C III, N III, C II, and N II in some velocity components. The far-UV spectral coverage of the FUSE data provides information on high-order Lyman series absorption, which we use to calculate the optical depths and line and continuum covering fractions in the intrinsic HI absorbing gas in a self-consistent fashion. The UV continuum flux of Mrk 279 decreases by a factor of ~7.5 over the time spanning these observations and we discuss the implications of the response of the absorption features to this change. From arguments based on the velocities, profile shapes, covering fractions and variability of the UV absorption, we conclude that some of the absorption components, particularly those showing prominent low-ionization lines, are likely associated with the host galaxy of Mrk 279, and possibly with its interaction with a close companion galaxy, while the remainder arises in a nuclear outflow.

astro-ph↗

arXiv2003

Outflows vs. Clouds in AGN Intrinsic Absorbers

We discuss the crucial role of a dynamical picture in the analysis of AGN intrinsic absorbers data. High quality FUSE data of Mrk 279 are used to demonstrate that the line of sight covering fraction is a strong function of velocity. In Mrk 279, as well as in most cases where the data is of high enough quality, the shape of the absorption troughs is mainly determined by the velocity-dependent covering fraction. We argue that the traditional ``cloud'' picture of AGN outflows is hard pressed to account for the velocity-dependent covering fraction, as well as for the highly super-thermal width of the troughs and the detached trough phenomenon. A disk outflow picture naturally explains these features and furthermore, is using the simplest reservoir for the outflowing material: The accretion disk around the black hole. Accounting for velocity dependent covering can drastically increase the inferred ionic column density of the analyzed trough, an increase which is amplified in the total column density and ionization parameter solution for the AGN outflow trough.

astro-ph↗

arXiv2003

Contrasting the UV and X-ray O VI Column Density Inferred for the Outflow in NGC 5548

We compare X-ray and UV spectroscopic observations of NGC 5548. Both data sets show O VI absorption troughs associated with the AGN outflow from this galaxy. We find that the robust lower limit on the column density of the O VI X-ray trough is seven times larger than the column density found in a study of the O VI UV troughs. This discrepancy suggests that column densities inferred for UV troughs of Seyfert outflows are often severely underestimated. We identify the physical limitations of the UV Gaussian modeling as the probable explanation of the O VI column density discrepancy. Specifically, Gaussian modeling cannot account for a velocity dependent covering fraction, and it is a poor representation for absorption associated with a dynamical outflow. Analysis techniques that use a single covering fraction value for each absorption component suffer from similar limitations. We conclude by suggesting ways to improve the UV analysis.

astro-ph↗

arXiv2002

AGN Outflows: Analysis of the Absorption Troughs

With the advent of Chandra and XMM, X-ray lines spectroscopy of AGN outflows is off to an exciting start. In this paper we illuminate some of the complications involved in extracting the outflow's physical conditions (ionization equilibrium, total column density, chemical abundances) from such spectroscopic data. To do so we use the example provided by high-quality FUSE and HST observations of the outflow seen in NGC 985. We show how simple determinations of the column density in the UV absorption lines often severely underestimate the real column densities. This is due to strong non-black saturation of the absorption troughs, where in many cases the UV line profile mainly reflects the velocity-dependent covering factor rather than the column density distribution. We then show that underestimating individual ionic column densities by a factor of 5 can cause a two orders of magnitude error in the inferred total column density of the outflow. In some case this will be enough to associate the UV and X-ray absorber with the same outflowing material. Finally, we note that the UV spectra of NGC 985 have 10--20 times the resolution and 2--5$\times$ the S/N per resolution element compared with the best available Chandra spectra of similar objects. Therefore, in the X-ray band non-black saturation and velocity-dependent covering factor effects will only become abundantly clear using vastly more capable future X-ray telescopes. However, by taking into consideration the lessons learned from the UV band, we can greatly improve the quality of physical constraints we extract from current X-ray data of AGN outflows.

astro-ph↗

arXiv2002

On the Column Density of AGN Outflows: the Case of NGC 5548

We re-analyze the HST high resolution spectroscopic data of the intrinsic absorber in NGC 5548 and find that the C IV absorption column density is at least four times larger than previously determined. This increase arises from accounting for the kinematical nature of the absorber and from our conclusion that the outflow does not cover the narrow emission line region in this object. The improved column density determination begins to bridge the gap between the high column densities measured in the X-ray and the low ones previously inferred from the UV lines. Combined with our findings for outflows in high luminosity quasars these results suggest that traditional techniques for measuring column densities: equivalent width, curve-of-growth and Gaussian modeling, are of limited value when applied to absorption associated with AGN outflows.

astro-ph↗

arXiv2001

Photoionization Models for BALQSO PG 0946+301

This is a companion paper to the manuscript entitled: ``HST STIS Observations Of PG 0946+301: The Highest Quality UV Spectrum of a BALQSO'' (Arav et al. 2001; accepted for publication in ApJ). Here we present photoionization-modeling results for ALL the ionic column density constraints found in these data, most of which we were unable to include in the printed version of the paper.

astro-ph↗

arXiv2001

HST STIS Observations of PG 0946+301: the Highest Quality UV Spectrum of a BALQSO

We describe deep (40 orbits) HST/STIS observations of the BALQSO PG 0946+301 and make them available to the community. These observations are the major part of a multi-wavelength campaign on this object aimed at determining the ionization equilibrium and abundances (IEA) in broad absorption line (BAL) QSOs. We present simple template fits to the entire data set, which yield firm identifications for more than two dozen BALs from 18 ions and give lower limits for the ionic column densities. We find that the outflow's metalicity is consistent with being solar, while the abundance ratio of phosphorus to other metals is at least ten times solar. These findings are based on diagnostics that are not sensitive to saturation and partial covering effects in the BALs, which considerably weakened previous claims for enhanced metalicity. Ample evidence for these effects is seen in the spectrum. We also discuss several options for extracting tighter IEA constraints in future analyses, and present the significant temporal changes which are detected between these spectra and those taken by the HST/FOS in 1992.

astro-ph↗

arXiv2000

Keck Hires Observations of the QSO First J104459.6+365605: Evidence for a Large Scale Outflow

This paper presents an analysis of a Keck HIRES spectrum of the QSO FIRST J104459.6+365605. The line of sight towards the QSO contains two clusters of outflowing clouds that give rise to broad blue shifted absorption lines. The outflow velocities of the clouds range from -200 to -1200 km/s and from -3400 to -5200 km/s, respectively. The width of the individual absorption lines ranges from 50 to more than 1000 km/s. The most prominent absorption lines are those of Mg II, Mg I, and Fe II. The low ionization absorption lines occur at the same velocities as the most saturated Mg II lines, showing that the Fe II, Mg I and Mg II line forming regions must be closely associated. Many absorption lines from excited states of Fe II are present, allowing a determination of the population of several low lying energy levels. From this we determine an electron density in the Fe II line forming regions of 4000 per cubic cm. Modelling the ionization state of the absorbing gas with this value of the electron density as a constraint, we find that the distance between the Fe II and Mg I line forming region and the continuum source is of order 700 parsec. From the correspondence in velocity between the Fe II, Mg I and Mg II lines we infer that the Mg II lines must be formed at the same distance. The Mg II absorption fulfills the criteria for Broad Absorption Lines defined by Weymann et al. (1991). This large distance is surprising, since BALs are generally thought to be formed in outflows at a much smaller distance from the nucleus.

astro-ph↗

arXiv2000

QSO 2359-1241: A Bright, Highly Polarized, Radio-Moderate, Reddened, Low-Ionization Broad Absorption Line Quasar

We report the discovery of a bright quasar (E=15.8, z=0.868) associated with the flat spectrum radio source NVSS J235953-124148. This quasar we designate QSO 2359-1241 possesses a rare combination of extreme properties that make it of special interest. These properties include: intrinsic high-velocity outflow seen in absorption for both high and low-ionization species, high optical polarization (about 5%), significant radio emission, and dust reddening. The dereddened absolute magnitude of QSO 2359-1241 places it among the three most optically luminous quasars known at z<1. High-resolution spectroscopy and a detailed analysis of the optical/ultraviolet absorption features will be given in a companion paper (Arav et al 2000).

astro-ph↗

arXiv2000

The Intrinsic Absorber in QSO 2359-1241: Keck and HST Observations

We present detailed analyses of the absorption spectrum seen in QSO 2359-1241 (NVSS J235953-124148). Keck HIRES data reveal absorption from twenty transitions arising from: He I, Mg I, Mg II, Ca II, and Fe II. HST data show broad absorption lines (BALs) from Al III 1857, C IV 1549, Si IV 1397, and N V 1240. Absorption from excited Fe II states constrains the temperature of the absorber to 2000K < T < 10,000K and puts a lower limit of 10^5 cm^{-3} on the electron number density. Saturation diagnostics show that the real column densities of He I and Fe II can be determined, allowing to derive meaningful constraints on the ionization equilibrium and abundances in the flow. The ionization parameter is constrained by the iron, helium and magnesium data to -3.0 < log(U) < -2.5 and the observed column densities can be reproduced without assuming departure from solar abundances. From comparison of the He I and Fe II absorption features we infer that the outflow seen in QSO 2359-1241 is not shielded by a hydrogen ionization front and therefore that the existence of low-ionization species in the outflow (e.g., Mg II, Al III, Fe II) does not necessitate the existence of such a front. We find that the velocity width of the absorption systematically increases as a function of ionization and to a lesser extent with abundance. Complementary analyses of the radio and polarization properties of the object are discussed in a companion paper (Brotherton et al. 2000).

astro-ph↗

arXiv1999

What Determines the Depth of BALs? Keck HIRES Observations of BALQSO 1603+300

We find that the depth and shape of the broad absorption lines (BALs) in BALQSO 1603+3002 are determined largely by the fraction of the emitting source which is covered by the BAL flow. In addition, the observed depth of the BALs is poorly correlated with their real optical depth. The implication of this result is that abundance studies based on direct extraction of column densities from the depth of the absorption troughs are unreliable. Our conclusion is based on analysis of unblended absorption features of two lines from the same ion (in this case the Si IV doublet), which allows unambiguous separation of covering factor and optical depth effects. The complex morphology of the covering factor as a function of velocity suggests that the BALs are produced by several physically separated outflows. The covering factor is ion dependent in both depth and velocity width. We also find evidence that in BALQSO 1603+3002 the flow does not cover the broad emission line region.

astro-ph↗

arXiv1998

HST Observations of the Broad Absorption Line Quasar PG 0946+301

We analyze HST and ground based spectra of the brightest BALQSO in the UV: PG~0946+301. A detailed study of the absorption troughs as a function of velocity is presented, facilitated by the use of a new algorithm to solve for the optical depth as a function of velocity for multiplet lines. We find convincing evidence for saturation in parts of the troughs. This supports our previous assertion that saturation is common in BALs and therefore cast doubts on claims for very high metallicity in BAL flows. Due to the importance of BAL saturation we also discuss its evidence in other objects. In PG~0946+301 large differences in ionization as a function of velocity are detected and our findings supports the hypothesis that the line of sight intersects a number of flow components that combine to give the appearance of the whole trough. Based on the optical depth profiles, we develop a geometrical-kinematical model for the flow. We have positively identified 16 ions of 8 elements (H~I, C~III, C~IV, N~III, N~IV, N~V, O~III, O~IV, O~V, O~VI, Ne~V, Ne~VIII, P~V, Si~IV, S~V, S~VI) and have a probable identifications of Mg~X and S~IV. Unlike earlier analysis of IUE data, we find no evidence for BALs arising from excited ionic states in the HST spectrum of PG~0946+301.

astro-ph↗

arXiv1998

PG 0946+301: the Rosetta Stone of BALQSOs?

We describe the motivation and features of a multiwavelength spectroscopic campaign on broad absorption line (BAL) QSO PG~0946+301. The main goal of this project is to determine the ionization equilibrium and abundances (IEA) in BAL outflows. Previous studies of IEA in BALQSOs were based on the assumption that the BALs are not saturated so that the column densities inferred from the apparent optical depths are realistic. This critical assumption is at odds with several recent observations and with analysis of existing data which indicate that the absorption troughs are heavily saturated even when they are not black. In addition, X-ray observations, which are important for constraining the ionizing continuum, were not available for those objects that had UV spectral data. Quantifying the level of saturation in the BALs necessitates UV spectroscopy with much higher S/N and broader spectral coverage than currently exist. After taking into account the capabilities of available observatories, our best hope for a substantial improvement in understanding the IEA in BALQSOs is to concentrate considerable observational resources on the most promising object. Our studies of available HST and ground-based spectra show that PG~0946+301 is by far the best candidate for such a program. This BALQSO is at least five times brighter, shortward of 1000 A rest frame, than any other object, and due to its low redshift it has an especially sparse Lya forest. At the same time PG~0946+301 is a typical BALQSO and therefore its IEA should be representative. To this effect we are developing a multiwavelength spectroscopic campaign (UV, FUV, X-ray and optical) on BALQSO PG 0946+301. We discuss the goals and feasibility of each observational component: HST, FUSE, ASCA and ground-based.

astro-ph↗

arXiv1998

Are AGN Broad Emission Lines Formed by Discrete Clouds? --- Analysis of Keck High Resolution Spectroscopy of NGC 4151

We search for a direct signature of discrete ''clouds'' in the broad line region (BLR) of the Seyfert galaxy NGC 4151. For this purpose we apply cross correlation (CC) analysis to a high resolution KECK spectrum of the galaxy. No such signature is found in the data. In order for cloud models to be compatible with this result, there must be at least $\sim3\times10^7$ emitting clouds in the BLR, where the limit is based on simulation of a homogeneous cloud population. More realistic distributions increase the lower limit to above $10^8$. These numbers are an order of magnitude improvement on our previous limit from Mrk 335, where the improvement comes from higher S/N, broader lines, and refined simulations. Combined with the predicted upper limit for the number of emitting clouds in NGC 4151 ($10^6-10^7$), the derived lower limit puts a strong constraint on the cloud scenario in the BLR of this object. Similar constraints can be paled on models where the emission originates in streams and sheets. Thus, this investigation suggests that the BELs in NGC 4151, and by extension in all AGNs, are not made of an ensemble of discrete independent emitters.

astro-ph↗

arXiv1997

What is the redshift of gamma-ray burst 970508?

A Bayesian likelihood analysis is used to constrain the redshift of the optical transient associated with gamma-ray burst 970508, under the assumption that the absorption at redshift 0.835 is not physically associated with the transient. The maximum-likelihood, expectation value and 95-percent upper limit on the OT redshift are 0.835, 1.3 and 1.9 respectively.

astro-ph↗