arXiv subjects

A. J. Barger

Publications and source records attributed to A. J. Barger.

At least 91 records · Page 5Linked to original sources

arXiv2001

Supermassive Black Hole Accretion History Inferred from a Large Sample of Chandra Hard X-ray Sources

We describe the optical, near-infrared, and radio properties of a sample of hard (2-7 keV) X-ray sources detected in a deep Chandra observation of the field surrounding the Abell 370 cluster. We combine these data with similar observations of the Chandra Deep Field-North and the Hawaii Survey Field SSA13 to obtain a sample of 69 hard X-ray sources (45 are spectroscopically identified) with extremely deep 20 cm observations. We find that about 4% of the >Lstar galaxy population is X-ray luminous at any time and hence that black hole accretion has a duration of about half a Gyr. We find that about 30% of the summed 2-7 keV flux from our total sample is from sources at z<1. We estimate the bolometric luminosities of accretion onto supermassive black holes and the mass inflow rates. The time history of the accretion rate density is evaluated, and its integrated value is reasonably consistent with the value inferred from the local black hole mass to bulge mass ratio.

astro-ph↗

arXiv2001

Multi-wavelength observations of serendipitous Chandra sources

We describe progress in a programme to study a sample of sources typical of those which contribute a large fraction of the hard X-ray background. The sources are selected from the fields of approximately 10 ks Chandra cluster observations with follow-up in the near-infrared and optical. The X-ray data indicate that many of these are powerful, obscured sources. Two hard objects which are lensed by the cluster are good candidates for X-ray type-2 quasars, with luminosities of the order 10^{44}-10^{45} erg s^{-1} and obscuring column densities of the order 10^{23} cm^{-2}. We find that the sources are bright in the infra-red (K typically 17-18). From Keck optical spectra and photometric redshifts, we find that the host galaxies are consistent with being early-type, massive hosts at redshifts ranging from 0.2 to 3, with median redshift approximately 1.5. The two obscured quasars also have mid-infrared detections and 850-micron upper-limits implying that the surrounding dust is at warm-to-hot temperatures (T of the order of 1000 K).

astro-ph↗

arXiv2001

Multi-wavelength observations of serendipitous Chandra X-ray sources in the field of A2390

We present optical spectra and near-infrared imaging of a sample of serendipitous X-ray sources detected in the field of Chandra observations of the A2390 cluster of galaxies. The sources have 0.5-7 keV fluxes of 0.6-8 x 10^{-14} erg cm^{-2} s^{-1} and lie around the break in the 2-10 keV source counts. They therefore are typical of sources dominating the X-ray Background in that band. Twelve of the fifteen targets for which we have optical spectra show emission lines, most of which have soft X-ray spectra. Including photometric redshifts and published spectra we have redshifts for seventeen of the sources, ranging from z~0.2 up to z~3 with a peak between z=1-2. Ten of our sources have hard X-ray spectra indicating a spectral slope flatter than that of a typical unabsorbed quasar. Two hard sources that are gravitationally lensed by the foreground cluster are obscured quasars, with intrinsic 2-10 keV luminosities of 0.2-3 x 10^{45} erg/s and absorbing columns of N_H>10^{23} cm^{-2}. Both were detected by ISO, showing that the absorbed flux emerges in the far-infrared.

astro-ph↗

arXiv2001

Detecting high redshift evolved galaxies as the hosts of optically faint hard X-ray sources

We combine deep Subaru near-infrared images of the massive lensing clusters A2390 and A370 with Keck optical data to map the spectral energy distributions (SEDs) of Chandra X-ray sources lying behind the clusters. The three sources behind A2390 are found to have extremely red colors with SEDs consistent with evolved galaxies at redshifts z>1.4. One source has extremely anomalous colors, which we interpret as evidence for a type Sa SED at a redshift around 2.5. The photometric redshift of another source has been confirmed at z=1.467 from near-infrared spectroscopy using the CISCO spectrograph on Subaru. Mapping of optically faint hard X-ray sources may prove to be an extremely efficient way to locate luminous evolved galaxies at high redshifts.

astro-ph↗

arXiv2001

The Chandra Deep Survey of the Hubble Deep Field North Area. II. Results from the Caltech Faint Field Galaxy Redshift Survey Area

We present results from a 221.9 ks Chandra exposure of the HDF-N and its vicinity, concentrating on the 8.6' X 8.7' area covered by the Caltech Faint Field Galaxy Redshift Survey (the `Caltech area'). The minimum detectable fluxes in the 0.5-2 keV and 2-8 keV bands are 1.3e-16 cgs and 6.5e-16 cgs, respectively and a total of 82 sources are detected. More than 80% of the extragalactic X-ray background in the 2-8 keV band is resolved. Redshifts are available for 96% of the sources with R<23; the redshift range is 0.1-3.5 with most sources having z < 1.5. Eight of the X-ray sources are located in the HDF-N itself, including two not previously reported. A population of X-ray faint, optically bright, nearby galaxies emerges at soft-band fluxes of ~< 3e-16 cgs. We set the tightest constraints to date on the X-ray emission properties of microJy radio sources, mid-infrared sources detected by ISO, and very red (R-K_s > 5.0) objects. Where both the infrared and the X-ray coverage are deepest, 75% of the X-ray sources are detected by ISO; the high X-ray to infrared matching rate bodes well for future sensitive infrared observations of faint X-ray sources. Four of the 33 very red objects that have been identified in the Caltech area by Hogg et al. (2000) are detected in X-rays; these four are among our hardest Chandra sources, and we argue that they contain moderately luminous obscured AGN. Overall, however, the small Chandra detection fraction suggests a relatively small AGN content in the optically selected very red object population. (Abridged)

astro-ph↗

arXiv2000

The Nature of the Hard X-ray Background Sources: Optical, Near-infrared, Submillimeter, and Radio Properties

With recent Chandra observations, at least 60% of the 2-10 keV background is now resolved into discrete sources. Here we present deep optical, NIR, submm, and 20 cm (radio) images, as well as high-quality optical spectra, of a complete sample of 20 hard X-ray sources in a deep Chandra observation of the SSA13 field. The thirteen I<23.5 galaxies have redshifts in the range 0.1 to 2.6. Two are quasars, five show AGN signatures, and six are z<1.5 luminous bulge-dominated galaxies whose spectra show no obvious optical AGN signatures. The seven spectroscopically unidentified sources have colors that are consistent with evolved early galaxies at z=1.5-3. Only one hard X-ray source is significantly detected in an ultradeep submm map; its millimetric redshift is in the range z=1.2-2.4. None of the remaining 19 sources is detected in the submm. These results probably reflect the fact that the 850-micron flux limits obtainable with SCUBA are quite close to the expected fluxes from obscured AGN. The hard X-ray sources have an average L(FIR)/L(2-10 keV)~60, similar to that of local obscured AGN. The same ratio for a sample of submm selected sources is in excess of 1100, suggesting that their FIR light is primarily produced by star formation. Our data show that luminous hard X-ray sources are common in bulge-dominated optically luminous galaxies. We use our measured bolometric corrections with the 2-10 keV EBL to infer the growth of supermassive black holes. Even with a high radiative efficiency of accretion (e=0.1), the black hole mass density required to account for the observed light is comparable to the local black hole mass density. (Abridged)

astro-ph↗

arXiv2000

Resolving the extragalactic hard X-ray background

The origin of the hard (2-10 keV) X-ray background has remained mysterious for over 35 years. Most of the soft (0.5-2 keV) X-ray background has been resolved into discrete sources, which are primarily quasars; however, these sources do not have the flat spectral shape required to match the X-ray background spectrum. Here we report the results of an X-ray survey 30 times more sensitive than previous studies in the hard band and four times more sensitive in the soft band. The sources detected in our survey account for at least 75 per cent of the hard X-ray background. The mean X-ray spectrum of these sources is in good agreement with that of the background. The X-ray emission from the majority of the detected sources is unambiguously associated with either the nuclei of otherwise normal bright galaxies or optically faint sources, which could either be active nuclei of dust enshrouded galaxies or the first quasars at very high redshifts.

astro-ph↗

arXiv2000

Mapping the Evolution of High Redshift Dusty Galaxies with Submillimeter Observations of a Radio-Selected Sample

Direct submm imaging has recently revealed the 850-micron background to be mostly composed of a population of distant ultraluminous infrared galaxies, but identifying the optical/NIR counterparts to these sources has proved difficult due to the poor submm spatial resolution. However, the proportionality of both cm and submm data to the star formation rate suggests that high resolution radio continuum maps with subarcsecond positional accuracy can be exploited to locate submm sources. In this paper we present results from a targeted SCUBA survey of micro-Jansky radio sources in the flanking fields of the Hubble Deep Field. Even with relatively shallow 850-micron SCUBA observations (>6 mJy at 3-sigma), we were successful at making submm detections of optical/NIR-faint (I>24 and K~21-22) radio sources, and our counts closely match the bright counts from submm surveys. Redshift estimates can be made from the ratio of the submm flux to the radio flux across the 100 GHz break in the spectral energy distribution. This millimetric redshift estimation places the bright submm population at z=1-3 where it forms the high redshift tail of the faint radio population. The star formation rate density (SFRD) due to ultraluminous infrared galaxies increases by more than two orders of magnitude from z~0 to z~1-3. The SFRD at high redshift inferred from our >6 mJy submm observations is comparable to that observed in the UV/optical. (Abridged)

astro-ph↗

arXiv1999

High Resolution Radio Imaging of Distant Submillimeter Galaxies

Using a combination of radio and optical imaging at 0.1\arcs to 0.2\arcs resolution with the VLA/MERLIN and HST has led to a breakthrough in our understanding of radio emission from distant (0.1 4-6) and submillimeter sources with S > 2 mJy at 850 microns as measured with SCUBA on the JCMT. The far-infrared luminosities of these galaxies exceeds even the most intense starbursts found in the local universe (e.g., Arp 220), suggesting they are in the process of converting the bulk of their gass mass into stars. These galaxies, completely absent in optical surveys, constitute 50% - 90% of the star-formation density in the distant Universe. Given the poor sub-mm resolution (15\arcs) of the SCUBA/JCMT images, we use the 0.2\arcs radio imaging as a surrogate in order to understand the nature of the dominant emission mechanism driving the FIR luminosity (AGN vs. star-formation). Upcoming developments in radio instrumentation (the Expanded VLA Array and Square Kilometer Array) will soon increase sensitivity and resolution orders of magnitude, providing a natural complement to parallel developments in sub-mm facilities (e.g., ALMA). With dual radio continuum and sub-mm surveys of the distant Universe, a census of galaxy evolution to the earliest cosmic epochs (z = 5-30) will soon be possible.

astro-ph↗

arXiv1999

Optically Faint Microjansky Radio Sources

We report on the identifications of radio sources from our survey of the Hubble Deep Field and the SSA13 fields, both of which comprise the deepest radio surveys to date at 1.4 GHz and 8.5 GHz respectively. About 80% of the microjansky radio sources are associated with moderate redshift starburst galaxies or AGNs within the I magnitude range of 17 to 24 with a median of I = 22 mag. Thirty-one (20%) of the radio sources are: 1) fainter than $I>$25 mag, with two objects in the HDF $I_{AB}>$28.5, 2) often identified with very red objects $I-K>$4, and 3) not significantly different in radio properties than the brighter objects. We suggest that most of these objects are associated with heavily obscured starburst galaxies with redshifts between 1 and 3. However, other mechanisms are discussed and cannot be ruled out with the present observations.

astro-ph↗

arXiv1999

The Discovery of ERO Counterparts to Faint Submm Galaxies

We use deep ground-based imaging in the near-IR to search for counterparts to the luminous submm sources in the catalog of Smail et al (1998). For the majority of the submm sources the near-IR imaging supports the counterparts originally selected from deep optical images. However, in two cases (10% of the sample) we find a relatively bright near-IR source close to the submm position, sources that were unidentified in the deep HST and ground-based R-band images used in Smail et al (1998). We place limits on colours of these sources from deep high-resolution Keck II imaging and find they have 2-sigma limits of (I-K)>6.8 and (I-K)>6.0 respectively. Both sources thus class as EROs. Using the spectral properties of the submm source in the radio and submm we argue that these EROs are probably the source of the submm emission, rather than the bright spiral galaxies previously identified by Smail et al. (1998). From the surface density of these submm-bright EROs we suggest that this class accounts for the majority of the reddest members of the ERO population, in good agreement with the preliminary conclusions of pointed submm observations of individual EROs. We conclude that the most extreme EROs represent a population of dusty, ultraluminous galaxies at high redshifts; further study of these will provide insights into the nature of star formation in obscured galaxies in the early Universe. The identification of similar counterparts in blank field submm surveys will be extremely difficult owing to their faintness (K~20.5, I>26.5). Finally, we discuss the radio and submm properties of the two submm-bright EROs discovered here and suggest that both galaxies lie at z>2.

astro-ph↗

arXiv1999

The view of galaxy formation from Hawaii: Seeing the dark side of the universe

The strength of the submillimeter background light shows directly that much of the energy radiated by star formation and AGN is moved to far infrared wavelengths. However, it is only as this background at 850 microns has been resolved with direct submillimeter imaging that we have seen that it is created by a population of ultraluminous (or near ultraluminous) infrared galaxies (ULIGs) which appear to lie at relatively high redshifts (z>1). Mapping the redshift evolution of this major portion of the universal star formation has been difficult because of the poor submillimeter spatial resolution, but this difficulty can be overcome by using extremely deep cm continuum radio observations to obtain precise astrometric information since the bulk of the brighter submillimeter sources have detectable radio counterparts. However, with this precise position information available, we find that most of the submillimeter sources are extremely faint in the optical and near infrared (I>>24 and K=21-22) and inaccessible to optical spectroscopy. Rough photometric redshift estimates can be made from combined radio and submillimeter energy distributions. We shall refer to this procedure as millimetric redshift estimation to distinguish it from photometric estimators in the optical and near IR. These estimators place the bulk of the submillimeter population at z=1-3, where it corresponds to the high redshift tail of the faint cm radio population. While still preliminary, the results suggest that the submillimeter population appears to dominate the star formation in this redshift range by almost an order of magnitude over the mostly distinct populations selected in the optical-ultraviolet.

astro-ph↗

arXiv1999

The Submillimeter Search for Very High Redshift Galaxies

Identifying the optical/near-infrared (NIR) counterparts to the distant submillimeter (submm) source population has proved difficult due to poor submm spatial resolution. However, the proportionality of both centimeter and submm data to the star formation rate suggests that high resolution radio continuum maps with subarcsecond positional accuracy could be exploited to locate submm sources. We targeted with SCUBA a sample of micro-Jy radio sources in the flanking fields of the Hubble Deep Field selected from the uniform (8 micro-Jy at 1-sigma) 1.4 GHz VLA image of Richards (1999). We find that the majority of bright (>6 mJy) submm sources have detectable radio counterparts. With the precise positions from the radio, we also find that these submm sources are extremely faint in the optical and NIR (I>>24 and K=21-22) and are therefore inaccessible to optical spectroscopy. Redshift estimates can, however, be made from the shape of the spectral energy distribution in the radio and submm. This procedure, which we refer to as millimetric redshift estimation, places the bright submm population at z=1-3, where it forms the high redshift tail of the faint radio population.

astro-ph↗

arXiv1999

Optical-Infrared Colour Distribution of a Statistically-Complete Sample of Faint Field Spheroidal Galaxies

In hierarchical models, where spheroidals are primarily produced via the merger of disk galaxies, the number of intrinsically red systems at faint limits will be substantially lower than that expected in "traditional" models where the bulk of star formation was completed at high redshifts. We analyse the optical--near-infrared colour distribution of a large flux-limited sample of field spheroidal galaxies identified morphologically from archival Hubble Space Telescope data. The $I_{814}-HK'$ colour distribution for a sample jointly limited at $I_{814}<$23 mag and $HK'<$19.5 mag is used to constrain the star formation history of elliptical and S0 galaxies. We detect a significant deficit of intrinsically red spheroids. From limited spectroscopic and deeper photometric redshift data, we demonstrate that this deficit results from a strong evolutionary effect. The space density of spheroidals of all colours at faint limits is lower than expected for a population of constant comoving density. We conclude that field ellipticals cannot have formed all of their stars at high redshift, as has been suggested for the bulk of their clustered counterparts. We compare alternative evolutionary histories with our data, including an analytical representation of recent models based on the hierarchical assembly of ellipticals, which provides a better, though not ideal fit. The accumulation of spectroscopic redshifts for our sample will allow a quantitative discrimination between models with extended star formation.

astro-ph↗

arXiv1999

Resolving the Submillimeter Background: the 850-micron Galaxy Counts

Recent deep blank field submillimeter surveys have revealed a population of luminous high redshift galaxies that emit most of their energy in the submillimeter. The results suggest that much of the star formation at high redshift may be hidden to optical observations. In this paper we present wide-area 850-micron SCUBA data on the Hawaii Survey Fields SSA13, SSA17, and SSA22. Combining these new data with our previous deep field data, we establish the 850-micron galaxy counts from 2 mJy to 10 mJy with a >3-sigma detection limit. The area coverage is 104 square arcmin to 8 mJy and 7.7 square arcmin to 2.3 mJy. The differential 850-micron counts are well described by the function n(S)=N_0/(a+S^3.2), where S is the flux in mJy, N_0=3.0 x 10^4 per square degree per mJy, and a=0.4-1.0 is chosen to match the 850-micron extragalactic background light. Between 20 to 30 per cent of the 850-micron background resides in sources brighter than 2 mJy. Using an empirical fit to our >2 mJy data constrained by the EBL at lower fluxes, we argue that the bulk of the 850-micron extragalactic background light resides in sources with fluxes near 1 mJy. The submillimeter sources are plausible progenitors of the present-day spheroidal population.

astro-ph↗

arXiv1999

Redshift Distribution of the Faint Submillimeter Galaxy Population

We present a Keck II LRIS spectroscopic follow-up study of the possible optical counterparts to a flux-limited sample of galaxies selected from an 850-micron survey of massive lensing clusters using the SCUBA bolometer array on the JCMT. These sources represent a population of luminous dusty galaxies responsible for the bulk of the 850-micron background detected by COBE and thus for a substantial fraction of the total far-infrared emission in the Universe. We present reliable redshifts for 20 galaxies and redshift limits for a further 4 galaxies selected from the error-boxes of 14 submm sources. Two other submm detections in the sample have no obvious optical counterparts, and the final submm source was only identified from imaging data after the completion of our spectroscopic observations. The optical identifications for 4 of the submm sources have been confirmed through either their detection in CO at mm-wavelengths (two pairs of galaxies at z=2.55 and z=2.80) or from the characteristics of their spectral energy distributions (two of the central cD galaxies in the lensing clusters). Plausible arguments based on the optical spectral properties (starburst or AGN signatures) of the counterparts allow us to identify a further two likely counterparts at z=1.06 and 1.16. For the remaining 8 cases, it is not always clear which, if any, of the optical sources identified are the true counterparts. Possible counterparts for these have redshifts ranging from z=0.18 to z=2.11. Working with the current identifications, we suggest that the majority of the extragalactic background light in the submm is emitted by sources at z<3 and hence that the peak activity in highly-obscured sources (both AGN and starbursts) lies at relatively modest redshifts. (Abridged)

astro-ph↗

arXiv1999

Molecular Gas in the z=2.565 Submillimeter Galaxy SMM J14011+0252

We report the detection of CO(3-2) emission from the submillimeter-selected luminous galaxy SMM J14011+0252. The optical counterpart of the submillimeter source has been identified as a merger system with spectral characteristics consistent with a starburst at z=2.565. The CO emission confirms the optical identification of the submillimeter source and implies a molecular gas mass of 5x10^{10}h_{75}^{-2} M(sun), after correcting for a lensing amplification factor of 2.75. The large molecular gas mass and the radio emission are consistent with the starburst interpretation of the source. These results are similar to those found for SMM J02399-0136, which was the first submillimeter selected CO source found at high redshift. The CO detections of these two high-redshift submillimeter galaxies suggest the presence of massive reservoirs of molecular gas which is consistent with the inferred high rates of star-formation (1000 M(sun)/yr). These two systems appear to be associated with merger events which may evolve into present day luminous elliptical galaxies.

astro-ph↗

arXiv1998

Constraints on the Early Formation of Field Elliptical Galaxies

We present the results of an HK' wide-field survey encompassing the Hubble Deep Field and its Flanking Fields. Our wide-field survey provides uniform coverage of an area 61.8 square arcmin to a depth equivalent to K=20.1 at 5-sigma. We have also imaged the Hubble Deep Field in HK', providing uniform coverage of an area 7.8 square arcmin to a depth equivalent to K=21.2 at 5-sigma. Using these data in combination with new deep UH 8K V and I imaging obtained on the CFHT, we find only a small population of objects with colors redder than an equivalent I-K=4, the color expected for an evolved elliptical at z>1. We infer that only a fraction of the local field elliptical population with M_K<-23.4 could have formed in single bursts at high redshift.

astro-ph↗