arXiv subjects

L. L. Cowie

Publications and source records attributed to L. L. Cowie.

At least 55 records · Page 3Linked to original sources

arXiv2008

A Highly Complete Spectroscopic Survey of the GOODS-N Field

We present a table of redshifts for 2907 galaxies and stars in the 145 square arcmin HST ACS GOODS-North, making this the most spectroscopically complete redshift sample obtained to date in a field of this size. We also include the redshifts, where available, in a table containing just under 7000 galaxies from the ACS area with K_s(AB)<24.5 measured from a deep K_s image obtained with WIRCam on the CFHT, as well as in a table containing 1016 sources with NUV(AB)<25 and 478 sources with FUV(AB)<25.5 (there is considerable overlap) measured from the deep GALEX images in the ACS area. Finally, we include the redshifts, where available, in a table containing the 1199 24 micron sources to 80 uJy measured from the wider-area Spitzer GOODS-North. The redshift identifications are greater than 90% complete to magnitudes of F435W(AB)=24.5, F850LP(AB)=23.3, and K_s(AB)=21.5 and to 24 micron fluxes of 250 uJy. An extensive analysis of these data will appear in a parallel paper, but here we determine how efficient color-selection techniques are at identifying high-redshift galaxies and Active Galactic Nuclei. We also examine the feasibility of doing tomography of the intergalactic medium with a 30 m telescope.

astro-ph↗

arXiv2008

Measuring the Sources of the Intergalactic Ionizing Flux

We use a wide-field (0.9 square degree) X-ray sample with optical and GALEX ultraviolet observations to measure the contribution of Active Galactic Nuclei (AGNs) to the ionizing flux as a function of redshift. Our analysis shows that the AGN contribution to the metagalactic ionizing background peaks around z=2. The measured values of the ionizing background from the AGNs are lower than previous estimates and confirm that ionization from AGNs is insufficient to maintain the observed ionization of the intergalactic medium (IGM) at z>3. We show that only sources with broad lines in their optical spectra have detectable ionizing flux and that the ionizing flux seen in an AGN is not correlated with its X-ray color. We also use the GALEX observations of the GOODS-N region to place a 2-sigma upper limit of 0.008 on the average ionization fraction fnu(700 A)/fnu(1500 A) for 626 UV selected galaxies in the redshift range z=0.9-1.4. We then use this limit to estimate an upper bound to the galaxy contribution in the redshift range z=0-5. If the z~1.15 ionization fraction is appropriate for higher redshift galaxies, then contributions from the galaxy population are also too low to account for the IGM ionization at the highest redshifts (z>4).

astro-ph↗

arXiv2008

The OPTX Project I: The Flux and Redshift Catalogs for the CLANS, CLASXS, and CDF-N fields

We present the redshift catalogs for the X-ray sources detected in the Chandra Deep Field North (CDF-N), the Chandra Large Area Synoptic X-ray Survey (CLASXS), and the Chandra Lockman Area North Survey (CLANS). The catalogs for the CDF-N and CLASXS fields include redshifts from previous work, while the redshifts for the CLANS field are all new. For fluxes above 10^-14 ergs cm^-2 s^-1 (2-8 keV) we have redshifts for 76% of the sources. We extend the redshift information for the full sample using photometric redshifts. The goal of the OPTX Project is to use these three surveys, which are among the most spectroscopically complete surveys to date, to analyze the effect of spectral type on the shape and evolution of the X-ray luminosity functions and to compare the optical spectral types with the X-ray spectral properties. We also present the CLANS X-ray catalog. The nine ACIS-I fields cover a solid angle of ~0.6 square degrees and reach fluxes of 7x10^-16 ergs cm^-2 s^-1 (0.5-2 keV) and 3.5x10^-15 ergs cm^-2 s^-1 (2-8 keV). We find a total of 761 X-ray point sources. Additionally, we present the optical and infrared photometric catalog for the CLANS X-ray sources, as well as updated optical and infrared photometric catalogs for the X-ray sources in the CLASXS and CDF-N fields. The CLANS and CLASXS surveys bridge the gap between the ultradeep pencil-beam surveys, such as the CDFs, and the shallower, very large-area surveys. As a result, they probe the X-ray sources that contribute the bulk of the 2-8 keV X-ray background and cover the flux range of the observed break in the logN-logS distribution. We construct differential number counts for each individual field and for the full sample.

astro-ph↗

arXiv2008

An Integrated Picture of Star Formation, Metallicity Evolution, and Galactic Stellar Mass Assembly

We present an integrated study of star formation and galactic stellar mass assembly from z=0.05-1.5 and galactic metallicity evolution from z=0.05-0.9 using a very large and highly spectroscopically complete sample selected by rest-frame NIR bolometric flux in the GOODS-N. We assume a Salpeter IMF and fit Bruzual & Charlot (2003) models to compute the galactic stellar masses and extinctions. We determine the expected formed stellar mass density growth rates produced by star formation and compare them with the growth rates measured from the formed stellar mass functions by mass interval. We show that the growth rates match if the IMF is slightly increased from the Salpeter IMF at intermediate masses (~10 solar masses). We investigate the evolution of galaxy color, spectral type, and morphology with mass and redshift and the evolution of mass with environment. We find that applying extinction corrections is critical when analyzing galaxy colors; e.g., nearly all of the galaxies in the green valley are 24um sources, but after correcting for extinction, the bulk of the 24um sources lie in the blue cloud. We find an evolution of the metallicity-mass relation corresponding to a decrease of 0.21+/-0.03 dex between the local value and the value at z=0.77 in the 1e10-1e11 solar mass range. We use the metallicity evolution to estimate the gas mass of the galaxies, which we compare with the galactic stellar mass assembly and star formation histories. Overall, our measurements are consistent with a galaxy evolution process dominated by episodic bursts of star formation and where star formation in the most massive galaxies (>1e11 solar masses) ceases at z<1.5 because of gas starvation. (Abstract abridged)

astro-ph↗

arXiv2007

Properties of a Gamma Ray Burst Host Galaxy at z ~ 5

We describe the properties of the host galaxy of the gamma-ray burst GRB060510B based on a spectrum of the burst afterglow obtained with the Gemini North 8m telescope. The galaxy lies at a redshift of z = 4.941 making it the fourth highest spectroscopically identified burst host. However, it is the second highest redshift galaxy for which the quality of the spectrum permits a detailed metallicity analysis. The neutral hydrogen column density has a logarithmic value of 21.0--21.2 cm^-2 and the weak metal lines of Ni, S and Fe show that the metallicity is in excess of a tenth of solar which is far above the metallicities in damped Lyman alpha absorbers at high redshift. The tightest constraint is from the Fe lines which place [Fe/H] in excess of -0.8. We argue that the results suggest that metallicity bias could be a serious problem with inferring star formation from the GRB population and consider how future higher quality measurements could be used to resolve this question.

astro-ph↗

arXiv2007

A New Population of High Redshift Short-Duration Gamma-Ray Bursts

The redshift distribution of the short-duration GRBs is a crucial, but currently fragmentary, clue to the nature of their progenitors. Here we present optical observations of nine short GRBs obtained with Gemini, Magellan, and the Hubble Space Telescope. We detect the afterglows and host galaxies of two short bursts, and host galaxies for two additional bursts with known optical afterglow positions, and five with X-ray positions (<6'' radius). In eight of the nine cases we find that the most probable host galaxies are faint, R~23-26.5 mag, and are therefore starkly different from the first few short GRB hosts with R~17-22 mag and z<0.5. Indeed, we measure spectroscopic redshifts of z~0.4-1.1 for the four brightest hosts. A comparison to large field galaxy samples, as well as the hosts of long GRBs and previous short GRBs, indicates that the fainter hosts likely reside at z>1. Our most conservative limit is that at least half of the five hosts without a known redshift reside at z>0.7 (97% confidence level), suggesting that about 1/3-2/3 of all short GRBs originate at higher redshifts than previously determined. This has two important implications: (i) We constrain the acceptable age distributions to a wide lognormal (sigma>1) with tau~4-8 Gyr, or to a power law, P(tau)~tau^n, with -1<n<0; and (ii) the inferred isotropic energies, E_{gamma,iso}~10^50-10^52 erg, are significantly larger than ~10^48-10^49 erg for the low redshift short GRBs, indicating a large spread in energy release or jet opening angles. Finally, we re-iterate the importance of short GRBs as potential gravitational wave sources and find a conservative Advanced LIGO detection rate of ~2-6 yr^-1.

astro-ph↗

arXiv2006

The MicroJansky Radio Galaxy Population

We use highly spectroscopically complete observations of the radio sources from the VLA 1.4 GHz survey of the HDF-N region to study the faint radio galaxy population and its evolution. We spectrally classify the sources into four spectral types: absorbers, star formers, Seyfert galaxies, and broad-line AGNs, and we analyze their properties by type. We supplement the spectroscopic redshifts with photometric redshifts measured from the rest-frame UV to MIR spectral energy distributions. Using deep X-ray observations of the field, we do not confirm the existence of an X-ray-radio correlation for star-forming galaxies. We also do not observe any correlations between 1.4 GHz flux and R magnitude or redshift. We find that the radio powers of the host galaxies rise dramatically with increasing redshift, while the optical properties of the host galaxies show at most small changes. Assuming that the locally determined FIR-radio correlation holds at high redshifts, we estimate total FIR luminosities for the radio sources. We note that the FIR luminosity estimates for any radio-loud AGNs will be overestimates. Considering only the radio sources with quasar-like bolometric luminosities, we find a maximum ratio of candidate highly-obscured AGNs to X-ray-luminous (>10^42 ergs/s) sources of about 1.9. We use source-stacking analyses to measure the X-ray surface brightnesses of various X-ray and radio populations. We find the contributions to the 4-8 keV light from our candidate highly-obscured AGNs to be very small, and hence these sources are unable to account for the light that has been suggested may be missing at these energies.

astro-ph↗

arXiv2006

A Near-Infrared Analysis of the Submillimeter Background and the Cosmic Star-Formation History

We use new deep near-infrared (NIR) and mid-infrared (MIR) observations to analyze the 850$~μ$m image of the GOODS HDF-N region. We show that much of the submillimeter background at this wavelength is picked out by sources with $H(AB)$ or 3.6um (AB)<23.25 (1.8 uJy). These sources contribute an 850um background of 24\pm2 Jy deg^-2. This is a much higher fraction of the measured background (31-45 Jy deg^-2) than is found with current 20cm or 24um samples. Roughly one-half of these NIR-selected sources have spectroscopic identifications, and we can assign robust photometric redshifts to nearly all of the remaining sources using their UV to MIR spectral energy distributions. We use the redshift and spectral type information to show that a large fraction of the 850um background light comes from sources with z=0-1.5 and that the sources responsible have intermediate spectral types. Neither the elliptical galaxies, which have no star formation, nor the bluest galaxies, which have little dust, contribute a significant amount of 850um light, despite the fact that together they comprise approximately half of the galaxies in the sample. The galaxies with intermediate spectral types have a mean flux of 0.40\pm0.03 mJy at 850um and 9.1\pm0.3 uJy at 20cm. The redshift distribution of the NIR-selected 850um light lies well below that of the much smaller amount of light traced by the more luminous, radio- selected submillimeter sources. We therefore require a revised star-formation history with a lower star-formation rate at high redshifts. We use a stacking analysis of the 20cm light in the NIR sample to show that the star-formation history of the total 850um sample is relatively flat down to z~1 and that half of the total star formation occurs at redshifts z<1.4.

astro-ph↗

arXiv2006

HST and Spitzer Observations of the Host Galaxy of GRB 050904: A Metal-Enriched, Dusty Starburst at z=6.295

We present deep Hubble Space Telescope and Spitzer Space Telescope observations of the host galaxy of GRB 050904 at z=6.295. The host is detected in the H-band and marginally at 3.6 micron. From these detections, and limits in the z'-band and 4.5 micron, we infer an extinction-corrected absolute magnitude, M(UV)=-20.7 mag, or ~L*, a substantial star formation rate of 15 solar masses per year, and a stellar mass of a few 10^9 solar masses. A comparison to the published sample of spectroscopically-confirmed galaxies at z>5.5 reveals that the host of GRB 050904 would evade detection and/or confirmation in any of the current surveys due to the lack of detectable Ly-alpha emission, which is likely the result of dust extinction (A[1200]~1.5 mag). This suggests that not all luminous starburst galaxies at z~6 are currently being accounted for. Most importantly, using the metallicity of Z~0.05 solar inferred from the afterglow absorption spectrum, our observations indicate for the first time that the observed evolution in the mass- and luminosity-metallicity relations from z=0 to z~2 continues on to z>6. The ease of measuring redshifts and metallicities from the afterglow emission suggests that in tandem with the next generation ground- and space-based telescopes, a GRB mission with dedicated near-IR follow-up can provide unique information on the evolution of stars and galaxies through the epoch of re-ionization.

astro-ph↗

arXiv2006

Cosmological Implications of the Very High Redshift GRB 050904

We report near simultaneous multi-color (RIYJHK) observations made with the MAGNUM 2m telescope of the gamma ray burst GRB 050904 detected by the SWIFT satellite. The spectral energy distribution shows a very large break between the I and J bands. Using intergalactic transmissions measured from high redshift quasars we show that the observations place a 95% confidence lower limit of z=6.18 on the object, consistent with a later measured spectroscopic redshift of 6.29 obtained by Kawai et al. (2005) with the Subaru telescope. We show that the break strength in the R and I bands is consistent with that measured in the quasars. Finally we consider the implications for the star formation history at high redshift.

astro-ph↗

arXiv2006

Spatial Correlation Function of the Chandra Selected Active Galactic Nuclei

We present the spatial correlation function analysis of non-stellar X-ray point sources in the Chandra Large Area Synoptic X-ray Survey of Lockman Hole Northwest (CLASXS). Our 9 ACIS-I fields cover a contiguous solid angle of 0.4 deg^2 and reach a depth of 3x10^-15 c.g.s in the 2-8 keV band. We supplement our analysis with data from the Chandra Deep Field North (CDFN). The addition of this field allows better probe of the correlation function at small scales. A total of 233 and 252 sources with spectroscopic information are used in the study of the CLASXS and CDFN fields respectively. We calculate both redshift-space and projected correlation functions in comoving coordinates, averaged over the redshift range of 0.1<z<3.0, for both CLASXS and CDFN fields for a standard cosmology with Ω_Λ = 0.73, Ω_{M} = 0.27, and H_0 = 0.71. The correlation function for the CLASXS field over scales of 3 Mpc <s< 200 Mpc can be modeled as a power-law of the form ξ(s) = (s/s_0)^{-γ}, with γ= 1.6^{+0.4}_{-0.3} and s_0 = 8.0^{+1.4}_{-1.5} Mpc. The redshift-space correlation function for CDFN on scales of 1 Mpc$<s<$~100 Mpc is found to have a similar correlation length $s_0 = 8.55^{+0.75}_{-0.74}$ Mpc, but a shallower slope ($γ= 1.3 \pm 0.1$). The real-space correlation functions derived from the projected correlation functions, are found to be $r_0 = 8.1^{+1.2}_{-2.2}$ Mpc, and $γ= 2.1 \pm 0.5$ for the CLASXS field, and $r_0 = 5.8^{+1.0}_{-1.5}$ Mpc, $γ= 1.38^{+0.12}_{-0.14}$ for the CDFN field. By comparing the real- and redshift-space correlation functions in the combined CLASXS and CDFN samples, we are able to estimate the redshift distortion parameter $β= 0.4 \pm 0.2$ at an effective redshift $z = 0.94$.(abridged)

astro-ph↗

arXiv2005

The afterglow and elliptical host galaxy of the short gamma-ray burst GRB 050724

Despite a rich phenomenology, gamma-ray bursts (GRBs) are divided into two classes based on their duration and spectral hardness -- the long-soft and the short-hard bursts. The discovery of afterglow emission from long GRBs was a watershed event, pinpointing their origin to star forming galaxies, and hence the death of massive stars, and indicating an energy release of about 10^51 erg. While theoretical arguments suggest that short GRBs are produced in the coalescence of binary compact objects (neutron stars or black holes), the progenitors, energetics, and environments of these events remain elusive despite recent localizations. Here we report the discovery of the first radio afterglow from a short burst, GRB 050724, which unambiguously associates it with an elliptical galaxy at a redshift, z=0.257. We show that the burst is powered by the same relativistic fireball mechanism as long GRBs, with the ejecta possibly collimated in jets, but that the total energy release is 10-1000 times smaller. More importantly, the nature of the host galaxy demonstrates that short GRBs arise from an old (>1 Gyr) stellar population, strengthening earlier suggestions, and providing support for coalescing compact object binaries as the progenitors.

astro-ph↗

arXiv2005

The Number Density of Intermediate and High Luminosity Active Galactic Nuclei at z~2-3

We use the combination of the 2 Ms Chandra X-ray image, new J and H band images, and the Spitzer IRAC and MIPS images of the Chandra Deep Field-North to obtain high spectroscopic and photometric redshift completeness of high and intermediate X-ray luminosity sources in the redshift interval z=2-3. We measure the number densities of z=2-3 active galactic nuclei (AGNs) and broad-line AGNs in the rest-frame 2-8 keV luminosity intervals 10^44-10^45 and 10^43-10^44 ergs/s and compare with previous lower redshift results. We confirm a decline in the number densities of intermediate-luminosity sources at z>1. We also measure the number density of z=2-3 AGNs in the luminosity interval 10^43-10^44.5 ergs/s and compare with previous low and high-redshift results. Again, we find a decline in the number densities at z>1. In both cases, we can rule out the hypothesis that the number densities remain flat to z=2-3 at above the 5-sigma level.

astro-ph↗

arXiv2005

Optical identification of ISO far-infrared sources in the Lockman Hole using a deep VLA 1.4 GHz continuum survey

By exploiting the far-infrared(FIR) and radio correlation, we have performed the Likelihood-Ratio analysis to identify optical counterparts to the far-infrared sources in the Lockman Hole. Using the likelihood ratio analysis and the associated reliability, 44 FIR sources have been identified with radio sources. Redshifts have been obtained for 29 out of 44 identified sources. One hyper-luminous infrared galaxy (HyLIRG) with and four ultraluminous infrared galaxies (ULIRGs) are identified in our sample. The space density of the FIR sources at z = 0.3-0.6 is 4.6\times 10^{-5}Mpc^{-3}, implying a rapid evolution of the ULIRG population. Most of \ISO FIR sources have their FIR-radio ratios similar to star-forming galaxies ARP 220 and M82. At least seven of our FIR sources show evidence for the presence of an active galactic nucleus (AGN) in optical emission lines, radio continuum excess, or X-ray activity. Three out of five (60%) of the ULIRG/HyLIRGs are AGN galaxies. Five of the seven AGN galaxies are within the ROSAT X-ray survey field, and two are within the XMM-Newton survey fields. X-ray emission has been detected in only one source, 1EX030, which is optically classified as a quasar. The non-detection in the XMM-Newton 2-10 keV band suggests a very thick absorption obscuring the central source of the two AGN galaxies. Several sources have an extreme FIR luminosity relative to the optical R-band, L(90μ\mathrm{m})/L(R) > 500, which is rare even among the local ULIRG population. While source confusion or blending might offer an explanation in some cases, they may represent a new population of galaxies with an extreme activity of star formation in an undeveloped stellar system -- i.e., formation of bulges or young ellipticals.

astro-ph↗

arXiv2005

The Afterglows, Redshifts, and Properties of Swift Gamma-Ray Bursts

We present optical, near-IR, and radio follow up of sixteen Swift bursts, including our discovery of nine afterglows and a redshift determination for three. These observations, supplemented by data from the literature, provide an afterglow recovery rate of 60% in the optical/near-IR, much higher than in previous missions (BeppoSAX, HETE-2, INTEGRAL, and IPN). The optical/near-IR afterglows of Swift events are on average 1.7 mag fainter at t=12 hr than those of previous missions. The X-ray afterglows are similarly fainter compared to those of pre-Swift bursts. In the radio the limiting factor is the VLA threshold and the detection rate for Swift bursts is similar to that for past missions. The redshift distribution of pre-Swift bursts peaked at z~1, whereas the five Swift bursts with measured redshifts are distributed evenly between 1.3 and 3.2. From these results we conclude that (i) the pre-Swift distributions were biased in favor of bright events and low redshift events, (ii) the higher sensitivity and accurate positions of Swift result in a better representation of the true burst redshift and brightness distributions (which are higher and dimmer, respectively), and (iii) as many as 1/3 of the bursts can be optically dark, as a result of a high redshift and/or dust extinction. We remark that the apparent lack of low redshift, low luminosity Swift bursts, and the lower event rate compared to pre-launch estimates (90 vs. 150 per year), are the result of a threshold that is similar to that of BATSE. In view of these inferences, afterglow observers may find it advisable to make significant changes in follow up strategies of Swift events. [abridged]

astro-ph↗

arXiv2004

The Cosmic Evolution of Hard X-ray Selected Active Galactic Nuclei

We use highly spectroscopically complete deep and wide-area Chandra surveys to determine the cosmic evolution of hard X-ray-selected AGNs. We determine hard X-ray luminosity functions (HXLFs) for all spectral types and for broad-line AGNs (BLAGNs) alone. At z<1.2, both are well described by pure luminosity evolution. Thus, all AGNs drop in luminosity by almost an order of magnitude over this redshift range. We show that this observed drop is due to AGN downsizing. We directly compare our BLAGN HXLFs with the optical QSO LFs and find that the optical QSO LFs do not probe faint enough to see the downturn in the BLAGN HXLFs. We rule out galaxy dilution as a partial explanation for the observation that BLAGNs dominate the number densities at the higher X-ray luminosities, while optically-narrow AGNs (FWHM<2000 km/s) dominate at the lower X-ray luminosities by measuring the nuclear UV/optical properties of the Chandra sources using the HST ACS GOODS-North data. The UV/optical nuclei of the optically-narrow AGNs are much weaker than expected if they were similar to the BLAGNs. We therefore postulate the need for a luminosity dependent unified model. Alternatively, the BLAGNs and the optically-narrow AGNs could be intrinsically different source populations. We cover both interpretations by constructing composite spectral energy distributions--including long-wavelength data from the MIR to the submillimeter--by spectral type and by X-ray luminosity. We use these to infer the bolometric corrections (from hard X-ray luminosities to bolometric luminosities) needed to map the accretion history. We determine the accreted supermassive black hole mass density for all spectral types and for BLAGNs alone using the observed evolution of the hard X-ray energy density production rate and our inferred bolometric corrections.

astro-ph↗

arXiv2004

The Chandra Large Area Synoptic X-ray Survey (CLASXS) of the Lockman Hole-Northwest: The X-ray Catalog

We present the X-ray catalog and basic results from our Chandra Large Area Synoptic X-ray Survey (CLASXS) of the Lockman Hole-Northwest field. Our 9 ACIS-I fields cover a contiguous solid angle of ~0.4 sq. deg. and reach fluxes of 5E-16 cgs (0.4-2 keV) and 3E-15 cgs (2-8keV). Our survey bridges the gap between ultradeep pencil-beam surveys, such as the Chandra Deep Fields (CDFs), and shallower, large area surveys, allowing a better probe of the X-ray sources that contribute most of the 2-10 keV cosmic X-ray background (CXB). We find a total of 525 X-ray point sources and 4 extended sources. At ~10E-14 cgs 2-8 keV, our number counts are significantly higher than those of several non-contiguous, large area surveys. On the other hand, the integrated flux from the CLASXS field, combined with ASCA and Chandra ultradeep surveys, is consistent with results from other large area surveys, within the variance of the CXB. Spectral evolution is seen in the hardening of the sources at fluxes below 1E-14 cgs Above 4E1-14 cgs(0.4-8 keV), ~60 of the sources are variable. Four extended sources in CLASXS is consistent with the previously measured LogN-LogS of galaxy clusters. We report the discovery of a gravitational lensing arc associated with one of these sources. (Abridged)

astro-ph↗

arXiv2004

An Optical Catalog of the Chandra Large Area Synoptic X-ray Survey Sources

We present photometric and spectroscopic observations of the X-ray sources detected in the wide-area, moderately deep Chandra Large Area Synoptic X-ray Survey of the Lockman Hole-Northwest field. We have B, V, R, I, and z' photometry for 521 (99%) of the 525 sources in the X-ray catalog and spectroscopic redshifts for 271 (52%), including 20 stars. We do not find evidence for redshift groupings of the X-ray sources, like those found in the Chandra Deep Field surveys, because of the larger solid angle covered by this survey. We separate the X-ray sources by optical spectral type and examine the colors, apparent and absolute magnitudes, and redshift distributions for the broad-line and non-broad-line active galactic nuclei. Combining our wide-area survey with other Chandra and XMM-Newton hard X-ray surveys, we find a definite lack of luminous, high accretion rate sources at z<1, consistent with previous observations that showed that supermassive black hole growth is dominated at low redshifts by sources with low accretion rates.

astro-ph↗