arXiv subjects

C. R. Keeton

Publications and source records attributed to C. R. Keeton.

At least 37 records · Page 2Linked to original sources

arXiv2003

A Gravitationally Lensed Quasar with Quadruple Images Separated by 14.62 Arcseconds

Gravitational lensing is a powerful tool for the study of the distribution of dark matter in the Universe. The cold-dark-matter model of the formation of large-scale structures predicts the existence of quasars gravitationally lensed by concentrations of dark matter so massive that the quasar images would be split by over 7 arcsec. Numerous searches for large-separation lensed quasars have, however, been unsuccessful. All of the roughly 70 lensed quasars known, including the first lensed quasar discovered, have smaller separations that can be explained in terms of galaxy-scale concentrations of baryonic matter. Although gravitationally lensed galaxies with large separations are known, quasars are more useful cosmological probes because of the simplicity of the resulting lens systems. Here we report the discovery of a lensed quasar, SDSS J1004+4112, which has a maximum separation between the components of 14.62 arcsec. Such a large separation means that the lensing object must be dominated by dark matter. Our results are fully consistent with theoretical expectations based on the cold-dark-matter model.

astro-ph↗

arXiv2003

Gravitational Lensing Statistics in Universes Dominated by Dark Energy

We study lens statistics in flat, low-density universes with different equations of state $w=p_Q/ρ_Q$ for the dark energy component. Dark energy modifies the distance-redshift relation and the mass function of dark matter halos leading to changes in the lensing optical depth as a function of image separation. Those effects must, however, be distinguished from effects associated with the structure of dark matter halos. Baryonic cooling causes galaxy-mass halos to have different central density profiles than group- and cluster-mass halos, which causes the distribution of normal arcsecond-scale lenses to differ from the distribution of ``wide-separation'' ($Δθ\gtrsim 4\arcsec$) lenses. Fortunately, the various parameters related to cosmology and halo structure have very different effects on the overall image separation distribution: (1) the abundance of wide-separation lenses is exremely sensitive (by orders of magnitude) to the distribution of ``concentration'' parameters for massive halos modeled with the Navarro-Frenk-White profile; (2) the transition between normal and wide-separation lenses depends mainly on the mass scale where baryonic cooling ceases to be efficient; and (3) dark energy has effects at all image separation scales. While current lens samples cannot usefully constrain all of the parameters, ongoing and future imaging surveys should discover hundreds or thousands of lenses and make it possible to disentangle the various effects and constrain all of the parameters simultaneously. (abridged)

astro-ph↗

arXiv2003

Self-Similar Models for the Mass Profiles of Early-type Lens Galaxies

We introduce a self-similar mass model for early-type galaxies, and constrain it using the aperture mass-radius relations determined from the geometries of 22 gravitational lenses. The model consists of two components: a concentrated component which traces the light distribution, and a more extended power-law component (rho propto r^-n) which represents the dark matter. We find that lens galaxies have total mass profiles which are nearly isothermal, or slightly steeper, on the several-kiloparsec radial scale spanned by the lensed images. In the limit of a single-component, power-law radial profile, the model implies n=2.07+/-0.13, consistent with isothermal (n=2). Models in which mass traces light are excluded at >99 percent confidence. An n=1 cusp (such as the Navarro-Frenk-White profile) requires a projected dark matter mass fraction of f_cdm = 0.22+/-0.10 inside 2 effective radii. These are the best statistical constraints yet obtained on the mass profiles of lenses, and provide clear evidence for a small but non-zero dark matter mass fraction in the inner regions of early-type galaxies. In addition, we derive the first strong lensing constraint on the relation between stellar mass-to-light ratio (Upsilon) and galaxy luminosity (L): Upsilon propto L^[0.14 (+0.16)(-0.12)], which is consistent with the relation suggested by the fundamental plane. Finally, we apply our self-similar mass models to current problems regarding the interpretation of time delays and flux ratio anomalies in gravitational lens systems.

astro-ph↗

arXiv2002

The Evolution of a Mass-Selected Sample of Early-Type Field Galaxies

We investigate the evolution of mass-selected early-type field galaxies using a sample of 28 gravitational lenses spanning the redshift range 0 < z < 1. Based on the redshift-dependent intercept of the fundamental plane in the rest frame B band, we measure an evolution rate of d log (M/L)_B / dz = -0.56 +/- 0.04 (all errors are 1 sigma unless noted) if we directly compare to the local intercept measured from the Coma cluster. Re-fitting the local intercept helps minimize potential systematic errors, and yields an evolution rate of d log (M/L)_B / dz = -0.54 +/- 0.09. An evolution analysis of properly-corrected aperture mass-to-light ratios (defined by the lensed image separations) is closely related to the Faber-Jackson relation. In rest frame B band we find an evolution rate of d log (M/L)_B / dz = -0.41 +/- 0.21, a present-day characteristic magnitude of M_{*0} = -19.70 + 5 log h +/- 0.29 (assuming a characteristic velocity dispersion of sigma_{DM*} = 225 km/s), and a Faber-Jackson slope of gamma_{FJ} = 3.29 +/- 0.58. The measured evolution rates favor old stellar populations (mean formation redshift z_f > 1.8 at 2 sigma confidence for a Salpeter initial mass function and a flat Omega_m =0.3 cosmology) among early-type field galaxies, and argue against significant episodes of star formation at z < 1.

astro-ph↗

arXiv2001

Cusped Mass Models Of Gravitational Lenses

Recent observations of galaxy luminosity profiles and dark matter simulations find luminosity and mass distributions characterized by central cusps rather than finite core radii. We introduce and implement a set of cusped ellipsoidal lens models which include limits similar to the Jaffe, Hernquist, ηand NFW models and apply them to the gravitational lenses APM 08279+5255 and B 1933+503. A successful model of APM 08279+5255 with its central, odd image requires a very shallow cusp, $γ\ltorder 0.4$ where $ρ\propto r^{-γ}$ as $r\to 0$, which is similar to a core rather than the favored $1 \ltorder γ\ltorder 2$ cusps. B~1933+503, by contrast, is well modeled with a steep density cusp, $1.6 \ltorder γ\ltorder 2.0$.

astro-ph↗

arXiv2001

Constraints on Galaxy Density Profiles from Strong Gravitational Lensing: The Case of B 1933+503

We consider a wide range of parametric mass models for B 1933+503, a ten-image radio lens, and identify shared properties of the models with the best fits. The approximate rotation curves varies by less than 8.5% from the average value between the innermost and the outermost image (1.5h^{-1} kpc to 4.1h^{-1} kpc) for models within 1 σof the best fit, and the radial dependence of the shear strength and angle also have common behavior for the best models. The time delay between images 1 and 6, the longest delay between the radio cores, is Δt = (10.6^{+2.4}_{-1.1})h^{-1} days (Ω_0=0.3, λ_0=0.7) including all the modeling uncertainties. Deeper infrared observations, to more precisely register the lens galaxy with the radio images and to measure the properties of the Einstein ring image of the radio source's host galaxy, would significantly improve the model constraints and further reduce the uncertainties in the mass distribution and time delay.

astro-ph↗

arXiv2000

B1359+154: A Six Image Lens Produced by a z=1 Compact Group of Galaxies

HST V and I-band observations show that the gravitational lens B1359+154 consists of six images of a single z_s=3.235 radio source and its star-forming host galaxy, produced by a compact group of galaxies at z_l = 1. VLBA observations at 1.7 GHz strongly support this conclusion, showing six compact cores with similar low-frequency radio spectra. B1359+154 is the first example of galaxy-scale gravitational lensing in which more than four images are observed of the same background source. The configuration is due to the unique lensing mass distribution: three primary lens galaxies lying on the vertices of a triangle separated by 0.7 arcsec (4/h kpc), inside the 1.7 arcsec diameter Einstein ring defined by the radio images. The gravitational potential has additional extrema within this triangle, creating a pair of central images that supplement the ``standard'' four-image geometry of the outer components. Simple mass models consisting of three lens galaxies constrained by HST and VLBA astrometry naturally reproduce the observed image positions but must be finely-tuned to fit the flux densities.

astro-ph↗

arXiv2000

Multi-Frequency Analysis of the New Wide-Separation Gravitational Lens Candidate RX J0921+4529

We report the discovery of a new two-image gravitational lens candidate. The system RX J0921+4529 contains two z_s=1.66 quasars separated by 6."93 with an H band magnitude difference of $Δm=1.39$. The HST NIC2 H band images reveal an H=18.2 spiral galaxy between the quasar images, which is probably a member of a z_l=0.32 X-ray cluster centered on the field. We detect an extended source near the fainter quasar image but not in the brighter image. If this extended source is the host galaxy of the fainter quasar, then the system is a binary quasar rather than a gravitational lens. VLA observations at 3.6 cm reveal emission from the lens galaxy at the flux level of 1 mJy and a marginal detection of the brighter quasar.

astro-ph↗

arXiv2000

What Fraction of Gravitational Lens Galaxies Lie in Groups?

We predict how the observed variations in galaxy populations with environment affect the number and properties of gravitational lenses in different environments. Two trends dominate: lensing strongly favors early-type galaxies, which tend to lie in dense environments, but dense environments tend to have a larger ratio of dwarf to giant galaxies than the field. The two effects nearly cancel, and the distribution of environments for lens and non-lens galaxies are not substantially different (lens galaxies are slightly less likely than non-lens galaxies to lie in groups and clusters). We predict that about 20% of lens galaxies are in bound groups (defined as systems with a line-of-sight velocity dispersion sigma in the range 200 < sigma < 500 km/s), and another roughly 3% are in rich clusters (sigma > 500 km/s). Therefore at least roughly 25% of lenses are likely to have environments that significantly perturb the lensing potential. If such perturbations do not significantly increase the image separation, we predict that lenses in groups have a mean image separation that is about 0.2'' smaller than that for lenses in the field and estimate that 20-40 lenses in groups are required to test this prediction with significance. The tail of the distribution of image separations is already illuminating. Although lensing by galactic potential wells should rarely produce lenses with image separations theta >~ 6'', two such lenses are seen among 49 known lenses, suggesting that environmental perturbations of the lensing potential can be significant. Further comparison of theory and data will offer a direct probe of the dark halos of galaxies and groups and reveal the extent to which they affect lensing estimates of cosmological parameters.

astro-ph↗

arXiv2000

The Importance of Einstein Rings

We develop a theory of Einstein rings and demonstrate it using the infrared Einstein ring images of the quasar host galaxies observed in PG1115+080, B1608+656 and B1938+666. The shape of an Einstein ring accurately and independently determines the shape of the lens potential and the shape of the lensed host galaxy. We find that the host galaxies of PG1115+080, B1608+656 and B1938+666 have axis ratios of 0.58+/-0.02, 0.69+/-0.02 and 0.62+/-0.15 including the uncertainties in the lens models. The Einstein rings break the degeneracies in the mass distributions or Hubble constants inferred from observations of gravitational lenses. In particular, the Einstein ring in PG1115+080 rules out the centrally concentrated mass distributions that lead to a high Hubble constant (H_0>60 km/s Mpc) given the measured time delays. Deep, detailed observations of Einstein rings will be revolutionary for constraining mass models and determining the Hubble constant from time delay measurements.

astro-ph↗

arXiv2000

The Fundamental Plane of Gravitational Lens Galaxies and The Evolution of Early-Type Galaxies in Low Density Environments

Most gravitational lenses are early-type galaxies in relatively low density environments -- a ``field'' rather than a ``cluster'' population. We show that field early-type galaxies with 0 < z < 1, as represented by the lens galaxies, lie on the same fundamental plane as those in rich clusters at similar redshifts. We then use the fundamental plane to measure the combined evolutionary and K-corrections for early-type galaxies in the V, I and H bands. Only for passively evolving stellar populations formed at z > 2 (H_0=65 km/s Mpc, Omega_0=0.3, Lambda_0=0.7) can the lens galaxies be matched to the local fundamental plane. The high formation epoch and the lack of significant differences between the field and cluster populations contradict many current models of the formation history of early-type galaxies. Lens galaxy colors and the fundamental plane provide good photometric redshift estimates with an empirical accuracy of -0.03 +/- 0.11 for the 17 lenses with known redshifts. A mass model dominated by dark matter is more consistent with the data than either an isotropic or radially anisotropic constant M/L mass model, and a radially anisotropic model is better than an isotropic model.

astro-ph↗

arXiv2000

Analytic Time Delays and H_0 Estimates for Gravitational Lenses

We study gravitational lens time delays for a general family of lensing potentials, which includes the popular singular isothermal elliptical potential and singular isothermal elliptical density distribution but allows general angular structure. Using a novel approach, we show that the time delay can be cast in a very simple form, depending only on the observed image positions. Including an external shear changes the time delay proportional to the shear strength, and varying the radial profile of the potential changes the time delay approximately linearly. These analytic results can be used to obtain simple estimates of the time delay and the Hubble constant in observed gravitational lenses. The naive estimates for four of five time delay lenses show surprising agreement with each other and with local measurements of H_0; the complicated Q 0957+561 system is the only outlier. The agreement suggests that it is reasonable to use simple isothermal lens models to infer H_0, although it is still important to check this conclusion by examining detailed models and by measuring more lensing time delays.

astro-ph↗

arXiv2000

Gravitational Lenses With More Than Four Images: I. Classification of Caustics

We study the problem of gravitational lensing by an isothermal elliptical density galaxy in the presence of a tidal perturbation. When the perturbation is fairly strong and oriented near the galaxy's minor axis, the lens can produce image configurations with six or even eight highly magnified images lying approximately on a circle. We classify the caustic structures in the model and identify the range of models that can produce such lenses. Sextuple and octuple lenses are likely to be rare because they require special lens configurations, but a full calculation of the likelihood will have to include both the existence of lenses with multiple lens galaxies and the strong magnification bias that affects sextuple and octuple lenses. At optical wavelengths these lenses would probably appear as partial or complete Einstein rings, but at radio wavelengths the individual images could probably be resolved.

astro-ph↗

arXiv2000

The Host Galaxy of the Lensed Quasar Q 0957+561

Infrared images of the Q 0957+561 gravitational lens obtained with the Hubble Space Telescope show two large (\sim 5 arcsec) lensed images of the z_s=1.41 quasar host galaxy. Parts of the host galaxy are doubly-imaged like the quasar, while other parts are quadruply-imaged. The distortions of the host galaxy offer the best probe yet of the global structure of the lensing potential, which is essential for determining the Hubble constant from the measured time delay. The distortions are inconsistent with the predictions of previously published lens models, which invalidates those models and their implications for H_0. New models show that the distortions finally break the long-standing degeneracy between the shape of the lens galaxy and the tidal shear contributed by the cluster containing the lens galaxy. The shape of the lens galaxy's mass distribution must be remarkably similar to the shape of its luminosity distribution, and most models that produce reasonable values for the Hubble constant roughly match the observed ellipticity gradient and isophote twist of the lens galaxy. Also, the cluster must be non-spherical and produce a relatively small tidal shear. Although there are still degeneracies in the lens models that lead to a 25% uncertainty in the derived value of the Hubble constant, there are also strong prospects for new observations to further improve the constraints and reduce the uncertainties.

astro-ph↗

arXiv1999

The Evolution of Gravitational Lens Galaxies

Most gravitational lens galaxies are early-type galaxies in relatively low density environments. We show that they lie on the same fundamental plane as early-type galaxies in both local and distant rich clusters. Their surface brightness evolution requires a typical star formation epoch of z=2-3, almost indistinguishable from that of rich cluster galaxies at comparable redshifts. The restricted galaxy type range of the lenses means that photometric redshifts work well even with only 1-3 filter photometry. We make preliminary measurements of the mass and luminosity functions of the lens galaxies, and find they are consistent with the standard model used for deriving cosmological limits using lens statistics. As expected for a mass-weighted sample, they are more massive and more luminous than the overall early-type galaxy population.

astro-ph↗

arXiv1999

A New Wide-Separation Gravitational Lens Candidate: RXJ 0921+4529

The lens candidate RXJ 0921+4529 consists of two z_s=1.66 quasar separated by 6."93 with an H band magnitude difference of Δm=1.39. The lens appears to be a z_l=0.31 X-ray cluster, including a m_H=18.5 late-type galaxy lying between the quasar images. We detect an extended source overlapping the faint quasar but not the bright quasar. If this extended source is the host galaxy of the fainter quasar, then the system is a quasar binary rather than a gravitational lens.

astro-ph↗

arXiv1999

The Quasar Pair Q 1634+267 A, B and the Binary QSO vs. Dark Lens Hypotheses

Deep HST/NICMOS H (F160W) band observations of the z=1.96 quasar pair Q 1634+267A,B reveal no signs of a lens galaxy to a 1 sigma threshold of approximately 22.5 mag. The minimum luminosity for a normal lens galaxy would be a 6L_* galaxy at z > 0.5, which is 650 times greater than our detection threshold. Our observation constrains the infrared mass-to-light ratio of any putative, early-type, lens galaxy to (M/L)_H > 690h_65 (1200h_65) for Omega_0=0.1 (1.0) and H_0=65h_65 km/s/Mpc. We would expect to detect a galaxy somewhere in the field because of the very strong Mg II absorption lines at z=1.1262 in the Q 1634+267 A spectrum, but the HST H-band, I-band (F785LP) and V-band (F555W) images require that any associated galaxy be very under-luminous less than 0.1 L^*_H (1.0 L^*_I) if it lies within less than 40 h^{-1} (100 h^{-1}) kpc from Q 1634+267 A,B. While the large image separation (3.86 arcsec) and the lack of a lens galaxy strongly favor interpreting Q 1634+267A,B as a binary quasar system, the spectral similarity remains a puzzle. We estimate that at most 0.06% of randomly selected quasar pairs would have spectra as similar to each other as the spectra of Q 1634+267 A and B. Moreover, spectral similarities observed for the 14 quasar pairs are significantly greater than would be expected for an equivalent sample of randomly selected field quasars. Depending on how strictly we define similarity, we estimate that only 0.01--3% of randomly drawn samples of 14 quasar pairs would have as many similar pairs as the observational sample.

astro-ph↗

arXiv1999

Dust and Extinction Curves in Galaxies with z>0: The Interstellar Medium of Gravitational Lens Galaxies

We determine 37 differential extinctions in 23 gravitational lens galaxies over the range 0 < z_l < 1. Only 7 of the 23 systems have spectral differences consistent with no differential extinction. The median differential extinction for the optically-selected (radio-selected) subsample is E(B-V)=0.04 (0.06) mag. The extinction is patchy and shows no correlation with impact parameter. The median total extinction of the bluest images is E(B-V)=0.08 mag, although the total extinction distribution is dominated by the uncertainties in the intrinsic colors of quasars. The directly measured extinction distributions are consistent with the mean extinction estimated by comparing the statistics of quasar and radio lens surveys, thereby confirming the need for extinction corrections when using the statistics of lensed quasars to estimate the cosmological model. A disjoint subsample of two face-on, radio-selected spiral lenses shows both high differential and total extinctions, but standard dust-to-gas ratios combined with the observed molecular gas column densities overpredict the amount of extinction by factors of 2-5. For several systems we can estimate the extinction law, ranging from R_V=1.5+/-0.2 for a z_l=0.96 elliptical, to R_V=7.2+/-0.1 for a z_l=0.68 spiral. For the four radio lenses where we can construct non-parametric extinction curves we find no evidence for gray dust over the IR-UV wavelength range. The dust can be used to estimate lens redshifts with reasonable accuracy, although we sometimes find two degenerate redshift solutions.

astro-ph↗