arXiv subjects
Publications and source records attributed to J. A. R. Caldwell.
We describe the data reduction and object cataloging for the GEMS survey, a large-area (800 arcmin^2) two-band (F606W and F850LP) imaging survey with the Advanced Camera for Surveys on HST, centered on the Chandra Deep Field South.
We combine high-resolution images from GEMS with redshifts and spectral energy distributions from COMBO-17 to explore the morphological types of galaxies that dominate the z~0.7 UV luminosity density. We analysed rest-frame 280 nm and V-band luminosities of 1483 galaxies with 0.65<z< 0.75, combining these with visual morphological classifications from F850LP images (approximately rest-frame V-band) taken with HST/ACS on the Extended Chandra Deep Field South. We derive UV luminosity functions and j_280 luminosity densities for spheroid-dominated galaxies, spiral galaxies, Magellanic irregulars, and clearly-interacting galaxies with morphologies suggestive of ongoing major mergers. We check the reliability of GEMS morphologies against the deeper GOODS images and quantify an incompleteness of the GEMS merger identification at the faint end. We derive the fractions of the global UV luminosity density j_280 originating from the galaxy types, and find that spiral galaxies and Magellanic irregulars dominate with about 40% each. Interacting and merging galaxies account for roughly 20% of j_280, while the contribution of early types is negligible. These results imply that the strong decline in the UV luminosity density of the Universe observed from z~1 until today is dominated by the decreasing UV luminosities of normal spiral galaxies, accompanied by the migration of UV-luminous star formation in irregular galaxies to systems of progressively lower mass and luminosity. (abridged)
We analyze the photometric data obtained by PLANET and OGLE on the caustic-crossing binary-lens microlensing event OGLE-2002-BLG-069. Thanks to the excellent photometric and spectroscopic coverage of the event, we are able to constrain the lens model up to the known ambiguity between close and wide binary lenses. The detection of annual parallax in combination with measurements of extended-source effects allows us to determine the mass, distance and velocity of the lens components for the competing models. While the model involving a close binary lens leads to a Bulge-Disc lens scenario with a lens mass of M=(0.51 +- 0.15) M_sol and distance of D_L=(2.9 +- 0.4) kpc, the wide binary lens solution requires a rather implausible binary black-hole lens (M >=126 M_sol). Furthermore we compare current state-of-the-art numerical and empirical models for the surface brightness profile of the source, a G5III Bulge giant. We find that a linear limb-darkening model for the atmosphere of the source star is consistent with the data whereas a PHOENIX atmosphere model assuming LTE and with no free parameter does not match our observations.
One third of present-day spirals host optically visible strong bars that drive their dynamical evolution. However, the fundamental question of how bars evolve over cosmological times has yet to be addressed, and even the frequency of bars at intermediate redshifts remains controversial. We investigate the frequency of bars out to z~1.0 drawing on a sample of 1590 galaxies from the GEMS survey, which provides morphologies from HST ACS two-color images, and highly accurate redshifts from the COMBO-17 survey. We identify spiral galaxies using the Sersic index, concentration parameter, and rest-frame color. We characterize bars and disks by fitting ellipses to F606W and F850LP images, taking advantage of the two bands to minimize bandpass shifting. We exclude highly inclined (i>60 deg) galaxies to ensure reliable morphological classifications, and apply completeness cuts of M_v <= -19.3 and -20.6. More than 40% of the bars that we detect have semi major axes a<0.5" and would be easily missed in earlier surveys without the small PSF of ACS. The bars that we can reliably detect are fairly strong (with ellipticities e>=0.4) and have a in the range ~1.2-13 kpc. We find that the optical fraction of such strong bars remains at ~(30% +- 6%) from the present-day out to look-back times of 2-6 Gyr (z~0.2-0.7) and 6-8 Gyr (z~0.7-1.0); it certainly shows no sign of a drastic decline at z>0.7. Our findings of a large and similar bar fraction at these three epochs favor scenarios in which cold gravitationally unstable disks are already in place by z~1, and where on average bars have a long lifetime (well above 2 Gyr). The distributions of structural bar properties in the two slices are, however, not statistically identical and therefore allow for the possibility that the bar strengths and sizes may evolve over time.
Bars drive the dynamical evolution of disk galaxies by redistributing mass and angular momentum, and they are ubiquitous in present-day spirals. Early studies of the Hubble Deep Field reported a dramatic decline in the rest-frame optical bar fraction f_opt to below 5% at redshifts z>0.7, implying that disks at these epochs are fundamentally different from present-day spirals. The GEMS bar project, based on ~8300 galaxies with HST-based morphologies and accurate redshifts over the range 0.2-1.1, aims at constraining the evolution and impact of bars over the last 8 Gyr. We present early results indicating that f_opt remains nearly constant at ~30% over the range z=0.2-1.1,corresponding to lookback times of ~2.5-8 Gyr. The bars detected at z>0.6 are primarily strong with ellipticities of 0.4-0.8. Remarkably, the bar fraction and range of bar sizes observed at z>0.6 appear to be comparable to the values measured in the local Universe for bars of corresponding strengths. Implications for bar evolution models are discussed.
We present the stellar main sequence luminosity function (LF) of the disrupted, low-mass, low-concentration globular cluster Palomar 5 and its well-defined tidal tails, which emanate from the cluster as a result of its tidal interaction with the Milky Way. The results of our deep (B ~ 24.5) wide-field photometry unequivocally indicate that preferentially fainter stars were removed from the cluster so that the LF of the cluster's main body exhibits a significant degree of flattening compared to other globular clusters. There is clear evidence of mass segregation, which is reflected in a radial variation of the LFs. The LF of the tidal tails is distinctly enhanced with faint, low-mass stars. Pal 5 exhibits a binary main sequence, and we estimate a photometric binary frequency of roughly 10%. Also the binaries show evidence of mass segregation with more massive binary systems being more strongly concentrated toward the cluster center.
We have performed HST imaging of a sample of 23 high-redshift (1.8<z<2.75) AGN, drawn from the COMBO-17 survey. The sample contains moderately luminous quasars (M_B ~ -23). The data are part of the GEMS imaging survey that provides high resolution optical images obtained with ACS in two bands (F606W and F850LP), sampling the rest-frame UV flux of the targets. To deblend the AGN images into nuclear and resolved (host galaxy) components we use a PSF subtraction technique that is strictly conservative with respect to the flux of the host galaxy. We resolve the host galaxies in both filter bands in 9 of the 23 AGN, whereas the remaining 14 objects are considered non-detections, with upper limits of less than 5% of the nuclear flux. However, when we coadd the unresolved AGN images into a single high signal-to-noise composite image we find again an unambiguously resolved host galaxy. The recovered host galaxies have apparent magnitudes of 23.0<F606W<26.0 and 22.5<F850LP<24.5 with rest-frame UV colours in the range -0.2<(F606W-F850LP)_obs<2.3. The rest-frame absolute magnitudes at 200 nm are -20.0<M_200nm<-22.2. The photometric properties of the composite host are consistent with the individual resolved host galaxies. We find that the UV colors of all host galaxies are substantially bluer than expected from an old population of stars with formation redshift z=5, independent of the assumed metallicities. These UV colours and luminosities range up to the values found for Lyman-break galaxies (LBGs) at z=3. Our results agree with the recent discovery of enhanced blue stellar light in AGN hosts at lower redshifts. [abridged]
We present the results from a study of the host galaxies of 15 optically selected AGNs with 0.5<z<1.1 from GEMS. GEMS is a Hubble Space Telescope imaging survey of a ~28"x28" contiguous field centered on the Chandra Deep Field South in the F606W and F850LP filter bands. It incorporates the SEDs and redshifts of ~10000 objects, obtained by the COMBO-17 project. We have detected the host galaxies of all 15 AGNs in the F850LP-band (and 13/15 in the F606W-band), recovering their fluxes, morphologies and structural parameters. We find that 80% of the host galaxies have early-type (bulge-dominated) morphologies, while the rest have structures characteristic of late-type (disk-dominated) galaxies. We find that 25% of the early types, and 30% of the late types, exhibit disturbances consistent with galaxy interactions. The hosts show a wide range of colors, from those of red-sequence galaxies to blue colors consistent with ongoing star formation. Roughly 70% of the morphologically early-type hosts have rest-frame blue colors, a much larger fraction than those typical of non-active morphologically early-type galaxies in this redshift and luminosity range. Yet, we find that the early-type hosts are structurally similar to red-sequence ellipticals, inasmuch as they follow an absolute magnitude versus half-light size correlation that are consistent with the mean relation for early-type galaxies at similar redshifts.
Microlensing is the only known direct method to measure the masses of stars that lack visible companions. In terms of microlensing observables, the mass is given by M=(c^2/4G)\tilde r_E θ_E and so requires the measurement of both the angular Einstein radius, θ_E, and the projected Einstein radius, \tilde r_E. Simultaneous measurement of these two parameters is extremely rare. Here we analyze OGLE-2003-BLG-238, a spectacularly bright (I_min=10.3), high-magnification (A_max = 170) microlensing event. Pronounced finite source effects permit a measurement of θ_E = 650 uas. Although the timescale of the event is only t_E = 38 days, one can still obtain weak constraints on the microlens parallax: 4.4 AU < \tilde r_E < 18 AU at the 1 σlevel. Together these two parameter measurements yield a range for the lens mass of 0.36 M_sun < M < 1.48 M_sun. As was the case for MACHO-LMC-5, the only other single star (apart from the Sun) whose mass has been determined from its gravitational effects, this estimate is rather crude. It does, however, demonstrate the viability of the technique. We also discuss future prospects for single-lens mass measurements.
We have mapped the AGN luminosity function and its evolution between z=1 and z=5 down to apparent magnitudes of $R<24$. Within the GEMS project we have analysed HST-ACS images of many AGN in the Extended Chandra Deep Field South, enabling us to assess the evolution of AGN host galaxy properties with cosmic time.
We discuss high-resolution, time-resolved spectra of the caustic exit of the binary microlensing event OGLE 2002-BUL-69 obtained with UVES on the VLT. The source star is a G5III giant in the Galactic Bulge. During such events, the source star is highly magnified, and a strong differential magnification around the caustic resolves its surface. Using an appropriate model stellar atmosphere generated by the NextGEN code we obtained a model light curve for the caustic exit and compared it with a dense set of photometric observations obtained by the PLANET microlensing follow up network. We further compared predicted variations in the H alpha equivalent width with those measured from our spectra. While the model and observations agree in the gross features, there are discrepancies suggesting shortcomings in the model, particularly for the H alpha line core, where we have detected amplified emission from the stellar chromosphere as the source star's trailing limb exited the caustic. This achievement became possible by the provision of the OGLE-III Early Warning System, a network of small telescopes capable of nearly-continuous round-the-clock photometric monitoring, on-line data reduction, daily near-real-time modelling in order to predict caustic crossing parameters, and a fast and efficient response of a 8m-class telescope to a ``Target-Of-Opportunity'' observation request.
We present a prescription to correct large-scale intensity variations affecting imaging data taken with the Wide Field Imager (WFI) at the MPG/ESO 2.2 m telescope at the European Southern Observatory at La Silla in Chile. Such smoothly varying, large-scale gradients are primarily caused by non-uniform illumination due to stray light, which cannot be removed using standard flatfield procedures. By comparing our observations to the well-calibrated, homogeneous multi-colour photometry from the Sloan Digital Sky Survey we characterise the intensity gradients across the camera by second-order polynomials. The application of these polynomials to our data removes the gradients and reduces the overall scatter. We also demonstrate that applying our correction to an independent WFI dataset significantly reduces its large-scale variations, indicating that our prescription provides a generally valid and simple tool for calibrating WFI photometry.
With round-the-clock monitoring of galactic bulge microlensing events, the PLANET experiment constrains the abundance and can yield the discovery of planets down to the mass of earth around galactic disk and bulge stars. Data taken until 1999 imply that less than 1/3 of bulge M-dwarfs are surrounded by jupiter-mass companions at orbital radii between 1 and 4 AU. The current rate of microlensing alerts allows 15-25 jupiters and 1-3 earths to be probed per year.
We present the main sequence luminosity function of the tidally disrupted globular cluster Palomar 5 and its tidal tails. For this work we analyzed imaging data obtained with the Wide Field Camera at the INT (La Palma) and data from the Wide Field Imager at the MPG/ESO 2.2 m telescope at La Silla down to a limiting magnitude of approximately 24.5 mag in B. Our results indicate that preferentially fainter stars were removed from the cluster so that the LF of the cluster's main body exhibits a significant degree of flattening compared to other GCs. This is attributed to its advanced dynamical evolution. The LF of the tails is, in turn, enhanced with faint, low-mass stars, which we interpret as a consequence of mass segregation in the cluster.
We obtain high-precision limb-darkening measurements in five bands (V, V_E, I_E, I, and H) for the K3 III (Teff=4200 K, [Fe/H]=+0.3, log(g)=2.3) source of the Galactic bulge microlensing event EROS BLG-2000-5. These measurements are inconsistent with the predictions of atmospheric models at >10 sigma. While the disagreement is present in all bands, it is most apparent in I, I_E and V_E, in part because the data are better and in part because the intrinsic disagreement is stronger. We find that when limb-darkening profiles are normalized to have unit total flux, the I-band models for a broad range of temperatures all cross each other at a common point. The solar profile also passes through this point. However, the profile as measured by microlensing does not. We conjecture that the models have incorporated some aspect of solar physics that is not shared by giant atmospheres.
Due to their extremely small luminosity compared to the stars they orbit, planets outside our own Solar System are extraordinarily difficult to detect directly in optical light. Careful photometric monitoring of distant stars, however, can reveal the presence of exoplanets via the microlensing or eclipsing effects they induce. The international PLANET collaboration is performing such monitoring using a cadre of semi-dedicated telescopes around the world. Their results constrain the number of gas giants orbiting 1--7 AU from the most typical stars in the Galaxy. Upgrades in the program are opening regions of ``exoplanet discovery space'' -- toward smaller masses and larger orbital radii -- that are inaccessible to the Doppler velocity technique.
We analyze PLANET photometric observations of the caustic-crossing binary-lens microlensing event, EROS BLG-2000-5, and find that modeling the observed light curve requires incorporation of the microlens parallax and the binary orbital motion. The projected Einstein radius (~r_E = 3.61 +/- 0.11 AU) is derived from the measurement of the microlens parallax, and we are also able to infer the angular Einstein radius (theta_E = 1.38 +/- 0.12 mas) from the finite source effect on the light curve, combined with an estimate of the angular size of the source given by the source position in a color-magnitude diagram. The lens mass, M = 0.612 +/- 0.057 M_sun, is found by combining these two quantities. This is the first time that parallax effects are detected for a caustic-crossing event and also the first time that the lens mass degeneracy has been completely broken through photometric monitoring alone. The combination of the projected Einstein radius and the angular Einstein radius also allows us to conclude that the lens lies in the near side of the disk, within 2.6 kpc of the Sun, while the radial velocity measurement indicates that the source is a Galactic bulge giant.
We analyze PLANET and MACHO observations of MACHO 99-BLG-47, the first nearly-normal microlensing event for which high signal-to-noise-ratio data reveal a well-covered, short-duration anomaly. This anomaly occurs near the the peak of the event. Short-duration anomalies near the peak of otherwise normal events are expected to arise both from extreme-separation (either very close or very wide), roughly equal-mass binary lenses, and from planetary systems. We show that the lens of MACHO 99-BLG-47 is in fact an extreme-separation binary, not a planetary system, thus demonstrating for the first time that these two important classes of events can be distinguished in practice. However, we find that the wide-binary and close-binary lens solutions fit the data equally well, and cannot be distinguished even at Delta-chi^2=1. This degeneracy is qualitatively much more severe than the one identified for MACHO 98-SMC-1 because the present degeneracy spans two rather than one dimension in the magnification field and does not require significantly different blending fractions. In the appendix, we explore this result, and show that it is related to the symmetry in the lens equation.