Samuel J. Schmidt

PHAT: PHoto-z Accuracy Testing

Context. Photometric redshifts (photo-zs) have become an essential tool in extragalactic astronomy. Many current and upcoming observing programmes require great accuracy of photo-zs to reach their scientific goals. Aims. Here we introduce PHAT, the PHoto-z Accuracy Testing programme, an international initiative to test and compare different methods of photo-z estimation. Methods. Two different test environments are set up, one (PHAT0) based on simulations to test the basic functionality of the different photo-z codes, and another one (PHAT1) based on data from the GOODS survey including 18-band photometry and similar to 2000 spectroscopic redshifts. Results. The accuracy of the different methods is expressed and ranked by the global photo-z bias, scatter, and outlier rates. While most methods agree very well on PHAT0 there are differences in the handling of the Lyman-alpha forest for higher redshifts. Furthermore, different methods produce photo-z scatters that can differ by up to a factor of two even in this idealised case. A larger spread in accuracy is found for PHAT1. Few methods benefit from the addition of mid-IR photometry. The accuracy of the other methods is unaffected or suffers when IRAC data are included. Remaining biases and systematic effects can be explained by shortcomings in the different template sets (especially in the mid-IR) and the use of priors on the one hand and an insufficient training set on the other hand. Some strategies to overcome these problems are identified by comparing the methods in detail. Scatters of 4-8% in Delta z/(1 + z) were obtained, consistent with other studies. However, somewhat larger outlier rates (\textgreater 7.5% with Delta z/(1 + z) \textgreater 0.15; \textgreater 4.5% after cleaning) are found for all codes that can only partly be explained by AGN or issues in the photometry or the spec-z catalogue. Some outliers were probably missed in comparisons of photo-zs to other, less complete spectroscopic surveys in the past. There is a general trend that empirical codes produce smaller biases than template-based codes. Conclusions. The systematic, quantitative comparison of different photo-z codes presented here is a snapshot of the current state-of-the-art of photo-z estimation and sets a standard for the assessment of photo-z accuracy in the future. The rather large outlier rates reported here for PHAT1 on real data should be investigated further since they are most probably also present (and possibly hidden) in many other studies. The test data sets are publicly available and can be used to compare new, upcoming methods to established ones and help in guiding future photo-z method development.

DISCOVERY OF A DISSOCIATIVE GALAXY CLUSTER MERGER WITH LARGE PHYSICAL SEPARATION

We present DLSCL J0916.2+2951 (z = 0.53), a newly discovered major cluster merger in which the collisional cluster gas has become dissociated from the collisionless galaxies and dark matter (DM). We identified the cluster using optical and weak-lensing observations as part of the Deep Lens Survey. Our follow-up observations with Keck, Subaru, Hubble Space Telescope, and Chandra show that the cluster is a dissociative merger and constrain the DM self-interaction cross-section σDM m –1 DM 7 cm2 g–1. The system is observed at least 0.7 ± 0.2 Gyr since first pass-through, thus providing a picture of cluster mergers 2-5 times further progressed than similar systems observed to date. This improved temporal leverage has implications for our understanding of merging clusters and their impact on galaxy evolution.

Cosmic Shear Results from the Deep Lens Survey - II: Full Cosmological Parameter Constraints from Tomography

We present a tomographic cosmic shear study from the Deep Lens Survey (DLS), which, providing a limiting magnitude r_{lim}~27 (5 sigma), is designed as a pre-cursor Large Synoptic Survey Telescope (LSST) survey with an emphasis on depth. Using five tomographic redshift bins, we study their auto- and cross-correlations to constrain cosmological parameters. We use a luminosity-dependent nonlinear model to account for the astrophysical systematics originating from intrinsic alignments of galaxy shapes. We find that the cosmological leverage of the DLS is among the highest among existing >10 sq. deg cosmic shear surveys. Combining the DLS tomography with the 9-year results of the Wilkinson Microwave Anisotropy Probe (WMAP9) gives Omega_m=0.293_{-0.014}^{+0.012}, sigma_8=0.833_{-0.018}^{+0.011}, H_0=68.6_{-1.2}^{+1.4} km/s/Mpc, and Omega_b=0.0475+-0.0012 for LCDM, reducing the uncertainties of the WMAP9-only constraints by ~50%. When we do not assume flatness for LCDM, we obtain the curvature constraint Omega_k=-0.010_{-0.015}^{+0.013} from the DLS+WMAP9 combination, which however is not well constrained when WMAP9 is used alone. The dark energy equation of state parameter w is tightly constrained when Baryonic Acoustic Oscillation (BAO) data are added, yielding w=-1.02_{-0.09}^{+0.10} with the DLS+WMAP9+BAO joint probe. The addition of supernova constraints further tightens the parameter to w=-1.03+-0.03. Our joint constraints are fully consistent with the final Planck results and also the predictions of a LCDM universe.

Mapping the Galaxy Color-Redshift Relation: Optimal Photometric Redshift Calibration Strategies for Cosmology Surveys

Calibrating the photometric redshifts of ≳ 10^9 galaxies for upcoming weak lensing cosmology experiments is a major challenge for the astrophysics community. The path to obtaining the required spectroscopic redshifts for training and calibration is daunting, given the anticipated depths of the surveys and the difficulty in obtaining secure redshifts for some faint galaxy populations. Here we present an analysis of the problem based on the self-organizing map, a method of mapping the distribution of data in a high-dimensional space and projecting it onto a lower-dimensional representation. We apply this method to existing photometric data from the COSMOS survey selected to approximate the anticipated Euclid weak lensing sample, enabling us to robustly map the empirical distribution of galaxies in the multidimensional color space defined by the expected Euclid filters. Mapping this multicolor distribution lets us determine where—in galaxy color space—redshifts from current spectroscopic surveys exist and where they are systematically missing. Crucially, the method lets us determine whether a spectroscopic training sample is representative of the full photometric space occupied by the galaxies in a survey. We explore optimal sampling techniques and estimate the additional spectroscopy needed to map out the color–redshift relation, finding that sampling the galaxy distribution in color space in a systematic way can efficiently meet the calibration requirements. While the analysis presented here focuses on the Euclid survey, similar analysis can be applied to other surveys facing the same calibration challenge, such as DES, LSST, and WFIRST.

Recovering redshift distributions with cross-correlations: pushing the boundaries

Determining accurate redshift distributions for very large samples of objects has become increasingly important in cosmology. We investigate the impact of extending cross-correlation based redshift distribution recovery methods to include small scale clustering information. The major concern in such work is the ability to disentangle the amplitude of the underlying redshift distribution from the influence of evolving galaxy bias. Using multiple simulations covering a variety of galaxy bias evolution scenarios, we demonstrate reliable redshift recoveries using linear clustering assumptions well into the non-linear regime for redshift distributions of narrow redshift width. Including information from intermediate physical scales balances the increased information available from clustering and the residual bias incurred from relaxing of linear constraints. We discuss how breaking a broad sample into tomographic bins can improve estimates of the redshift distribution, and present a simple bias removal technique using clustering information from the spectroscopic sample alone.

Tomographic magnification of Lyman-break galaxies in the Deep Lens Survey

Using about 450 000 galaxies in the Deep Lens Survey, we present a detection of the gravitational magnification of z > 4 Lyman-break galaxies by massive foreground galaxies with 0.4 7.5σ with a fixed σ8 and is found to be consistent with the expected redshift dependence of the Wilkinson Microwave Anisotropy Probe 7 Λ cold dark matter cosmology.

THE STAR FORMATION HISTORY OF GALAXIES MEASURED FROM INDIVIDUAL PIXELS. I. THE HUBBLE DEEP FIELD NORTH

We analyze the photometric information contained in individual pixels of galaxies in the Hubble Deep Field North (HDF-N) using a new technique, pixel-z, that combines predictions of evolutionary synthesis models with photometric redshift template fitting. Each spectral energy distribution template is a result of modeling of the detailed physical processes affecting gas properties and star formation efficiency. The criteria chosen to generate the spectral energy distribution templates is that of sampling a wide range of physical characteristics such as age, star formation rate, obscuration, and metallicity. A key feature of our method is the sophisticated use of error analysis to generate error maps that define the reliability of the template fitting on pixel scales and allow for the separation of the interplay among dust, metallicity, and star formation histories. This technique offers a number of advantages over traditional integrated color studies. As a first application, we derive the star formation and metallicity histories of galaxies in the HDF-N. Our results show that the comoving density of star formation rate, determined from the UV luminosity density of sources in the HDF-N, increases monotonically with redshift out to at least redshift of 5. This behavior can plausibly be explained by a smooth increase of the UV luminosity density with redshift coupled with an increase in the number of star-forming regions as a function of redshift. We also find that the information contained in individual pixels in a galaxy can be linked to its morphological history. Finally, we derive the metal enrichment rate history of the universe and find it in good agreement with predictions based on the evolving H I content of Ly? QSO absorption-line systems.

Inferring the redshift distribution of the cosmic infrared background

Cross-correlating the Planck High Frequency Instrument (HFI) maps against quasars from the Sloan Digital Sky Survey (SDSS) DR7, we estimate the intensity distribution of the Cosmic Infrared Background (CIB) over the redshift range 0 < z < 5.We detect redshift-dependent spatial cross-correlations between the two datasets using the 857, 545 and 353 GHz channels and we obtain upper limits at 217 GHz consistent with expectations. At all frequencies with detectable signal we infer a redshift distribution peaking around z ~ 1.2 and find the recovered spectrum to be consistent with emission arising from star forming galaxies. By assuming simple modified blackbody and Kennicutt relations, we estimate dust and star formation rate density as a function of redshift, finding results consistent with earlier multiwavelength measurements over a large portion of cosmic history. However, we note that, lacking mid-infrared coverage, we are not able to make an accurate determination of the mean temperature for the dust responsible for the CIB. Our results demonstrate that clustering-based redshift inference is a valuable tool for measuring the entire evolution history of the cosmic star formation rate from a single and homogeneous dataset.

Improved photometric redshifts via enhanced estimates of system response, galaxy templates and magnitude priors

Wide, deep photometric surveys require robust photometric redshift estimates (photo-zs) for studies of large-scale structure. These estimates depend critically on accurate photometry. We describe the improvements to the photometric calibration and the photo-z estimates in the Deep Lens Survey (DLS) from correcting three of the inputs to the photo-z calculation: the system response as a function of wavelength, the spectral energy distribution templates, and template prior probabilities as a function of magnitude. We model the system response with a physical model of the MOSAIC cameras CCD, which corrects a 0.1 magnitude discrepancy in the colours of type M2 and later stars relative to the SDSS z-band photometry. We provide our estimated z-band response function for the use of other surveys that used MOSAIC before its recent detector upgrade. The improved throughput curve, template set, and Bayesian prior lead to a 20 per cent reduction in photo-z scatter and a reduction of the bias by a factor of more than two. This paper serves as both a photo-z data release description for DLS and a guide for testing the quality of photometry and resulting photo-zs generally.

Galaxy-Mass Correlations on 10 Mpc Scales in the Deep Lens Survey

We examine the projected correlation of galaxies with mass from small scales (<few hundred kpc), where individual dark matter halos dominate, out to 15 Mpc, where correlated large-scale structure dominates. We investigate these profiles as a function of galaxy luminosity and redshift. Selecting 0.8 million galaxies in the Deep Lens Survey, we use photometric redshifts and stacked weak gravitational lensing shear tomography out to radial scales of 1° from the centers of foreground galaxies. We detect correlated mass density from multiple halos and large-scale structure at radii larger than the virial radius and find the first observational evidence for growth in the galaxy-mass correlation on 10 Mpc scales with decreasing redshift and fixed range of luminosity. For a fixed range of redshift, we find a scaling of projected halo mass with rest-frame luminosity similar to previous studies at lower redshift. We control systematic errors in shape measurement and photometric redshift, enforce volume completeness through absolute magnitude cuts, and explore residual sample selection effects via simulations.