Weak Gravitational Lensing with COSMOS: Galaxy Selection and Shape Measurements

TL;DR

The paper addresses precision weak lensing from space by constructing a PSF- and CTE-corrected galaxy shape catalog from the COSMOS ACS data. It develops a full pipeline including data reduction, completeness assessment, photometric redshift distribution, PSF modelling with effective focus tracking, RRG-based shape measurements, and shear calibration via extensive simulations, culminating in a final lensing catalog of about 3.9×10^5 galaxies over 1.64 deg^2 with an effective density near 66 arcmin^-2. The work quantifies intrinsic shape noise, showing a near-constant floor of $\sigma_{int} \approx 0.26$ across size, magnitude, and redshift, and demonstrates that the COSMOS catalog is well suited for 3D tomography and future wide-field space missions. These results provide critical inputs for survey design, photometric redshift strategies, and calibration requirements in upcoming weak lensing programs.

Abstract

With a primary goal of conducting precision weak lensing measurements from space, the COSMOS survey has imaged the largest contiguous area observed by the Hubble Space Telescope (HST) to date using the Advanced Camera for Surveys (ACS). This is the first paper in a series where we describe our strategy for addressing the various technical challenges in the production of weak lensing measurements from the COSMOS data. The COSMOS ACS catalog is constructed from 575 ACS/WFC tiles (1.64 deg^2) and contains a total 1.2x10^6 objects to a limiting magnitude of F814W=26.5. This catalog is made publicly available. The shapes of galaxies have been measured and corrected for the distortion induced by the time varying ACS Point Spread Function and for Charge Transfer Efficiency effects. Next, simulated images are used to derive the shear susceptibility factors that are necessary in order to transform shape measurements into unbiased shear estimators. Finally, for each galaxy, we derive a shape measurement error and utilize this quantity to extract the intrinsic shape noise of the galaxy sample. Interestingly, our results indicate that the intrinsic shape noise varies little with either size, magnitude or redshift. Representing a number density of 66 galaxies per arcmin^2, the final COSMOS weak lensing catalog contains 3.9x10^5 galaxies with accurate shape measurements. The properties of the COSMOS weak lensing catalog described throughout this paper will provide key input numbers for the preparation and design of next-generation wide field space missions.

Figures (12)

• Figure 1: Date of observation for each of the survey pointings. The PSF of the ACS/WFC varies on timescales that are much shorter than the period over which COSMOS was observed.
• Figure 9: The top panel shows the galaxy and stellar number counts as compared to the HDF. The dashed curve corresponds to images with a sun-angle of less than 70 degrees and the solid curve corresponds to images with a sun-angle greater than 70 degrees. Poisson error bars are also indicated but are very small. The bottom panel shows the point source selection for the catalog compared to stellar models computed from evolutionary tracks and constrained by local Hipparcos data.
• Figure 10: Completeness of the COSMOS F814W catalog as a function of total magnitude and FWHM determined by inserting fake objects into an ACS image. The thick contours show the percentage of fake objects recovered by SExtractor. The thin contours are the lines of constant surface brightness, in units of mag arcsec$^{-2}$, assuming a Gaussian profile. The grey points represent a random sample of objects from the COSMOS catalog plotted as a function of mag_auto and fwhm_image. The dashed horizontal line indicates the size of the ACS PSF. Note that the simulations only consider objects with Gaussian profiles whereas in reality, the COSMOS objects exhibit a wide variety of profiles.
• Figure 11: Median redshift of COSMOS compared to various photometric redshift surveys. For clarity, error bars are only shown for the COSMOS, CFHTLS and UDF surveys
• Figure 12: Adopted PSF model across the survey. The colors correspond to deviations in the apparent focus of the telescope away from nominal ($\mu m$). These are caused by expansion and contraction of the HST due to thermal variations. Note that the focus values are clustered and not randomly distributed. See Rhodes et al. 2007 for more details about the PSF pattern at particular focus positions.