Updated 7/12/2005 by J. Lotz Created 5/16/2005 by J. Lotz lotz@noao.edu ****************************************************************** 1) Extended Groth Strip HST ACS image general information 2) Image naming conventions and sizes 3) Multi-Drizzle Parameters 4) Photometry ** updated 7/12/05 ** 5) Astrometry ** updated 7/12/05 ** 6) Limiting magnitudes ** updated 11/04/05 ** ****************************************************************** 1) General Information-- This website contains the latest version (5/05) of the reduced ACS images and photometry catalogs for the Extended Groth Strip, taken for HST Cycle 13 program 10134, PI. Marc Davis. Go here for the MAST program information: http://archive.stsci.edu/cgi-bin/proposal_search?id=10134&mission=hst The EGS ACS data consists of 63 ACS pointings. Each pointing has been observed in two bandpasses, F606W (wide V) and F814W (I). Each reduced image has been drizzled from 4 exposures to produce final 8000 x 8000 pixel images with pixel sampling of 0.03" per pixel. The total exposure times, ABMAG and VEGA zeropoints are: F606W (V) 2260s 26.486 26.396 F814W (V) 2100s 25.937 25.501 2) Image naming conventions and sizes: The EGS tiles are in a 3 x 21 pattern, with the 01-01 tile at the south-east corner and the 21-03 tile in the north-west corner. The reduced and 0.03" per pixel drizzled F606W and F814W images are egs0101_v_drz03.fits -- 245M egs0101_i_drz03.fits -- 245M The inverse variance weight maps produced by Multi-drizzle are egs0101_v_wht03.fits -- 245M egs0101_i_wht03.fits -- 245M The segmentation map produced by SExtractor for the summed V+I images is egs0101_seg.fits -- 123 M The corresponding photometry catalog is egs0101_phot.cat -- 301 K Therefore, the ACS science images for each tiles require 1103 M, or ~68 GB for the entire EGS. 3) Multi-drizzling parameters: We downloaded the ACS pipeline-reduced images for each individual exposure from the MAST archive. This pipeline includes the basic bias subtractions and flat-fielding and provides a data-quality file for bad pixels. In order to achieve the best resolution possible, the images for each pointing/bandpass were "drizzled" together using the STSDAS Multidrizzle algorithim. These latest reductions were done at a higher pixel sampling (0.03" per pixel) than the previous reduction. We used the square kernel, a pixel dropsize p = 0.8, and final pixel size = 0.6 (0.03/0.05). The trade-off for higher resolution is increased correlation in the background noise in the final drizzled image. The Multi-drizzle algorithim also produces inverse variance (1/RMS^2) weight maps for each drizzled image. These weight maps include 1) effects of non-uniform exposure times over the "gaps" in ACS and masked cosmic-rays and 2) the "true" noise expected for the output images. Therefore, any photometry measurements should use the output weight maps to recover the true photometric errors. The Multidrizzle geometric distortion correction coefficients, relative shifts, and singly-drizzled images are available for those wishing to reconstruct the ACS PSF across the images. The pre-drizzle STScI pipeline-reduced individual images are also available or can be downloaded directly from MAST. 4) Photometry: The photometric catalogs and segmentation maps were produced using SExtractor ver 2.3 (Bertin & Arnout 1996). The objects were first detected in the summed V+I image convolved with a 9x9 pixel tophat function. The segmentation map, positions, ellipticities, and position angles are based on the summed V+I image. Then the photometry in each bandpass was measured using the V+I segmentation maps. Therefore the "isophotal" V and I magnitudes are measured by summing over the same pixels for each object, and the isophotal V-I colors should be a reasonable estimate of the total galaxy color. We adjusted the weight maps used by SExtractor such that regions of the final drizzled image with less than 1 exposure were not measured. This excludes most of the cosmic-rays found around the edges of each images from the photometric catalogs. The photometry catalogs has the following columns: # (1) Tile # (2) Galaxy -- SExtractor ID number; matches segmentation map # (3) Xcenter # (4) Ycenter # (5) RA_V -- RA from V image # (6) DEC_V -- DEC from V image # (7) STAR_V -- "stellarity index" > 0.8-0.9 likely to be stars # (8) V_MISO -- isophotal V mag (ABmag) # (9) V_MISOER #(10) V_MBEST -- "best" V mag (ABmag) -- rough correction for ap.effects #(11) V_MBESTER #(12) RA_I -- RA from I image #(13) DEC_I -- DEC from I image #(14) STAR_I -- "stellarity index" #(15) I_MISO -- isophotal I mag (ABmag) #(16) I_MISOER #(17) I_MBEST -- best I mag (ABmag) #(18) I_MBESTER #(19) VI_ISO -- isophotal V-I color #(20) VI_ISOER #(21) VI_BEST -- "best" V-I color #(22) VI_BESTER #(23) KRON_RAD -- Kron radius #(24) FWHM -- full width half max #(25) A/B #(26) PA 5) Astrometry: 7/12/05 The ACS images have been tied to the DEEP2/SDSS astrometry. These revised plate solutions are good to ~0.07" in RA and DEC. Images downloaded prior to 7/12/05 were not tied to DEEP astrometry and may have absolute offsets of a few arcsecs. 6) Limiting magnitude: for F606W, in a 5x5 pixel (0.15" x 0.15") box 5-sigma limiting mag = 28.75 (AB) 10-sigma = 27.94 (AB) for F814W, in a 5x5 pixel (0.15" x 0.15") box 5-sigma limiting mag = 28.10 (AB) 10-sigma = 27.30 (AB)