Ultradeep IRAC Imaging Over the HUDF and GOODS-South: Survey Design and Imaging Data Release

Labbé, I.; Oesch, P. A.; Illingworth, G. D.; van Dokkum, P. G.; Bouwens, R. J.; Franx, M.; Carollo, C. M.; Trenti, M.; Holden, B.; Smit, R.; González, V.; Magee, D.; Stiavelli, M.; Stefanon, M.

Abstract

The IRAC ultradeep field and IRAC Legacy over GOODS programs are two ultradeep imaging surveys at 3.6 and 4.5 μm with the Spitzer Infrared Array Camera (IRAC). The primary aim is to directly detect the infrared light of reionization epoch galaxies at z > 7 and to constrain their stellar populations. The observations cover the Hubble Ultra Deep Field (HUDF), including the two HUDF parallel fields, and the CANDELS/GOODS-South, and are combined with archival data from all previous deep programs into one ultradeep data set. The resulting imaging reaches unprecedented coverage in IRAC 3.6 and 4.5 μm ranging from >50 hr over 150 arcmin2, >100 hr over 60 sq arcmin2, to ˜200 hr over 5-10 arcmin2. This paper presents the survey description, data reduction, and public release of reduced mosaics on the same astrometric system as the CANDELS/GOODS-South Wide Field Camera 3 (WFC3) data. To facilitate prior-based WFC3+IRAC photometry, we introduce a new method to create high signal-to-noise PSFs from the IRAC data and reconstruct the complex spatial variation due to survey geometry. The PSF maps are included in the release, as are registered maps of subsets of the data to enable reliability and variability studies. Simulations show that the noise in the ultradeep IRAC images decreases approximately as the square root of integration time over the range 20-200 hr, well below the classical confusion limit, reaching 1σ point-source sensitivities as faint as 15 nJy (28.5 AB) at 3.6 μm and 18 nJy (28.3 AB) at 4.5 μm. The value of such ultradeep IRAC data is illustrated by direct detections of z = 7-8 galaxies as faint as HAB = 28.

ADS Listing