Predictions for Cosmological Infrared Surveys from Space with the Multiband Imaging Photometer for SIRTF (MIPS)

Abstract

We make predictions for the cosmological surveys to be conducted by MIPS/SIRTF at 24, 70 and 160 microns, for the GTO and the legacy programs, using the latest knowledge of the instrument. In addition to detector and cirrus confusion noise, we discuss in detail the derivation of the confusion noise due to extragalactic sources, that depends strongly on the shape of the source counts at a given wavelength and on the telescope and detector pixel sizes. We show that it is wise in general to compare the classical photometric criterion and the so called source density criterion to predict the confusion levels. We obtain, using the model of Lagache, Dole, & Puget (2002) limiting fluxes of 50 microJy, 3.2 mJy and 36 mJy at 24, 70 and 160 microns (resp.). We compute the redshift distributions of the detected sources at each wavelength, and show that they extend up to z ~ 2.7 at 24 microns and up to z ~ 2.5 at 70 and 160 microns, leading to resolve at most 69, 54 and 24% of the Cosmic Infrared Background (CIB) at 24, 70 and 160 microns (resp.). We estimate which galaxy populations will be used to derive the luminosity function evolution with redshift. We also give the redshift distributions of the unresolved sources in the FIR range, that dominates the fluctuations of the CIB, and a predicted power spectrum showing the feasibility of fluctuations (due to Poissonian & clustered source distributions) measurements. The main conclusion is that MIPS (and SIRTF in general) cosmological surveys will greatly improve our understanding of galaxy evolution by giving data with unprecedented accuracy in the mid and far infrared range.