Multiwavelength Study of Massive Galaxies at<i>z</i>∼2. I. Star Formation and Galaxy Growth

Abstract

Examining a sample of massive galaxies at 1.4&lt;z&lt;2.5 with K_{Vega}&lt;22 from the\nGreat Observatories Origins Deep Survey, we compare photometry from Spitzer at\nmid- and far-IR, to submillimeter, radio and rest-frame ultraviolet\nwavelengths, to test the agreement between different tracers of star formation\nrates (SFRs) and to explore the implications for galaxy assembly. For z~2\ngalaxies with moderate luminosities(L_{8um}&lt;10^{11}L_sun), we find that the SFR\ncan be estimated consistently from the multiwavelength data based on local\nluminosity correlations. However,20--30% of massive galaxies, and nearly all\nthose with L_{8um}&gt;10^{11}L_sun, show a mid-IR excess which is likely due to\nthe presence of obscured active nuclei, as shown in a companion paper. There is\na tight and roughly linear correlation between stellar mass and SFR for\n24um-detected galaxies. For a given mass, the SFR at z=2 was larger by a factor\nof ~4 and ~30 relative to that in star forming galaxies at z=1 and z=0,\nrespectively. Typical ultraluminous infrared galaxies (ULIRGs) at z=2 are\nrelatively 'transparent' to ultraviolet light, and their activity is long lived\n(~400 Myr), unlike that in local ULIRGs and high redshift\nsubmillimeter-selected galaxies. ULIRGs are the common mode of star formation\nin massive galaxies at z=2, and the high duty cycle suggests that major mergers\nare not the dominant trigger for this activity.Current galaxy formation models\nunderpredict the normalization of the mass-SFR correlation by about a factor of\n4, and the space density of ULIRGs by an orderof magnitude, but give better\nagreement for z&gt;1.4 quiescent galaxies.\n