HST F160W Imaging of Very Massive Galaxies at $1.5<z<3.0$: Diversity of Structures and the Effect of Close Pairs on Number Density Estimates

TL;DR

This study uses HST imaging to analyze the structures of very massive galaxies at redshifts 1.5 to 3, revealing diversity in sizes and the impact of close pairs on galaxy number density estimates, with implications for future surveys.

Contribution

It provides the first detailed analysis of very massive galaxy structures at high redshift and quantifies the effect of close pairs on stellar mass function measurements.

Findings

Approximately one-third of the galaxies are close pairs.

Quiescent galaxies at 2.5<z<3.0 are larger than expected.

Close pairs can reduce observed number densities by up to 1.5 times.

Abstract

We present a targeted follow-up Hubble Space Telescope WFC3 F160W imaging study of very massive galaxies $(\log(M_{\rm{star}}/M_{\odot})> 11.2)$ selected from a combination of ground-based near-infrared galaxy surveys (UltraVISTA, NMBS-II, UKIDSS UDS) at $1.5<z<3$. We find that these galaxies are diverse in their structures, with $\sim1/3$ of the targets being composed of close pairs, and span a wide range in sizes. At $1.5<z<2.5$, the sizes of both star-forming and quiescent galaxies are consistent with the extrapolation of the stellar mass-size relations determined at lower stellar masses. At $2.5<z<3.0$, however, we find evidence that quiescent galaxies are systematically larger than expected based on the extrapolation of the relation derived using lower stellar mass galaxies. We used the observed light profiles of the blended systems to decompose their stellar masses and investigate the effect of the close pairs on the measured number densities of very massive galaxies in the early universe. We estimate correction factors to account for close-pair blends and apply them to the observed stellar mass functions measured using ground-based surveys. Given the large uncertainties associated with this extreme population of galaxies, there is currently little tension between the (blending-corrected) number density estimates and predictions from theoretical models. Although we currently lack the statistics to robustly correct for close-pair blends, we show that this is a systematic effect which can reduce the observed number density of very massive galaxies by up to a factor of $\sim1.5$, and should be accounted for in future studies of stellar mass functions.