Observational determination of the galaxy bias from cosmic variance with a random pointing survey: Clustering of z~2 galaxies from Hubble's BoRG survey

Abstract

Gravitational clustering broadens the count-in-cells distribution of galaxies for surveys along uncorrelated (well-separated) lines of sight beyond Poisson noise. A number of methods have proposed to measure this excess "cosmic" variance to constrain the galaxy bias (i.e. the strength of clustering) independently of the two-point correlation function. Here we present an observational application of these methods using data from 141 uncorrelated fields (~700 arcmin$^2$ total) from Hubble's Brightest of Reionizing Galaxies (BoRG) survey. We use BoRG's broad-band imaging in optical and near infrared to identify N~1000 photometric candidates at z~2 through a combination of colour selection and photometric redshift determination, building a magnitude-limited sample with $m_{AB}\leq24.5$ in F160W. We detect a clear excess in the variance of the galaxy number counts distribution compared to Poisson expectations, from which we estimate a galaxy bias $b \approx 3.63 \pm 0.57$. When divided by SED-fit classification into ~400 early-type and ~600 late-type candidates, we estimate biases of $b_{early} \approx 4.06 \pm 0.67$ and $b_{late} \approx 2.98 \pm 0.98$ respectively. These estimates are consistent with previous measurements of the bias from the two-point correlation function, and demonstrate that with $N\gtrsim100$ sight-lines, each containing $N\gtrsim5$ objects, the counts-in-cell analysis provides a robust measurement of the bias. This implies that the method can be applied effectively to determine clustering properties (and characteristic dark-matter halo masses) of z~6-9 galaxies from a pure-parallel James Webb Space Telescope survey similar in design to Hubble's BoRG survey.