Tomographic halo model of the unWISE-Blue galaxies using cross-correlations with BOSS CMASS galaxies

Abstract: The halo model offers a framework for investigating galaxy clustering, and for understanding the growth of galaxies and the distribution of galaxies of different types. Here, we use the halo model to study the small-scale clustering and halo occupation distribution (HOD) of the unWISE-Blue galaxy sample, an infrared-selected sample of $\sim$100 million galaxies across the entire extragalactic sky at $z\sim 0.5$ $-$ similar redshifts to the Baryon Oscillation Spectroscopic Survey (BOSS) CMASS sample. Although the photometric unWISE galaxies cannot be easily split in redshift, we use their cross-correlation with the BOSS CMASS sample to tomographically probe the HOD of the unWISE galaxies at $0.45 < z < 0.75$. To do so, we develop a new method for applying the halo model to cross-correlations between a photometric sample and a spectroscopic sample in narrow redshift bins, incorporating halo exclusion, post-Limber corrections, and redshift-space distortions. We reveal strong evolution in the CMASS HOD, and modest evolution in the unWISE-Blue HOD. For unWISE-Blue, we find that the average bias and mean halo mass drop from $b = 1.6$ and $\log_{10}(M_{\mathrm{h}}/M_{\odot}) \sim 13.4$ at $z \sim 0.5$ to $b = 1.4$ and $\log_{10}(M_{\mathrm{h}}/M_{\odot}) \sim 13.1$ at $z \sim 0.7$, and that the satellite fraction drops modestly from $\sim$20% to $\sim$10% in the same range. These results are useful for creating mock samples of the unWISE-Blue galaxies. Furthermore, the techniques developed to obtain these results are applicable to other tomographic cross-correlations between photometric samples and narrowly-binned spectroscopic samples, such as clustering redshifts.