Building a better understanding of the massive high-redshift BOSS CMASS galaxies as tools for cosmology

We explore the massive bluer star-forming population of the Sloan Digital Sky Survey (SDSS) III/BOSS CMASS DR11 galaxies at z > 0.55 to quantify their differences, in terms of redshift-space distortions and large-scale bias, with respect to the luminous red galaxy sample. We perform a qualitative analysis to understand the significance of these differences and whether we can model and reproduce them in mock catalogues. Specifically, we measure galaxy clustering in CMASS on small and intermediate scales (0.1 less than or similar to r less than or similar to 50 h(-1) Mpc) by computing the two-point correlation function - both projected and redshift-space - of these galaxies, and a new statistic, Sigma(pi), able to separate the coherent and dispersed redshift-space distortion contributions and the large-scale bias. We interpret our clustering measurements by adopting a Halo Occupation Distribution (HOD) scheme that maps them on to high-resolution N-body cosmological simulations to produce suitable mock galaxy catalogues. The traditional HOD prescription can be applied to the red and the blue samples, independently, but this approach is unphysical since it allows the same mock galaxies to be either red or blue. To overcome this ambiguity, we modify the standard formulation and infer the red and the blue models by splitting the full mock catalogue into two complementary and non-overlapping submocks. This separation is performed by constraining the HOD with the observed CMASS red and blue galaxy fractions and produces reliable and accurate models.