Implementing a Robust Test of Galaxy Catalogue Completeness for Dark Siren Measurements of the Hubble Constant

Abstract

We present the application of a robust test of galaxy catalogue completeness to the gwcosmo pipeline. The method implements a straightforward statistical test for determining the apparent magnitude completeness limit of a magnitude-redshift sample. This offers an improved, less conservative approach compared with how galaxy catalogue completeness is currently estimated in the gwcosmo gravitational wave cosmology pipeline for determining the Hubble constant $H_{0}$. The test also does not require prior knowledge of the luminosity function, and thus returns a more robust estimate of the limiting apparent magnitude for a magnitude-redshift sample of galaxies. For GWTC-1 results using $B_{J}$-band photometry of galaxies in the GLADE catalogue, we find a $3.4\%$ improvement on the inference of $H_{0}$ using dark sirens only and a $1.3\%$ improvement for the combined posterior with GW170817. Using GLADE+, there is a $8.6\%$ improvement with dark sirens only and a $6.3\%$ improvement for the combined posterior with GW170817. However, the final posterior on $H_{0}$ using the GWTC-3 dataset with the GLADE+ $K$-band shows no improvement when applying the robust method. This is because the GLADE+ galaxy catalogue provides little or no coverage in the $K$-band for any of the GWTC-3 events. With the use of deeper galaxy catalogues in future gravitational wave cosmology analyses, the adoption of a less conservative estimate of magnitude completeness will become increasingly important.