Rapid localization of gravitational wave hosts with FIGARO

Abstract: The copious scientific literature produced after the detection of GW170817 electromagnetic counterpart demonstrated the importance of a prompt and accurate localization of the gravitational wave within the co-moving volume. In this letter, we present FIGARO, a ready to use and publicly available software that relies on Bayesian non-parametrics. FIGARO is designed to run in parallel with parameter estimation algorithms to provide updated three-dimensional volume localization information. Differently from any existing algorithms, the analytical nature of the FIGARO reconstruction allows a ranking of the entries of galaxy catalogues by their probability of being the host of a gravitational wave event, hence providing an additional tool for a prompt electromagnetic follow up of gravitational waves. We illustrate the features of FIGARO on binary black holes as well as on GW170817. Finally, we demonstrate the robustness of FIGARO by producing so-called pp-plots and we present a method based on information entropy to assess when, during the parameter estimation run, it is reasonable to begin releasing skymaps.