Abstract

Time delay cosmography with strongly lensed supernovae (SNe) is an exciting local probe of H0 that is independent of the local distance ladder. One of the most important ingredients in H0 estimates from strongly lensed SNe is the time delay between the appearance of the multiple images of the SN. In this talk, I describe GausSN – a new method for extracting time delays from multi-band photometric observations of resolved lensed SNe images using Gaussian Processes. Our methodology improves upon existing time delay estimation methods by including a fully Bayesian exploration of the parameter space and a novel treatment of microlensing, all with minimal assumptions about the underlying shape of the light curve. We demonstrate the ability of GausSN to recover accurate and precise time delays using simulations of lensed SNe data as expected from the Roman Space Telescope. Compared to existing methods, GausSN recovers time delays which are on average as close to the truth with better calibrated uncertainties. With the upcoming Rubin Observatory’s Legacy Survey of Space and Time and Roman Space Telescope, we expect to discover tens to hundreds of lensed SNe, which, with tools like GausSN, will provide important independent evidence for investigating the Hubble tension in the coming decade.