Cosmic Cartography: Photometric Redshifts for Next-Generation Sky Surveys

Knowing the distances to galaxies as measured by their cosmological redshift is crucial for studies of cosmology, galaxy evolution, and astronomical transients. The next generation of astronomical imaging surveys (like LSST, Euclid, and Roman Observatories) will all be critically dependent on estimates of galaxy redshifts from imaging data alone; the resulting measurements are called photometric redshifts or photo-z's. Traditional photo-z estimation methods only use measures of total light received from a galaxy (colors and magnitudes) as inputs, thereby, throwing away the rich pixel-level information present in images. Moreover, the uncertainty estimates produced by these methods are not statistically well defined and the availability of data to train these methods is scarce. I will present my work on developing new deep learning-based photo-z estimation methods that take images directly as inputs and provide state-of-the-art photo-z prediction accuracy while being interpretable and requiring less training data. I will also talk about a statistical formalism that I developed to produce well-calibrated photo-z uncertainty estimates that are method-agnostic and employ minimal assumptions. Finally, I will also provide an overview of our recent efforts to obtain spectroscopic samples to train for photo-z algorithms using the Dark Energy Spectroscopic Instrument (DESI).