SB/BNL Joint Cosmo seminar (at Stony Brook): Alex Malz, NYU: Probabilistic Photometric Redshifts for Next-Generation Galaxy Surveys
→
US/Eastern
Universe
Universe
Description
Upcoming and ongoing galaxy surveys will and do produce redshift probability distribution functions (PDFs) in addition to traditional photometric redshift (photo-z) point estimates for numerous and faint galaxies, in lieu of unattainable spectroscopic redshifts. The storage of photo-z PDFs may present a challenge with increasingly large catalogs as we face a trade-off between the accuracy of subsequent science measurements and storage costs. I present an approach to optimizing the choice of storage parametrization for a survey producing a catalog of 1-D PDFs and demonstrate it on LSST-like photo-z PDFs. Additionally, photo-z PDFs have been underutilized and misused in cosmological inference thus far, potentially leading to inappropriate propagation of redshift errors to cosmological parameter constraints. I present a mathematically self-consistent Bayesian hierarchical model for estimating the redshift distribution function, necessary for weak gravitational lensing, based on a catalog of photo-z PDFs, and compare it with traditional methods.