Description
Imaging Systematics Treatment for DESI ELGs: Effects on Constraining Primordial non-Gaussianity
Angular imaging systematics can introduce non-trivial spurious large-scale correlations in the observed target density along the sky. Hence, the Dark Energy Spectroscopic Instrument (DESI) survey, like other surveys, has an ongoing imaging systematic mitigation effort. This effect is strongest for the Emission Line Galaxies (ELGs) sample, and its treatment requires non-linear mitigation techniques, such as machine learning. Such mitigation methods must be employed with care and require rigorous validation as they can over-fit and remove modes at large scales, especially hindering the accurate measurement of primordial non-Gaussianity (PNG). My talk will give an overview of the DESI imaging systematics on the two-point clustering statistics including the BAO measurements and how one can validate and robustly extract PNG information using cross-tracers of ELGs, LRGs, and QSOs.
