Evidence Hub

Diffusion-based Galaxy Simulations for the Roman High Latitude Survey

Jake Lee

JPL Data, 2026

Abstract

The Nancy Grace Roman Space Telescope [Roman was formerly named the Wide-Field Infrared Survey Telescope (WFIRST). http://roman.gsfc.nasa.gov/] (hereafter Roman; Spergel 2015, Akeson 2019) will deliver a transformative view of the cosmos with its High Latitude Wide Area Survey (HLWAS; hereafter part of the HLS), a wide-field near-infrared imaging program covering 2,000 square degrees. A central pillar of the HLS is weak gravitational lensing (WL; Bartelmann & Schneider 2001), which will constrain the growth of large-scale structure and the nature of dark energy. Meeting Roman’s ambitious cosmological goals requires controlling shear systematics to unprecedented levels, with multiplicative biases m constrained to m ~ 3.2 x 10^-4 (Dore 2018). Achieving this accuracy depends critically on the realism of galaxy image simulations, which are used to calibrate shear measurements, test photometric redshifts, and evaluate the impact of blending. Simulating Roman-like observations is particularly challenging. The survey will encompass billions of galaxies across a wide range of redshifts, morphologies, and environments, many of which display irregular, clumpy, or multi-component structures only revealed at Hubble Space Telescope (HST; Lallo 2012) and James Webb Space Telescope (JWST; McElwain 2023) resolution. Traditional pipelines based on analytic light profiles—such as Sersic or double-Sersic models—offer interpretability and computational control, but they struggle to reproduce…

Read the paper

Related studies

Go deeper in Evidence Hub

Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.

Open this study Sign up free