Exploiting intrinsic galaxy correlations to improve redshift and properties

Photometric information will be central to Roman cosmology and galaxy evolution studies, used both to estimate redshifts and to measure distributions and scaling relations of luminosity, stellar mass, and specific star formation rate. Those same quantities can enhance cosmology analyses by separating populations with different intrinsic alignments or clustering for weak lensing and large-scale structure studies, characterizing hosts of Type Ia supernovae, or mapping foreground shear and magnification for strong lens systems and supernovae.

Training sets with well-measured properties remain limited, so new methods must extract the most from sparse data. A recently developed framework applies UMAP to encode optical–IR galaxy spectral energy distributions into a low-dimensional color space and then interpolates between objects with high-quality measurements to predict redshifts for complete photometric samples.

photometric redshifts, roman telescope, galaxy evolution, seds, umap, redshift estimation, stellar mass, star formation, intrinsic alignments, weak lensing