Approaches to tackle the redshift distribution calibration challenge for weak lensing cosmology
by
small seminar room
510A
Weak gravitational lensing and galaxy clustering are some of the most powerful probes of cosmology, with current surveys reaching percent-level precision on parameters such as S_8. As we enter the Rubin LSST and Roman era, statistical uncertainties will shrink dramatically, and systematic errors – particularly the calibration of the source redshift distribution n(z) – will become the dominant limitation. Achieving sub-percent precision of n(z) is therefore essential for realizing the full cosmological potential of next-generation imaging surveys.
This talk presents a coordinated program developed through my work with HSC, DESC, and DESI to tackle redshift distribution calibration from three complementary angles. First, I show how RAIL provides a unified framework for photometric redshift estimation, evaluation, and n(z) calibration, validated through simulations in the DESC Photo-z Data Challenge and already demonstrated on Rubin DP1 data. Second, I describe how DESI clustering redshifts are used to calibrate the redshift distribution of the HSC shear catalog, with methods that translate directly to the DESIxLSST overlap and extend calibration beyond z>1.5. Finally, I demonstrate how 3x2pt analysis enables internal redshift self-calibration and consistency checks of the redshift distribution.
These approaches form a coherent strategy to overcome the central systematic challenge for weak lensing cosmology in the LSST era.
Zoom link: https://bnl.zoomgov.com/j/16001410126?pwd=LO1VgLQipgjIfQpOlKB6szuFxr7L7g.1
Xiangchong Li