3:30–4:30 pm ERC 501
Host: Alex Drlica-Wagner
Giulia Giannini (UChicago/KICP) "Photometric redshift calibration in the Dark Energy Survey"
Photometric redshift calibration in the Dark Energy Survey” Abstract: Biased redshift calibration is one of the limiting factors of wide-field imaging surveys. Cosmological constraints from weak gravitational lensing suffer particularly from this misestimation, relying on correct redshift distributions of both the foreground (lenses) and background (sources) galaxies. The Dark Energy Survey (DES) for the analysis of the first three years of data (Y3) has developed a redshift calibration pipeline that combines the strengths of a photometry-based and a clustering-based technique. The first method leverages machine learning in a Bayesian scheme where deep fields act as intermediate steps between a high quality redshift sample and the DES wide field. The second method exploits instead the spatial overlap of another high quality redshift sample with the DES wide field, measuring the angular two-point cross-correlation between the two samples. The combination is performed through a joint likelihood approach, and it is particularly powerful as the dominant systematic uncertainties related to photometric measurements do not affect the clustering method, and vice versa. We show that using this technique for the redshift calibration of the DES Y3 lens sample has an impact on cosmological constraints, highlighting the importance of the redshift calibration in view of future multi-probe cosmological analyses.
