Modeling baryonic effects for future weak lensing survey

The modification of matter distribution due to baryonic physics is one of the major theoretical uncertainties in cosmological weak lensing measurements. Developing mitigation schemes for effects of baryons is crucial to assure robustness of cosmological analysis, and has potential to increase the constraining power by extracting information from small scale data. In this talk, I will introduce a framework to model baryonic effects through the principal component analysis (PCA). Utilizing several state-of-the-art hydrodynamical simulations, we learn the deviations of matter power spectra caused by various effects of baryonic physics and extract the dominant features of these diverse deviations into a few PC modes. By projecting our observational data vector into these PC modes, we can then efficiently mitigate baryonic uncertainty by throwing these PC modes away. I will forecast the performance of the PCA framework in marginalizing baryonic uncertainties for future LSST-like survey, and compare it with a halo-model based modeling method under several baryonic scenarios.

Speaker: Hung-Jin Huang, Carnegie Mellon Univ.