A direct detection method of gravitational shear-galaxy intrinsic alignment correlation in non-linear regimes for photometric surveys

Intrinsic alignment (IA) of galaxies is a serious source of contamination in weak-lensing signals and a primary systematic for photometric surveys. If IA is ignored, the determination of the dark energy equation of state can be biased by as much as 50%, and the matter power spectrum amplitude can be affected by up to 30%. Thus, it is important to develop methods to isolate the IA systematics and remove it from the cosmic shear signal with better accuracy. The self-calibration (SC) technique separates the gravitational shear-galaxy Intrinsic alignment (GI) correlation and the gravitational shear-gravitational shear (GG) correlation in the measured shape-shape correlations. After a direct measurement of the galaxy density-galaxy intrinsic alignment (gI) correlation in the surveys, the SC determines the GI correlation through a scaling relation for cross-correlating redshift bins. We develop a modification to the SC scaling relation which allows the mitigation of the GI correlation beyond the linear regime. The method is suitable for a non-linear galaxy bias model and works at a scale similar to the tidal alignment and tidal torquing (TATT) model for intrinsic ellipticity. We test this new SC with a toy model, and we observe the scaling relation is accurate within 5% for non-adjacent redshift bins and within 10% for adjacent redshift bins. Hence, the suppression of the GI contamination by a factor of 20 and 10 or larger is possible, respectively for non-adjacent and adjacent bins. We test the robustness of the new SC up to a variation of 20% in the IA parameters and observe a suppression factor of 10 is still achievable for the GI signal in all combinations of bins.