EFFECT OF MASKED REGIONS ON WEAK-LENSING STATISTICS

Abstract

Sky masking is unavoidable in wide-field weak lensing observations. We study how masks affect the measurement of statistics of matter distribution probed by weak gravitational lensing. We first use 1000 cosmological ray-tracing simulations to examine in detail the impact of masked regions on the weak lensing Minkowski Functionals (MFs). We consider actual sky masks used for a Subaru Suprime-Cam imaging survey. The masks increase the variance of the convergence field and the expected values of the MFs are biased. The bias then affects the non-Gaussian signals induced by the gravitational growth of structure. We then explore how masks affect cosmological parameter estimation. We calculate the cumulative signal-to-noise ratio S/N for masked maps to study the information content of lensing MFs. We show that the degradation of S/N for masked maps is mainly determined by the effective survey area. We also perform simple chi^2 analysis to show the impact of lensing MF bias due to masked regions. Finally, we compare ray-tracing simulations with data from a Subaru 2 deg^2 survey in order to address if the observed lensing MFs are consistent with those of the standard cosmology. The resulting chi^2/n_dof = 29.6/30 for three combined MFs, obtained with the mask effects taken into account, suggests that the observational data are indeed consistent with the standard LambdaCDM model. We conclude that the lensing MFs are powerful probe of cosmology only if mask effects are correctly taken into account.