1: \begin{abstract}

2: Sky masking is unavoidable in wide-field weak lensing observations.

3: We study how masks affect the measurement of statistics 

4: of matter distribution probed by weak gravitational lensing.

5: We first use 1000 cosmological ray-tracing simulations to examine 

6: in detail the impact of masked regions on the weak lensing Minkowski Functionals (MFs).

7: We consider actual sky masks used for a Subaru Suprime-Cam imaging survey.

8: The masks increase the variance of the convergence field

9: and the expected values of the MFs are biased.

10: The bias then affects the non-Gaussian signals 

11: induced by the gravitational growth of structure.

12: We then explore how masks affect cosmological parameter estimation. 

13: We calculate the cumulative signal-to-noise ratio S/N for masked maps 

14: to study the information content of lensing MFs.

15: We show that the degradation of S/N for masked maps is mainly 

16: determined by the effective survey area.

17: We also perform simple $\chi^2$ analysis to show the impact of 

18: lensing MF bias due to masked regions.

19: Finally, we compare ray-tracing simulations with %the lensing MFs 

20: data from a Subaru 2 deg$^2$ survey

21: in order to address if the observed lensing MFs are consistent 

22: with those of the standard cosmology.

23: The resulting $\chi^2/n_{\rm dof} = 29.6/30$ for three combined MFs,

24: obtained with the mask effects taken into account,  

25: suggests that the observational data are indeed consistent with 

26: the standard $\Lambda$CDM model.

27: We conclude that the lensing MFs are powerful probe of cosmology

28: only if mask effects are correctly taken into account.

29: \end{abstract}

30: