Effects of the Complex Mass Distribution of Dark Matter Halos on Weak‐Lensing Cluster Surveys

Abstract

Gravitational lensing effects arise from the light-ray deflection by all of the mass distribution along the line of sight. It is then expected that weak-lensing cluster surveys can provide us with true mass-selected cluster samples. With numerical simulations, we analyze the correspondence between peaks in the lensing convergence κ-map and dark matter halos. In particular, we emphasize the difference between the peak κ-value expected from a dark matter halo modeled as an isolated and spherical one, which exhibits a one-to-one correspondence with the halo mass at a given redshift, and that of the associated κ-peak from simulations. For halos with the same expected κ, their corresponding peak signals in the κ-map present a wide dispersion. At an angular smoothing scale of θG = 1', our study shows that for relatively large clusters, the complex mass distribution of individual clusters is the main reason for the dispersion. The projection effect of uncorrelated structures does not play significant roles. The triaxiality of dark matter halos accounts for a large part of the dispersion, especially for the tail on the high-κ side. Thus, lensing-selected clusters are not really mass-selected. To better predict κ-selected cluster abundance for a cosmological model, one has to take into account the triaxial mass distribution of dark matter halos. On the other hand, for a significant number of clusters, their mass distribution is even more complex than that described by the triaxial model. Our analyses find that large substructures affect the identification of lensing clusters considerably. They could show up as separate peaks in the κ-map and cause a misassociation of the whole cluster with a peak resulting only from a large substructure. The lower end dispersion of κ is attributed mostly to this substructure effect. For θG = 2', the projection effect can be significant and contributes to the dispersion at both high- and low-κ ends.