The splashback radius of optically selected clusters with Subaru HSC Second Public Data Release

Recent constraints on the splashback radius around optically selected galaxy clusters from the redMaPPer cluster-finding algorithm in the literature have shown that the observed splashback radius is $\sim 20\%$ smaller than that predicted by N-body simulations. We present analyses on the splashback features around $\sim 3000$ optically selected galaxy clusters detected by the independent cluster-finding algorithm CAMIRA over a wide redshift range of $0.1<z_{\rm cl}<1.0$ from the second public data release of the Hyper Suprime-Cam (HSC) Subaru Strategic Program covering $\sim 427~{\rm deg}^2$ for the cluster catalog. We detect the splashback feature from the projected cross-correlation measurements between the clusters and photometric galaxies over the wide redshift range, including for high redshift clusters at $0.7<z_{\rm cl}<1.0$, thanks to deep HSC images. We find that constraints from red galaxy populations only are more precise than those without any color cut, leading to $1\sigma$ precisions of $\sim 15\%$ at $0.4<z_{\rm cl}<0.7$ and $0.7<z_{\rm cl}<1.0$. These constraints are more consistent with the model predictions ($\lesssim 1\sigma$) than their $20\%$ smaller values as suggested by the previous studies with the redMaPPer ($\sim 2\sigma$). We also investigate selection effects of the optical cluster-finding algorithms on the observed splashback features by creating mock galaxy catalogs from a halo occupation distribution model, and find that such effects to be sub-dominant for the CAMIRA cluster-finding algorithm. We also find that the redMaPPer-like cluster-finding algorithm induces a smaller inferred splashback radius in our mock catalog, especially at lower richness, which can well explain the smaller splashback radii in the literature. In contrast, these biases are significantly reduced when increasing its aperture size.