Clustering of quasars in a wide luminosity range at redshift 4 with Subaru Hyper Suprime-Cam Wide-field imaging

Abstract We examine the clustering of quasars over a wide luminosity range, by utilizing 901 quasars at $\overline{z}_{\rm phot}\sim 3.8$ with −24.73 < M1450 < −22.23 photometrically selected from the Hyper Suprime-Cam Subaru Strategic Program (HSC-SSP) S16A Wide2 date release and 342 more luminous quasars at 3.4 < zspec < 4.6 with −28.0 < M1450 < −23.95 from the Sloan Digital Sky Survey that fall in the HSC survey fields. We measure the bias factors of two quasar samples by evaluating the cross-correlation functions (CCFs) between the quasar samples and 25790 bright z ∼ 4 Lyman break galaxies in M1450 < −21.25 photometrically selected from the HSC dataset. Over an angular scale of 10${^{\prime\prime}_{.}}$0 to 1000${^{\prime\prime}_{.}}$0, the bias factors are $5.93^{+1.34}_{-1.43}$ and $2.73^{+2.44}_{-2.55}$ for the low- and high-luminosity quasars, respectively, indicating no significant luminosity dependence of quasar clustering at z ∼ 4. It is noted that the bias factor of the luminous quasars estimated by the CCF is smaller than that estimated by the auto-correlation function over a similar redshift range, especially on scales below 40${^{\prime\prime}_{.}}$0. Moreover, the bias factor of the less-luminous quasars implies the minimal mass of their host dark matter halos is 0.3–2 × 1012 h−1 M⊙, corresponding to a quasar duty cycle of 0.001–0.06.