Observational Evidence for Large-Scale Gas Heating in a Galaxy Protocluster at z=2.30

Abstract: We report a $z=2.30$ galaxy protocluster (COSTCO-I) in the COSMOS field, where the Lyman-$\alpha$ forest as seen in the CLAMATO IGM tomography survey does not show significant absorption. This departs from the transmission-density relationship (often dubbed the fluctuating Gunn-Peterson approximation; FGPA) usually expected to hold at this epoch, which would lead one to predict strong Ly$\alpha$ absorption at the overdensity. For comparison, we generate mock Lyman-$\alpha$ forest maps by applying FGPA to constrained simulations of the COSMOS density field, and create mocks that incorporate the effects of finite sightline sampling, pixel noise, and Wiener filtering. Averaged over $r=15\,h^{-1}\,\mathrm{Mpc}$ around the protocluster, the observed Lyman-$\alpha$ forest is consistently more transparent in the real data than in the mocks, indicating a rejection of the null hypothesis that the gas in COSTCO-I follows FGPA ($p=0.0026$, or $2.79 \sigma$ significance). It suggests that the large-scale gas associated with COSTCO-I is being heated above the expectations of FGPA, which might be due to either large-scale AGN jet feedback or early gravitational shock heating. COSTCO-I is the first known large-scale region of the IGM that is observed to be transitioning from the optically-thin photoionized regime at Cosmic Noon, to eventually coalesce into an intra-cluster medium (ICM) by $z=0$. Future observations of similar structures will shed light on the growth of the ICM and allow constraints on AGN feedback mechanisms.