Cross-correlating Planck tSZ with RCSLenS weak lensing: implications for cosmology and AGN feedback

Abstract

We present measurements of the spatial mapping between (hot) baryons and the total matter in the Universe, via the cross-correlation between the thermal Sunyaev-Zeldovich (tSZ) map from Planck and the weak gravitational lensing maps from the Red Sequence Cluster Survey (RCSLenS). The cross-correlations are performed on the map level where all the sources (including diffuse intergalactic gas) contribute to the signal. We consider two configuration-space correlation function estimators, ξ y−κ and ξ y−γ t , and a Fourier space estimator, C y−κ , in our analysis. We detect a significant correlation out to three degrees of angular separation on the sky. Based on statistical noise only, we can report 13σ and 17σ detections of the cross-correlation using the configuration-space y − κ and y − γ t estimators, respectively. Including a heuristic estimate of the sampling variance yields a detection significance of 6σ and 8σ, respectively. A similar level of detection is obtained from the Fourier-space estimator, C y−κ . As each estimator probes different dynamical ranges, their combination improves the significance of the detection. We compare our measurements with predictions from the cosmo-OWLS suite of cosmological hydrodynamical simulations, where different galactic feedback models are implemented. We find that a model with considerable AGN feedback that removes large quantities of hot gas from galaxy groups and WMAP-7yr best-fit cosmological parameters provides the best match to the measurements. All baryonic models in the context of a Planck cosmology over-predict the observed signal. Similar cosmological conclusions are drawn when we employ a halo model with the observed 'universal' pressure profile.

Lensing and the Warm-hot Intergalactic Medium

“…Since at most 15% of the measured signal comes from unbound gas if we restrict the overall IGM contribution to be that fraction of the overall signal then only models with α ≤ 1.5 are compatible with the data. In other words, the mean temperature of the IGM free electron gas would beT e ≤ 7 × 10 6 K. The data from Hojjati et al (2016) shown in Fig 4a are even more restrictive. These authors compared their measurements against the predictions of the halo model and from numerical simulations that included diffuse gas.…”

“…What is more relevant is that the overall shape is very different. Hojjati et al (2016) found that the shape of the correlation function and power spectrum was strongly dependent on physical processes undergone by baryons in halos such as radiative cooling, star formation, supernovae winds and AGN feedback. For instance, AGNs expel gas to large distances from the center of halos, lowering the signal at small scales.…”

“…Search for absorption lines and SZ contributions have provided preliminary evidence of its existence. The κY C cross-correlation measured by Van Waerbeke et al (2014) and Hojjati et al (2016) probes the fluctuations on the electron pressure along the line of sight and its distribution but it is not yet a detection of the missing baryon component. As indicated by Hojjati et al (2015) about 50% of the signal comes from the small fraction of baryons within massive halos; at most, 15% of the cross-correlation power at ℓ ∼ 500 could come from unbound gas.…”

“…1b; dashed (blue), dot-dashed (red) and triple dot-dashed (green) lines correspond to z = 0.1, 0.3, 0.5 respectively. The more recent analysis by Hojjati et al (2016) uses the deeper RCSLenS catalog that includes all galaxies with mag r > 18. These authors provide a numerical fit of their lens distribution, plotted in Fig.…”

“…The latter contribution is degenerate with respect to cosmological and physical parameters and the data did not permit a robust inference (Battaglia et al 2015). Hojjati et al (2016) improved the statistical significance of the lensing -tSZ cross-correlation by using a larger weak lensing map derived from the Red Sequence Cluster Lensing Survey (RCSLenS) and found that their signal was best interpreted if AGN feedback removed a large quantity of hot gas from galaxy groups.…”