Title:The faint end of the red sequence galaxy luminosity function: unveiling surface brightness selection effects with the CLASH clusters

Abstract:Characterizing the evolution of the faint end of the cluster red sequence (RS) galaxy luminosity function (GLF) with redshift is a milestone in understanding galaxy evolution. However, the community is still divided in that respect, hesitating between an enrichment of the RS due to efficient quenching of blue galaxies from $z\sim1$ to present-day or a scenario in which the RS is built at a higher redshift and does not evolve afterwards. Recently, it has been proposed that surface brightness (SB) selection effects could possibly solve the literature disagreement, accounting for the diminishing of the RS faint population in ground based observations. We investigate this hypothesis by comparing the RS GLFs of 16 CLASH clusters computed independently from ground-based Subaru/Suprime-Cam and HST/ACS images in the redshift range $0.187\leq z\leq0.686$. We stack individual cluster GLFs in redshift and mass bins.
We find similar RS GLFs for space and ground based data, with a difference of 0.2$\sigma$ in the faint end parameter $\alpha$ when stacking all clusters together and a maximum difference of 0.9$\sigma$ in the case of the high redshift stack, demonstrating a weak dependence on the type of observations in the probed range of redshift and mass. When considering the full sample, we estimate $\alpha = -0.76 \pm 0.07$ and $\alpha = -0.78 \pm 0.06$ with HST and Subaru respectively. We note a mild variation of the faint end with redshift at a 1.7$\sigma$ and 2.6$\sigma$ significance. We investigate the effect of SB dimming by simulating our low redshift galaxies at high redshift. We measure an evolution in the faint end slope of less than 1$\sigma$ in this case, implying that the observed signature is moderately larger than one would expect from SB dimming alone, and indicating a true evolution in the faint end slope. (Abridged...)