Galaxy groups up to z=2.5 in deep near-infrared surveys : detection and quenched fractions

Obtaining large samples of galaxy groups and (proto-)cluster cores at redshift z > 1.5 is crucial to better understand cluster formation and the role environment plays in shaping galaxy properties. While dozens of these structures have now been studied, a systematic search for such systems has proven challenging. Exploiting deep near-infrared selected photometric data is a promising approach. We designed a new group finder algorithm, the DElaunay TEssellation ClusTer IdentiFication with photo-z (DETECTIFz) to detect galaxy groups in photometric data. The algorithm uses the joint PDF of redshift and stellar mass of galaxies to identify groups as stellar-mass over-densities. Using data from three of the deepest near-infrared surveys: UKIDSS UDS, VIDEO, and UltraVISTA, that were homogeneously reduced (the REFINE survey), we build a pure (>90%) sample of 448 candidate groups up to z=2.5 and study some of their properties. In this talk, I present results from Sarron & Conselice (accepted in MNRAS). I will detail the DETECTIFz algorithm and show its performances on mock galaxy catalogues. I will show some properties of our group catalogue. In particular, we find that galaxies with $10.25<\log (M_{\star }/{\mathrm{M}}_{\odot })<11$ have higher quenched fractions in groups than in the field up to z=2.2, the difference growing with decreasing redshift.