Abstract

We analyse the mass-size relation of similar to 400 quiescent massive ETGs (M-*/M-circle dot > 3 x 10(10)) hosted by massive clusters (M-200 similar to 2-7 x 10(14) M-circle dot) at 0.8 < z < 1.5, compared to those found in the field at the same epoch. Size is parametrized using the mass-normalized B-band rest-frame size, gamma = Re/M-11(0.57). We find that the gamma distributions in both environments peak at the same position, but the distributions in clusters are more skewed towards larger sizes. This tail induces average sizes similar to 30-40 per cent larger for cluster galaxies than for field galaxies of similar stellar mass, while the median sizes are statistically the same with a difference of similar to 10 +/- 10 per cent. Since this size difference is not observed in the local Universe, the evolution of average galaxy size at fixed stellar mass from z similar to 1.5 for cluster galaxies is less steep at more than 3 sigma (alpha(1 + z)(-0.53 +/- 0.04)) than the evolution of field galaxies (alpha(1 + z)(-0.92) (+/- 0.04)). The difference in evolution is not measured when the median values of. are considered: gamma(1 + z)(-0.84 +/- 0.04) in the field versus alpha(1 + z)(-0.71 +/- 0.05) in clusters. In our sample, the tail of large galaxies is dominated by galaxies with 3 x 10(10) < M-*/ M-circle dot < 10(11). At this low-mass end, the difference in the average size is better explained by the accretion of new galaxies that are quenched more efficiently in clusters and/ or by different morphological mixing in the cluster and field environments. If part of the size evolution would be due to mergers, the difference that we see between cluster and field galaxies could be caused by higher merger rates in clusters at higher redshift, when galaxy velocities are lower.