We analyse the stellar populations in the host galaxies of 53 X-ray selected optically dull active galactic nuclei (AGN) at 0.34<z<1.07 with ultra-deep (m=26.5) optical medium-band (R~50) photometry from the Survey for High-z Absorption Red and Dead
Sources (SHARDS). The spectral resolution of SHARDS allows us to consistently measure the strength of the 4000 AA break, Dn(4000), a reliable age indicator for stellar populations. We confirm that most X-ray selected moderate-luminosity AGN (L_X<10^44 erg/s) are hosted by massive galaxies (typically M*>10^10.5 M_sun) and that the observed fraction of galaxies hosting an AGN increases with the stellar mass. A careful selection of random control samples of inactive galaxies allows us to remove the stellar mass and redshift dependencies of the AGN fraction to explore trends with several stellar age indicators. We find no significant differences in the distribution of the rest-frame U-V colour for AGN hosts and inactive galaxies, in agreement with previous results. However, we find significantly shallower 4000 AA breaks in AGN hosts, indicative of younger stellar populations. With the help of a model-independent determination of the extinction, we obtain extinction-corrected U-V colours and light-weighted average stellar ages. We find that AGN hosts have younger stellar populations and higher extinction compared to inactive galaxies with the same stellar mass and at the same redshift. We find a highly significant excess of AGN hosts with Dn(4000)~1.4 and light weighted average stellar ages of 300-500 Myr, as well as a deficit of AGN in intrinsic red galaxies. We interpret failure in recognising these trends in previous studies as a consequence of the balancing effect in observed colours of the age-extinction degeneracy.
We report on results from the analysis of a stellar mass-selected (log M*>9.0) sample of 1644 galaxies at 0.65<z<1.1 with ultra-deep (m<26.5) optical medium-band (R~50) photometry from the Survey for High-z Absorption Red and Dead Sources (SHARDS). T
he spectral resolution of SHARDS allows us to consistently measure the strength of the 4000 Angstrom spectral break [Dn(4000), an excellent age indicator for the stellar populations of quiescent galaxies] for all galaxies at z~0.9 down to log M*9. The Dn(4000) index cannot be resolved from broad-band photometry, and measurements from optical spectroscopic surveys are typically limited to galaxies at least x10 more massive. When combined with the rest-frame U-V colour, Dn(4000) provides a powerful diagnostic of the extinction affecting the stellar population that is relatively insensitive to degeneracies with age, metallicity or star formation history. We use this novel approach to estimate de-reddened colours and light-weighted stellar ages for individual sources. We explore the relationships linking stellar mass, (U-V), and Dn(4000) for the sources in the sample, and compare them to those found in local galaxies. The main results are: a) both Dn(4000) and (U-V) correlate with M*. The dispersion in Dn(4000) values at a given M* increases with M*, while the dispersion for (U-V) decreases due to the higher average extinction prevalent in massive star-forming galaxies. b) for massive galaxies, we find a smooth transition between the blue cloud and red sequence in the intrinsic U-V colour, in contrast with other recent results. c) at a fixed stellar age, we find a positive correlation between extinction and stellar mass. d) the fraction of sources with declining or halted star formation increases steeply with the stellar mass, from ~5% at log M*~9.0-9.5 to ~80% at log M*>11, in agreement with downsizing scenarios.
