The properties of field elliptical galaxies at intermediate redshift -: III.: The Fundamental Plane and the evolution of stellar populations from z ≈ 0.4 to z=0

We report on the study of a sample of 25 field early-type galaxies, in the redshift range z approximate to0.1-0.5, selected on the basis of colours and morphology from the Medium Deep Survey (MDS) of the Hubble Space Telescope (HST). Surface photometry in two colours (F606W and F814W) and redshifts have been derived for all the galaxies in the sample, while velocity dispersions have been measured for 19 of the sample galaxies, as described in a companion paper. Our study of the evolution of the Fundamental Plane (FP) with redshift yields the following results. Field early-type galaxies define a tight FP out to z approximate to0.4, with scatter unchanged with respect to local samples, within the observational errors. The intermediate redshift FP is offset with respect to the local FP of the Coma cluster, in the sense that, for a given effective radius and velocity dispersion, galaxies are brighter than expected from the local relation. The offset of the FP is found to increase with redshift. The range of parameters covered by our sample is not sufficiently extended to measure the slopes of the FP at intermediate redshift. Similar results are found for the SBe-R-e relation, out to z approximate to0.5. The evolution of the FP (and of the SBe-R-e relation) is studied quantitatively with a Bayesian-Monte Carlo technique. By applying this technique, we find that the offset of the intercept of the FP (Delta gamma) with respect to the local FP increases as Delta gamma = tauz with the following 68 per cent limits: 0.33 < <tau> < 0.44 (for <Omega> = 1,Omega (Lambda) = 0) or 0.44 < <tau> < 0.56 (for <Omega> = 0.3, Omega (Lambda) = 0.7). In addition, we interpret the results in terms of the evolution of the stellar populations, under the assumption of passive evolution. In a single-burst scenario, the observed properties are consistent with those of a stellar population formed at z greater than or similar to2 (for Omega = 1, Omega (Lambda) = 0, H-0 = 50 km s(-1) Mpc(-1)) or 0.8 < z < 1.6 (for Omega = 0.3, Omega (Lambda) = 0.7, H-0 = 65 km s(-1) Mpc(-1)). If a small fraction of the stellar mass is formed in a secondary burst, the primary burst may have occurred at higher z. For example (for Omega = 0.3, Omega (Lambda) = 0.7, H-0 = 65 km s(-1) Mpc(-1)) the primary burst may have occurred at z greater than or similar to3 if a secondary burst with a tenth of the stellar mass occurred at 0.6 < z < 0.8. Finally, the intercept and scatter of the FP found for field early-type galaxies and for cluster data (taken from the literature) at z approximate to 0.3-0.4 are mutually consistent, within the observational errors. If higher redshift (up to z = 0.83) cluster data are considered, the ages of the stellar populations of field early-type galaxies inferred from a single-burst scenario are found to be marginally smaller than the ages derived for the cluster galaxies.