Disk galaxy scaling relations at intermediate redshifts - I. The Tully-Fisher and velocity-size relations

Aims. Galaxy scaling relations such as the Tully-Fisher relation (between the maximum rotation velocity V-max and luminosity) and the velocity-size relation (between V-max and the disk scale length) are powerful tools to quantify the evolution of disk galaxies with cosmic time. Methods. We took spatially resolved slit spectra of 261 field disk galaxies at redshifts up to z approximate to 1 using the FORS instruments of the ESO Very Large Telescope. The targets were selected from the FORS Deep Field and William Herschel Deep Field. Our spectroscopy was complemented with HST/ACS imaging in the F814W filter. We analyzed the ionized gas kinematics by extracting rotation curves from the two-dimensional spectra. Taking into account all geometrical, observational, and instrumental effects, these rotation curves were used to derive the intrinsic V-max. Results. Neglecting galaxies with disturbed kinematics or insufficient spatial rotation curve extent, V-max was reliably determined for 124 galaxies covering redshifts 0.05 < z < 0.97. This is one of the largest kinematic samples of distant disk galaxies to date. We compared this data set to the local B-band Tully-Fisher relation and the local velocity-size relation. The scatter in both scaling relations is a factor of similar to 2 larger at z approximate to 0.5 than at z approximate to 0. The deviations of individual distant galaxies from the local Tully-Fisher relation are systematic in the sense that the galaxies are increasingly overluminous toward higher redshifts, corresponding to an overluminosity Delta M-B = -(1.2 +/- 0.5) mag at z = 1. This luminosity evolution at given V-max is probably driven by younger stellar populations of distant galaxies with respect to their local counterparts, potentially combined with modest changes in dark matter mass fractions. The analysis of the velocity-size relation reveals that disk galaxies of a given V-max have grown in size by a factor of similar to 1.5 over the past similar to 8 Gyr, most likely through accretion of cold gas and/or small satellites. From scrutinizing the combined evolution in luminosity and size, we find that the galaxies that show the strongest evolution toward smaller sizes at z approximate to 1 are not those that feature the strongest evolution in luminosity, and vice versa.