The Tully-Fisher relation of distant field galaxies

We examine the evolution of the Tully-Fisher relation (TFR) using a sample of 89 field spirals for which we have measured confident rotation velocities (V-rot). This sample covers the redshift range 0.1 less than or similar to z less than or similar to 1, with a median of < z >= 0.33. The best-fitting TFR has a slope consistent with that measured locally, and we find no significant evidence for a change with redshift, although our sample is not large enough to well constrain this. By plotting the residuals from the local TFR versus redshift, we find evidence that these luminous (M-B less than or similar to M*(B)) spiral galaxies are increasingly offset from the local TFR with redshift, reaching a brightening of -1.0 +/- 0.5 mag, for a given V-rot, by z similar to 1. This is supported by fitting the TFR to our data in several redshift bins, suggesting a corresponding brightening of the TFR intercept. Since selection effects would generally increase the fraction of intrinsically bright galaxies at higher redshifts, we argue that the observed evolution is probably an upper limit. Previous studies have used an observed correlation between the TFR residuals and V-rot to argue that low-mass galaxies have evolved significantly more than those with higher mass. However, we demonstrate that such a correlation may exist purely due to an intrinsic coupling between the V-rot scatter and TFR residuals, acting in combination with the TFR scatter and restrictions on the magnitude range of the data, and therefore it does not necessarily indicate a physical difference in the evolution of galaxies with different V-rot. Finally, if we interpret the luminosity evolution derived from the TFR as due to the evolution of the star formation rate (SFR) in these luminous spiral galaxies, we find that SFR(z) proportional to (1 +z)(1.7 +/- 1.1). Notwithstanding the relatively large uncertainty, this evolution, which is probably overestimated due to selection effects, seems to be slower than the one derived for the overall field galaxy population. This suggests that the rapid evolution in the SFR density of the Universe observed since z similar to 1 is not driven by the evolution of the SFR in individual bright spiral galaxies.