Homogeneous velocity-distance data for peculiar velocity analysis. II. Calibration of field samples

This is the second in a series of papers in which we assemble and analyze a homogeneous catalog of peculiar velocity data. In Paper I, we treated two Tully-Fisher (TF) cluster samples. Here, we study four TF field samples: the I-band sample of Mathewson, Ford, & Buchhorn, the r-band samples of Willick and Courteau-Faber, and the H-band sample of Aaronson et al. as revised by Tormen & Burstein. In Paper III, we will combine these spiral samples with elliptical data to form the Mark III Catalog of Galaxy Peculiar Velocities, which will be used as input for POTENT and other velocity analyses. We derive the TF slope and scatter separately for each sample, correcting for selection biases and self-consistency determining internal-extinction coefficients. We find no evidence for a luminosity dependence of internal extinction. We fit both forward and inverse TF relations and, by comparing the two, validate the bias corrections applied to the forward fits. By matching TF distances for several hundred galaxies common to two or more samples, we determine the relative zero points of the various TF relations. The global zero point is set by the Han-Mould TF calibration carried out in Paper I. We calibrate the TF relations after grouping field galaxies in redshift space, making conservative use of preliminary TF relations based on Hubble flow distances. The differences between the preliminary and final TF scatter estimates indicate typical radial peculiar velocities of ∼250-300 km s^-1. We find observed scatters σ_TF ≈ 0.38-0.43 mag for the r- and I-band samples, larger than many recent estimates, and estimate the intrinsic TF scatter to be ∼0.30 mag. Based on simulated catalogs, we find that our observed scatter estimates are good to ≲ 5%.