Homogeneous Velocity-Distance Data for Peculiar Velocity Analysis. III. The Mark III Catalog of Galaxy Peculiar Velocities

This is the third in a series of papers in which we assemble and analyze a homogeneous catalog of peculiar velocity data. In Papers I and II, we described the Tully-Fisher (TF) redshift-distance samples that constitute the bulk of the catalog and our methodology for obtaining mutually consistent TF calibrations for these samples. In this paper, we supply further technical details of the treatment of the data and present a subset of the catalog in tabular form. The full catalog, known as the Mark III Catalog of Galaxy Peculiar Velocities, is available in accessible on-line databases, as described herein. The electronic catalog incorporates not only the TF samples discussed in Papers I and II but also elliptical galaxy D_n-σ samples originally presented elsewhere. The relative zero pointing of the elliptical and spiral data sets is discussed here.

The basic elements of the Mark III Catalog are the observables for each object (redshift, magnitude, velocity width, etc.) and inferred distances derived from the TF or D_n-σ relations. Distances obtained from both the forward and inverse TF relations are tabulated for the spirals. Malmquist bias-corrected distances are computed for each catalog object using density fields obtained from the IRAS 1.2 Jy redshift survey. Distances for both individual objects and groups are provided. A variety of auxiliary data, including distances and local densities predicted from the IRAS redshift survey reconstruction method, are tabulated as well. We study the distributions of TF residuals for three of our samples and conclude that they are well approximated as Gaussian. However, for the Mathewson et al. sample we demonstrate a significant decrease in TF scatter with increasing velocity width. We test for, but find no evidence of, a correlation between TF residuals and galaxy morphology. Finally, we derive transformations that map the apparent magnitude and velocity width data for each spiral sample onto a common system. This permits the application of analysis methods that assume that a unique TF relation describes the entire sample.