Comparing Spectroscopic and Photometric Stellar Mass Estimates

The purpose of this letter is to compare the quality of different methods forestimating stellar masses of galaxies. We compare the results of (a) fittingstellar population synthesis models to broad band colors from SDSS and 2MASS,(b) the analysis of spectroscopic features of SDSS galaxies (Kauffmann et al.2003), and, (c) a simple dynamical mass estimate based on SDSS velocitydispersions and effective radii. Knowing that all three methods can havesignificant biases, a comparison can help to establish their (relative)reliability. In this way, one can also probe the quality of the observationallycheap broadband color mass estimators for galaxies at higher redshift.Generally, masses based on broad-band colors and spectroscopic features agreereasonably well, with a rms scatter of only ~ 0.25 dex over almost 4 decades inmass. However, as may be expected, systematic differences do exist and have anamplitude of ~ 0.15 dex, corrleting with Halpha emission strength.Interestingly, masses from broad-band color fitting are in better agreementwith dynamical masses than masses based on the analysis of spectroscopicfeatures. In addition, the differences between the latter and the dynamicalmasses correlate with Halpha equivalent width, while this much less the casefor the broad-band masses. We conclude that broad band color mass estimators,provided they are based on a large enough wavelength coverage and use anappropriate range of ages, metallicities and dust extinctions, can yield fairlyreliable stellar masses for galaxies. This is a very encouraging result as suchmass estimates are very likely the only ones available at significant redshiftsfor some time to come.Comment: accepted for publication in ApJ