The effect of substructure on mass estimates of galaxies

Large galaxies are thought to form hierarchically, from the accretion and disruption of many smaller galaxies. Such a scenario should naturally lead to galactic phase-space distributions containing some degree of substructure. We examine the errors in mass estimates of galaxies and their dark halos made using the projected phase-space distribution of a tracer population (such as a globular cluster system or planetary nebulae) due to falsely assuming that the tracers are distributed randomly. The level of this uncertainty is assessed by applying a standard mass estimator to samples drawn from 11 random realizations of galaxy halos containing levels of substructure consistent with current models of structure formation. We find that substructure will distort our mass estimates by up to similar to 20% - a negligible error compared to statistical and measurement errors in current derivations of masses for our own and other galaxies. However, this represents a fundamental limit to the accuracy of any future mass estimates made under the assumption that the tracer population is distributed randomly, regardless of the size of the sample or the accuracy of the measurements.