Hubble Space Telescope proper motion (HSTPROMO) catalogs of Galactic globular clusters. III. Dynamical distances and mass-to-light ratios

TL;DR

This study uses Hubble Space Telescope data to estimate dynamical distances and mass-to-light ratios for 15 globular clusters, validating photometric methods and constraining stellar initial mass functions.

Contribution

First dynamical distance estimates for most clusters, validating photometric distances and constraining the initial mass function in globular clusters.

Findings

Dynamical and photometric distances agree within ~2%.

Dynamical M/L ratios align with stellar-population models assuming Chabrier or Kroupa IMFs.

Results strongly exclude a Salpeter IMF.

Abstract

We present dynamical distance estimates for 15 Galactic globular clusters and use these to check the consistency of dynamical and photometric distance estimates. For most of the clusters, this is the first dynamical distance estimate ever determined. We extract proper-motion dispersion profiles using cleaned samples of bright stars from the Hubble Space Telescope proper-motion catalogs recently presented in Bellini et al. (2014) and compile a set of line-of-sight velocity-dispersion profiles from a variety of literature sources. Distances are then estimated by fitting spherical, non-rotating, isotropic, constant mass-to-light (M/L) dynamical models to the proper-motion and line-of-sight dispersion profiles together. We compare our dynamical distance estimates with literature photometric estimates from the Harris (1996, 2010 edition) globular cluster catalog and find that the mean fractional difference between the two types is consistent with zero at just $-1.9 \pm 1.7 \%$. This indicates that there are no significant biases in either estimation method and provides an important validation of the stellar-evolution theory that underlies photometric distance estimates. The analysis also estimates dynamical M/L ratios for our clusters; on average, the dynamically-inferred M/L ratios agree with existing stellar-population-based M/L ratios that assume a Chabrier initial mass function (IMF) to within $-8.8 \pm 6.4 \%$, implying that such an IMF is consistent with our data. Our results are also consistent with a Kroupa IMF, but strongly rule out a Salpeter IMF. We detect no correlation between our M/L offsets from literature values and our distance offsets from literature values, strongly indicating that our methods are reliable and our results are robust.