Testing Substellar Models with Keck LGS AO Dynamical Masses

Trent Dupuy (IfA/Hawaii)

Substellar theoretical models are now widely used in many areas of active research, from the determination of the IMF to the characterization of directly imaged planets. However, despite their extensive use, these models have not been tested over much of the relevant parameter space. We are conducting an orbital monitoring program to determine the dynamical masses of brown dwarf binaries using Keck laser guide star adaptive optics (LGS AO). Our mass measurements provide the strongest tests of substellar models to date, as mass is the most fundamental parameter governing all objects' life histories. Keck LGS AO is ideal for monitoring these binaries, which have typical orbital separations of 50-150 milliarcsec, as its exceptional resolving power is crucial for determining precise astrometry and thereby precise masses (< 5%). Our program has already doubled the number of brown dwarfs with dynamical masses, extending such measurements to much lower temperatures than previous work. Our results show that theoretical models harbor significant systematic errors and do not accurately reproduce observed colors or temperatures, and we find that luminosities may be under predicted by a factor of ~2-3. Such a discrepancy in the predicted luminosity evolution (i.e., cooling rate) of brown dwarfs would have wide-ranging implications as, for example, these models are also used to predict the luminosity and radius evolution of extrasolar giant planets. Finally, we consider the growing sample of brown dwarf binaries as a whole and show how they will provide novel tests of star/brown dwarf formation models. We find that the observed eccentricity distribution can discriminate between different brown dwarf formation models that otherwise reproduce observations well.