Groundwater availability studies in the arid southwestern United States traditionally have assumed that groundwater discharge by evapotranspiration (ET ) from desert playas is a significant component of the groundwater budget. However, desert playa ET rates are poorly constrained by Bowen ratio energy budget (BREB) and eddy-covariance (EC) micrometeorological measurement approaches. Best attempts by previous studies to constrain ET from desert playas have resulted in ET rates that are within the measurement error of micrometeorological approaches. This study uses numerical models to further constrain desert playa ET rates that are within the measurement error of BREB and EC approaches, and to evaluate the effect of hydraulic properties and salinity-based groundwater density contrasts on desert playa ET rates. Numerical models simulated ET rates from desert playas in Death Valley, California and Dixie Valley, Nevada. Results indicate that actual ET rates from desert playas are significantly below the uncertainty thresholds of BREB- and EC-based micrometeorological measurements. Discharge from desert playas likely contributes less than 2% of total groundwater discharge from Dixie and Death Valleys, which suggests discharge from desert playas also is negligible in other basins. Simulation results also show that ET from desert playas primarily is limited by differences in hydraulic properties between alluvial fan and playa sediments and, to a lesser extent, by salinity-based groundwater density contrasts.