A reanalysis of multiwavelength observations of the pulsating DA white dwarf G117-B15A

In the light of recent developments in the study of the atmospheric properties of DA white dwarfs, we present a reanalysis of the multiwavelength observations of the ZZ Ceti star G117-B15A obtained by Robinson et al. Following a procedure similar to that used by these authors, we compare the observed amplitudes of the dominant pulsation mode of that star in six different bandpasses with theoretical amplitudes computed from model atmospheres in order to constrain the pulsation index I of the mode and the atmospheric parameters of G117-B15A. Our approach, however, allows for the sensitivity of the predicted pulsation amplitudes to the convective efficiency used in the model atmosphere calculations. Thus, we use three grids of model atmospheres based on the ML2, ML2/alpha = 0.6, and ML1 parameterization of the mixing-length theory (MLT). We find, as did Robinson ct al., that the observations are best explained if l = 1. However, for each version of the MLT, the best-fitting models form a different family of solutions in the T-eff-log g plane. Since we cannot discriminate fully between the various versions of the MLT on the basis of the pulsation data alone, we invoke constraints derived in an independent analysis of the time-averaged optical and ultraviolet spectra of G117-B15A. These constraints indicate that the pulsation results are incompatible with the time-averaged spectrum of that star if model atmospheres based on ML2 and ML1 convection are used in the analysis. We find, however, that a model of G117-B15A with T-eff similar to 11,500 K, log g similar to 8.0, and ML2/alpha = 0.6 convection is consistent with both the time-averaged spectroscopic observations and the observed pulsation amplitudes.