We present a detailed spectroscopic analysis of 115 helium-line (DB) and 28 cool, He-rich hydrogen-line (DA) white dwarfs based on atmosphere fits to optical spectroscopy and photometry. We find that 63% of our DB population show hydrogen lines, making them DBA stars. We also demonstrate the persistence of pure DB white dwarfs with no detectable hydrogen feature at low effective temperatures. Using state-of-the art envelope models, we next compute the total quantity of hydrogen, M H , that is contained in the outer convection zone as a function of effective temperature and atmospheric H/He ratio. We find that some (T eff , M H ) pairs cannot physically exist as a homogeneously mixed structure; such combination can only occur as stratified objects of the DA spectral type. On that basis, we show that the values of M H inferred for the bulk of the DBA stars are too large and incompatible with the convective dilution scenario. We also present evidence that the hydrogen abundances measured in DBA and cool, heliumrich white dwarfs cannot be globally accounted for by any kind of accretion mechanism onto a pure DB star. We suggest that cool, He-rich DA white dwarfs are most likely created by the convective mixing of a DA star with a thin hydrogen envelope; they are not cooled down DBA's. We finally explore several scenarios that could account for the presence of hydrogen in DBA stars.

We present a critical review of the determination of fundamental parameters of white dwarfs discovered by the Gaia mission. We first reinterpret color-magnitude and color-color diagrams using photometric and spectroscopic information contained in the Montreal White Dwarf Database (MWDD), combined with synthetic magnitudes calculated from a self-consistent set of model atmospheres with various atmospheric compositions. The same models are then applied to measure the fundamental parameters of white dwarfs using the so-called photometric technique, which relies on the exquisite Gaia trigonometric parallax measurements, and photometric data from Pan-STARRS, SDSS, and Gaia. In particular, we discuss at length the systematic effects induced by these various photometric systems. We then study in great detail the mass distribution as a function of effective temperature for the white dwarfs spectroscopically identified in the MWDD, as well as for the white dwarf candidates discovered by Gaia. We pay particular attention to the assumed atmospheric chemical composition of cool, non-DA stars. We also briefly revisit the validity of the mass-radius relation for white dwarfs, and the recent discovery of the signature of crystallization in the Gaia color-magnitude diagram for DA white dwarfs. We finally present evidence that the core composition of most of these white dwarfs is, in bulk, a mixture of carbon and oxygen, an expected result from stellar evolution theory, but never empirically well established before.

We revisit the problem of the formation of DB white dwarfs, as well as the origin of hydrogen in DBA stars, using a new set of envelope model calculations with stratified and mixed hydrogen/helium compositions. We first describe an approximate model to simulate the so-called convective dilution process, where a thin, superficial hydrogen radiative layer is gradually eroded by the underlying and more massive convective helium envelope, thus transforming a DA white dwarf into a DB star. We show that this convective dilution process is able to account for the large increase in the number of DB white dwarfs below T eff ∼ 20, 000 K, but that the residual hydrogen abundances expected from this process are still orders of magnitude lower than those observed in DBA white dwarfs. Scenarios involving the accretion of hydrogen from the interstellar medium or other external bodies have often been invoked to explain these overabundances of hydrogen. In this paper, we describe a new paradigm where hydrogen, initially diluted within the thick stellar envelope, is still present and slowly diffusing upward in the deeper layers of a T eff ∼ 20, 000 K white dwarf. When the convective dilution process occurs, the bottom of the mixed H/He convection zone sinks deep into the star, resulting in large amounts of hydrogen being dredged-up to the stellar surface, a phenomenon similar to that invoked in the context of DQ white dwarfs.

We use 1837 DA white dwarfs with high signal to noise ratio spectra and Gaia parallaxes to verify the absolute calibration and extinction coefficients for the Galaxy Evolution Explorer (GALEX). We use white dwarfs within 100 pc to verify the linearity correction to the GALEX data. We find that the linearity correction is valid for magnitudes brighter than 15.95 and 16.95 for the Far Ultraviolet (FUV) and Near Ultraviolet (NUV) bands, respectively. We also use DA white dwarfs beyond 250 pc to calculate extinction coefficients in the FUV and NUV bands; R FUV = 8.01 ± 0.07 and R NUV = 6.72 ± 0.04. These are consistent with the predicted extinction coefficients for Milky Way type dust in the FUV, but smaller than predictions in the NUV. With well understood optical spectra and state-of-the-art model atmosphere analysis, these white dwarfs currently provide the best constraints on the extinction coefficients for the GALEX data.

G183−35 is an unusual white dwarf that shows an Hα line split into five components, instead of the usual three components seen in strongly magnetic white dwarfs. Potential explanations for the unusual set of lines includes a double degenerate system containing two magnetic white dwarfs and/or rotational modulation of a complex magnetic field structure. Here we present time-resolved spectroscopy of G183−35 obtained at the Gemini Observatory. These data reveal two sets of absorption lines that appear and disappear over a period of about 4 hours. We also detect low-level (0.2%) variability in optical photometry at the same period. We demonstrate that the spectroscopic and photometric variability can be explained by the presence of spots on the surface of the white dwarf and a change in the average field strength from about 4.6 MG to 6.2 MG. The observed variability is clearly due to G183−35's relatively short spin period. However, rotational modulation of a complex magnetic field by itself cannot explain the changes seen in the central Hα component. An additional source of variability in the line profiles, most likely due to a chemically inhomogeneous surface composition, is also needed. We propose further observations of similar objects to test this scenario.