A new look at the pulsating DB white dwarf GD 358: Line-of-sight velocity measurements and constraints on model atmospheres^*

¹ Lund Observatory Box 43, 22100 Lund, Sweden e-mail: rubina@astro.lu.se
² Imperial College of Science, Technology, and Medicine, Blackett Laboratory, Prince Consort Road, London, SW7 2BZ, UK e-mail: rubina@ic.ac.uk
³ Astronomical Institute, Utrecht University, PO Box 80000, 3508 TA Utrecht, The Netherlands e-mail: M.H.vanKerkwijk@astro.uu.nl
⁴ Department of Astronomy and Astrophysics, University of Toronto, 60 St George Street, Toronto, Ontario M5S 3H8, Canada e-mail: mhvk@astro.utoronto.ca
⁵ Department of Physics and Astronomy, University of North Carolina, Chapel Hill, NC 27599-3255, USA e-mail: clemens@physics.unc.edu
⁶ Institut für Theoretische Physik und Astrophysik, Universität Kiel, 24098 Kiel, Germany e-mail: koester@astrophysik.uni-kiel.de

Corresponding author: R. Kotak, rubina@ph.ic.ac.uk

Received: 16 September 2002
Accepted: 30 October 2002

Abstract

We report on our findings of the bright, pulsating, helium atmosphere white dwarf GD 358, based on time-resolved optical spectrophotometry. We identify 5 real pulsation modes and at least 6 combination modes at frequencies consistent with those found in previous observations. The measured Doppler shifts from our spectra show variations with amplitudes of up to 5.5  at the frequencies inferred from the flux variations. We conclude that these are variations in the line-of-sight velocities associated with the pulsational motion. We use the observed flux and velocity amplitudes and phases to test theoretical predictions within the convective driving framework, and compare these with similar observations of the hydrogen atmosphere white dwarf pulsators (DAVs). The wavelength dependence of the fractional pulsation amplitudes (chromatic amplitudes) allows us to conclude that all five real modes share the same spherical degree, most likely, . This is consistent with previous identifications based solely on photometry. We find that a high signal-to-noise mean spectrum on its own is not enough to determine the atmospheric parameters and that there are small but significant discrepancies between the observations and model atmospheres. The source of these remains to be identified. While we infer T_eff  kK and from the mean spectrum, the chromatic amplitudes, which are a measure of the derivative of the flux with respect to the temperature, unambiguously favour a higher effective temperature, 27 kK, which is more in line with independent determinations from ultra-violet spectra.