HSTAND OPTICAL DATA REVEAL WHITE DWARF COOLING, SPIN, AND PERIODICITIES IN GW LIBRAE 3-4 YEARS AFTER OUTBURST

Abstract

Since the large amplitude 2007 outburst which heated its accreting, pulsating white dwarf, the dwarf nova system GW Librae has been cooling to its quiescent temperature. Our Hubble Space Telescope ultraviolet spectra combined with ground-based optical coverage during the third and fourth year after outburst show that the fluxes and temperatures are still higher than quiescence (T = 19,700 K and 17,300 K versus 16,000 K pre-outburst for a log g = 8.7 and d = 100 pc). The K wd of 7.6 ± 0.8 km s −1 determined from the C i λ1463 absorption line, as well as the gravitational redshift implies a white dwarf mass of 0.79 ± 0.08 M . The widths of the UV lines imply a white dwarf rotation velocity v sin i of 40 km s −1 and a spin period of 209 s (for an inclination of 11 deg and a white dwarf radius of 7 × 10 8 cm). Light curves produced from the UV spectra in both years show a prominent multiplet near 290 s, with higher amplitude in the UV compared to the optical, and increased amplitude in 2011 versus 2010. As the presence of this set of periods is intermittent in the optical on weekly timescales, it is unclear how this relates to the non-radial pulsations evident during quiescence.

GW Librae: a unique laboratory for pulsations in an accreting white dwarf

“…HST ultraviolet observations in 2010 revealed multiple closely spaced periodicities, but with the highest peak at 292 s. The signal was also detected in 2011 with HST near 293 s, with an amplitude more than twice as large as in 2010 (Szkody et al 2012). If this signal is the return of the ∼237 s period after outburst, it demonstrates the unique opportunity that DN outbursts offer to study the evolution of the pulsation spectrum as a function of white dwarf cooling.…”

“…While in 2002, 2010, and 2011 the system showed periodic variations on time-scales of ∼230-650 s, interpreted as non-radial white dwarf pulsations (Szkody et al 2002(Szkody et al , 2012, top and middle panels in Fig. 2), the new observations in 2013 reveal an intriguing and puzzling feature: the light curve is dominated by a large-amplitude variability spanning the entire HST visit.…”

“…Additionally, we retrieved from the MAST archive all Space Telescope Imaging Spectrograph (STIS) and COS ultraviolet data taken with the G140L grating, and using the time-tag mode: four STIS orbits taken prior to the 2007 outburst (for more details see Szkody et al 2002), and a total of three COS orbits post-outburst data (see Szkody et al 2012). Table 1 summarizes the observations and the time-averaged spectrum of each observation is shown in Fig.…”

“…We performed a bilinear interpolation of the grid in wavelength and effective temperature. We fixed log g at 8.35, which, using the white dwarf mass-radius relation derived from cooling models for white dwarfs (Holberg & Bergeron 2006;Kowalski & Saumon 2006;Bergeron et al 2011;Tremblay et al 2011), corresponds to previous estimates of the white dwarf mass in GW Lib based on the observed gravitational redshift (M WD = 0.84±0.02 M ; van Spaandonk et al 2010, M WD = 0.79±0.08 M ; Szkody et al 2012).…”

“…The stable pulsation signals reported by van Zyl et al (2000) were immediately swamped by the light from the accretion disc and an orbital superhump near 80 min became the dominant variability in the optical light curve (Kato, Maehara & Monard 2008;Bullock et al 2011). To date, several prominent signals have recurrently shown up (Copperwheat et al 2009;Schwieterman et al 2010;Bullock et al 2011;Vican et al 2011;Szkody et al 2012). Except for a signal near 290-300 s, all other signals have relatively long periods (∼19 min, ∼2 h, and ∼4 h).…”