Swift Detection of a Superflare from DG CVn

ATel #6121; S. Drake (CRESST/USRA/GSFC), R. Osten (STScI), K. L. Page (U. Leicester), J. A. Kennea (PSU), S. R. Oates (IAA-CSIC, UCL-MSSL), H. Krimm (CRESST/USRA/GSFC) and N. Gehrels (NASA/GSFC)
on 5 May 2014; 18:45 UT
Credential Certification: Kim Page (kpa@star.le.ac.uk)

Tweet

The Swift team reports the detection of a superflare from one of the stars in the close visual (0.17“) dM4e+dM4e flare star binary system DG CVn (G 165-8AB). The Burst Alert Telescope (BAT) triggered on DG CVn at 2014-04-23T21:07:08UT = T0 (trigger 596958 reported in GCN Circ. 16158), resulting in an automatic slew to the source. The partial coding was 93%. The hard X-ray source had a peak intensity in the BAT 15-50 keV band of ~300 mCrab or 0.06 count/cm^2/s. The BAT data cover the period from T0-239 to 963 s. The mask-weighted lightcurve shows a single peak from ~T0-40 s to 120 s and another weaker peak from ~T0+200 to 240 s. The time-averaged spectrum from -29 to 337 s is well fit by either a simple power-law or a bremsstrahlung model. The power law index of the time-averaged spectrum for the former model is 2.62+/-0.33, while the temperature of the latter is 26(+12,-8) keV. The fluence in the 15-150 keV band is (8.4+/-1.5)e-7 erg/cm^2 for the power-law model.

When the Swift X-Ray Telescope (XRT) started observing at T0+117 s, the soft X-ray 0.3-10 keV rate of DG CVn was ~100 count/s, corresponding to 5e-9 erg/cm^2/s, and then decayed moderately, reaching a count rate of ~50 count/s by ~328 s after the trigger. After a 4.2 ks gap in XRT observations of this field, the soft X-ray emission had declined to a level of 4-15 count/s, but, after a further gap, at T0+11 ks DG CVn was observed to have had a second, smaller flare back to a level of ~30 count/s. The source then decayed monotonically for ~10 days, with a power-law of alpha = 1.39+/-0.01 fitting the data after T0+10 ks, ignoring a third, much smaller flare (peak rate 0.7 count/s) which occurred at T0+460 ks. By the end of this period, the count rate had declined to ~0.07 count/s (swift.ac.uk/DGCVn.gif) or a soft X-ray flux of DG CVn of 3e-12 erg/cm^2/s, similar to the levels of previous detections of this source by ROSAT and XMM-Newton. Preliminary spectral analysis of the XRT data from T0+120 to 600 ks using a 3T APEC fit yields a temperature of 27(+6,-8) keV for the dominant high-T component in the first observation (T0+117 to T0+328 s), in agreement with the initial BAT measurement. The derived temperatures for this component decline rapidly with time to ~4 keV during (T0+)4-20 ks, 2.7 keV for 20-50 ks and 2.3 keV for 60-200 ks, except during the 2 later re-flares when they showed small increases.

The Swift UV/Optical Telescope (UVOT) began settled observations of the field of DG CVn at T0+125 s. The initial finding chart images in the white and u filters are heavily saturated. After the 4.2 ks gap in XRT/UVOT observations, UVOT observed in all 7 UVOT filters, with the optical filters (v, b and u) being saturated until ~T0+20 ks. Several more smaller flares were observed after the initial trigger, with the second and third brightest occurring at T0+10 ks and T0+80 ks after the trigger. The flares have decreased in peak brightness as the overall brightness has decreased. The temporal evolution is most pronounced in the UV filters.

The peak XRT flux corresponds to an X-ray luminosity of 1.9e32 erg/s at the 18 pc distance of this system (recently derived by Riedel et al. 2014, AJ, 147, 85) which is 1.5 times the combined systemic bolometric luminosity of 1.3e32 erg/s. Thus, like the 2008 EV Lac superflare detected by Swift (Osten et al. 2008, ATel #1499; 2010, ApJ, 721, 785), for a period of a few minutes the X-ray emission from this flare outshone all the light from its parent star. As discussed by Riedel et al. (2014), this binary system has kinematic, rotational and activity characteristics indicative of membership in the young star population scattered throughout the solar neighborhood rather than to the dominant Gyr-old thick disk population. These authors suggest a likely age of 30 Myr for DG CVn. Thus, the extreme activity of this system as manifested by this event is due to its extreme youth.