Flaring and periodic variations in the optical photometry of Swift J1910.2-0546
ATel #4246; Christopher Lloyd (Sussex), Arto Oksanen (Caisey Harlingten Observatory), Peter Starr (Warrumbungle Observatory), Graham Darlington (BAA VSS), Roger Pickard (BAA VSS)
on 9 Jul 2012; 19:57 UT
Credential Certification: Christopher Lloyd (c.lloyd@sussex.ac.uk)
Subjects: Optical, X-ray, Binary, Black Hole, Neutron Star, Transient
Referred to by ATel #: 4347
We report optical time-series photometry of the optical counterpart of the recently discovered high-energy transient Swift J1910.2-0546/MAXI J1910-057 (Krimm et al., ATel #4139, Usui et al., ATel #4140) obtained with 0.35-m to 0.5-m class telescopes in Australia, Chile and the UK. Five other observations were taken with the Faulkes Telescope South with an r' filter on 2012 June 14. Subsequent time-series observations were obtained on 2012 June 27, 28, 29, July 1, 6 and 7 with individual runs or data sets lasting for between 2 and 7 hours. These were unfiltered and reduced as V relative to stars of the preliminary BAA comparison sequence (R. Pickard and R. Miles, http://www.britastro.org/vss/swiftmaxi_xray_transient.htm ).
The mean magnitudes of the data segments are...
2012 Jun 14 2456092.14 r' = 15.49(1)
2012 Jun 27 2456105.87 (V) = 15.76(1)
2012 Jun 28 2456106.78 (V) = 15.80(2)
2012 Jun 28 2456107.48 (V) = 15.98(1)
2012 Jun 29 2456107.81 (V) = 15.96(2)
2012 Jul 01 2456110.02 (V) = 15.94(1)
2012 Jul 06 2456114.80 (V) = 15.85(1)
2012 Jul 06 2456115.05 (V) = 15.96(1)
2012 Jul 06 2456115.21 (V) = 16.05(1)
2012 Jul 06 2456115.25 (V) = 15.85(18)
2012 Jul 07 2456116.04 (V) = 15.99(1)
The optical counterpart was discovered on 2012 June 1 and initial magnitudes of g' = 16.0(1) and r' = 15.7(1) (Rau et al., ATel #4144), UVOT b = 16.14(4) (Kennea et al., ATel #4145) and R = 15.9 (Cenko and Ofek, ATel #4146) were reported. On June 4 the UVOT magnitude was essentially unchanged at b = 16.19(4) (Kennea et al., ATel #4149) although Britt et al. (ATel #4195) found a brightening of 0.1 magnitudes between June 1 and 4 from r' = 15.50(1) on June 1 to r' = 15.42(1) on June 3 and r' = 15.40(1) on June 4. They also reported flickering of up to 0.1 magnitudes but no periodic variation.
Our observation on June 14 is only ~ 0.1 magnitudes fainter than the maximum reported 10 days earlier. Over the 11 days from June 27 and July 7 the mean magnitude faded from (V) = 15.83 to 15.98, but with excursions of up to 0.3 magnitudes. On two occasions, June 28 and July 6, it faded by 0.2 magnitudes in less than one day but subsequently recovered by at least 0.1 magnitudes.
On July 6 it brightened by 0.4 magnitudes from (V) ~ 16.05 over 20 minutes but the decline was cut off by the end of observations. This flare has been tested against several check stars and is not an instrumental or calibration anomaly. The following day the magnitude had stabilised at (V) = 16.0, about 0.05 magnitudes brighter than the pre-flare level. Another rapid change occurred on June 28 when it brightened by 0.1 magnitudes from (V) ~ 16.1 over 1.7 hours, and while this was not as extreme as the flare it is clearly different from the other changes seen in the light curve.
All the other time-series observations show significant short time-scale variations of 0.05 to 0.1 magnitudes and generally appear cyclical. A Lomb-Scargle periodogram of the detrended data shows a clear series of spikes centred near f = 10.75 cycles/day with a weaker series around 4 cycles/day. The primary feature has two components which when fitted with a least squares Fourier series resolve to periods of 0.09353(2) and 0.09245(2) days with a full amplitude of 0.03 magnitudes. Both fit the data equally well but the shorter period is more nearly sinusoidal. The dominant alias has two other periods at 0.10155(3) and 0.10276(3) days which also fit the data well but are marginally less significant and have asymmetric light curves.
The time scale of these periods and the coherence of the variation argues against QPOs and may be evidence of an orbital variation. Most of the scatter in the observations can be accounted for by the periodic variation and observational errors so there is little space for VLFN in the optical and there is no clear indication of flickering.