The Changing Light Curve of [MA93]1640 (XMMU J010743.1-715953)

ATel #15159; P. C. Schmidtke (Arizona State University), A. P. Cowley (Arizona State University), A. Udalski (Warsaw University Observatory)
on 11 Jan 2022; 16:45 UT
Credential Certification: Paul Schmidtke (Paul.Schmidtke@asu.edu)

Tweet

Emission-line star [MA93]1640 is the optical counterpart of XMMU J010743.1-715953, an X-ray source that shows evidence of weak pulsations with P=153 s and has a hard spectrum typical of Be/X-ray binaries (Coe et al., 2012, MNRAS, 424, 282). The visible star is classified as type B2IV-Ve, with a variable Halpha line profile. The OGLE-III light curve has long-term brightness changes (I=16.95 to 16.45) that are superimposed by numerous sharp dips and spikes. These rapid fluctuations are most prominent when the source is faint. Coe et al. identified a possible periodicity at P=100.3 d. A light curve folded on this period (see Fig. 15 of Coe et al.) reveals a narrow dip that precedes a rapid rise to maximum light.

We have analyzed 10 years (2010-2020) of OGLE-IV observations, which are shown in the top panel of the accompanying figure. The pattern of dips/spikes and the general brightening present at the end of OGLE-III continued during OGLE-IV seasons 1-2, with the star reaching I=15.8 near JD 2456100. Thereafter, the character of these variations changed substantially. As the overall system brightness declined to I=16.5 during seasons 3-5 and then rose in seasons 6-10, the fluctuations were replaced with lower-amplitude variations. To facilitate the study of this change in optical behavior, long-term trends in the observations were removed by subtracting a second-order polynomial from each season of OGLE-IV data. We refer to these detrended values as I* magnitudes.

PDM analysis of I* data from seasons 1 and 2 of OGLE-IV reveals a signal at P=103.1+/-0.5 d, close to the period reported for OGLE-III observations. A plot of the folded light curve is shown in the figure, where phase zero corresponds to maximum light. Comparison of this plot with that presented by Coe et al. shows the dip that precedes maximum light has become shallower and broader. Furthermore, the peak, which is less pronounced in OGLE-IV data than in OGLE-III, is followed by large scatter at phase 0.2. The highest points near this phase were taken JD 2455445-2455459 and are readily visible in the raw light curve. The presence of this secondary brightening is difficult to reconcile with an orbital interpretation of the 103-d periodicity.

To study the character of variations after JD 2456100, we calculated a periodogram of I* data from OGLE-IV seasons 3-10. As shown in the accompanying figure, the highest peak occurs at P=51.44 d, but aliases close to 0.5 and 1.0 d have nearly the same power. When folded on each trial period, the light curve is a low-amplitude sinusoid. We note the 51.44-d period is approximately half that of the 100-103 d periodicity present when the star was more active. Evidence for an orbital signature in seasons 3-10 of OGLE-IV data is weak, and the observed signal is most likely caused by nonradial pulsations (NRPs) of the star. The light curves of the 0.5 and 1.0-d aliases have shapes and amplitudes that are consistent with this interpretation, but we cannot distinguish between them using the available observations. As an example, the OGLE-IV I* light curve folded on the best-fit sine wave at P=1.0170 d is shown in the bottom panel of the figure. The apparent strength of the 51-d signal in the periodogram is likely an artifact of data sampling, detrending errors, changes in NRP amplitude, or variations in period of the NRPs. Schmidtke et al. ( 2013, MNRAS, 431, 252) demonstrated that small changes in NRP period have the effect of shifting power to a long-period alias.

Photometry of [MA93]1640