This week we are following up two posts from weeks 268 and 306. These were about the star that seemed to have two periods in close proximity to each other. These two periods were either near two days or near 16hr. The reason we could not really decide which pair of periods was correct was our typical cadence of one day for the observations.
To determine the correct periods (2d or 16hr), we ran a campaign to get a better sampled dataset early December last year. We seem to have now almost all those images uploaded and processed – a few are still missing and will be added prior to our final analysis. However, the current data are already sufficient to decide on the right set of periods.
In the above plot we display the R-band data for the last two observing seasons, including the campaign (visible as the denser group of points towards the end of the light curve). In the top right panel the light curve is shown. The magnitudes are plotted as difference from the median magnitude. One can still see the ‘interference’ pattern, or beating, which is caused by the two periods being close together.
The top left panel shows the periodogram of the data. The highest peaks are now much more clearly near two days. The aliases near 16hr have been reduced, and a zoom into the new high cadence data clearly rules out the 16hr periods as being correct. The most prominent period is 1.9243d, and the second period is about 2.0848d, which results in the beat frequency of 25d.
The bottom plot shows the phase folded data, if the most prominent period is used. We have taken some high resolution spectra of the object as well. Unfortunately the star is quite faint even for the 2m telescope used. However, there are no real indication that we are seeing a binary system. Thus, any explanation of why there are two periods in the photometry will need to involve only one star. Some thinking to do, while we wait for the rest of the data to be processed….