Once we have identified a possible planet candidate from the NGTS data, the first thing we do is check to see if the transit signal is also visible in the Transiting Exoplanet Survey Satellite (TESS) data. Let’s take an example of one of the first objects we thought of as a planet candidate, Subject 69693101. The Planet Hunters NGTS light curve for this object looks promising, as do the light curves for the odd/even eclipses and secondary eclipses (not shown). Analysis of the NGTS data has estimated the radius of the orbiter to be 1.35 times the radius of Jupiter, which is large but not large enough to immediately rule it out.
So I decided to look at the TESS data for this star. Luckily, TESS has observed this star in two “sectors”, giving me about two months’ worth of data to look at (TESS observes an area of the sky, called a sector, for one month before moving on to the next, and is now revisiting the area to gather more data). The graph below shows the light curves for both sectors, with the x-axis showing the number of days since a fixed date (astronomers prefer to use a unit called Julian days, which is the number of days since January 1, 4713 BC, but this number is huge, so to make the numbers more manageable we subtract a standard value; in this case, we subtract 2.457 million days).
At first glance, the star appears to be highly variable. This could be due to intrinsic variability in the star, but such a periodic signal could be due to it being a binary system. The first step is to try to phase-fold the data onto the period detected by NGTS. This is the process of taking each section of the light curve that corresponds to what we think is the orbit of one of the candidates around the transit, and plotting the data on top of each other – that is, the transits should all be in the middle of the plot. The plot below shows the data folded onto a 2.51 day orbit detected by NGTS, with the grey dots showing the raw data and the red dots showing the data binned at 10 minute averages.
While this looks like an obvious transit signal, we were cautious when looking at it due to the star’s obvious variability. So we folded the data back to twice the detected period, and two transits at different depths were clearly visible, indicating that this is indeed an eclipsing binary system, and not a planetary candidate. While it is unfortunate that this particular candidate is not a planet, it does mean that we can focus our time and resources on more promising candidates that may actually be planets.
