Measurement of the light curve of exoplanet Qatar-1 b
Dr. Gerold Holtkamp, 2021-04-27
The Plan
A modern single-lens reflex camera, abbreviated DSLR (English for Digital Single-Lens Reflex) , is a powerful device. It can also be used in astrophotography - as long as the objects are bright enough. If the light intensity is low, the so-called noise becomes noticeable. If the recording chip of the DSLR were cooled (but it isn't!), this noise could be effectively reduced. For this reason, cameras specially developed for astrophotography are equipped with their own cooling of the recording chip. Unfortunately, my equipment (so far) only consists of DSLR cameras. The most powerful of these is the Canon 6D. With this I wanted to tackle a difficult task: detecting an exoplanet.
Exoplanets are planets orbiting a star other than our sun. But since all the stars and their possible planets are very far away from us - even those in our cosmic neighborhood - one cannot directly observe these exoplanets. In recent years, recordings have actually been made, but with an extremely high level of technical effort, which is currently unattainable for amateur astronomers. The most promising way for us to prove this is the so-called transit method. Direct pictures are not possible for amateur equipment. But there are stars whose planets pass right between the parent star and us observers on earth. So in this transit the star gets a little dimmer for us. "Dimmer" in this case means only a decrease in brightness of a few mmag. It is therefore a real challenge for a DSLR camera to detect this drop in brightness.
The Cassegrain telescope of the observatory of the natural scientific association Osnabrück has a mirror diameter of 60 cm and a focal length of 746 cm. Already at the beginning of January 2018 I had tried to prove exoplanet transits with this telescope and my Canon 6D. At that time, however, the telescope's original control software from 1992 was still installed for tracking. Modern autoguiding software such as PHD, which is responsible for fine tracking, could not be used. With the large focal length, however, it is absolutely necessary to have fine tracking, otherwise you will not get point-like stars. Accordingly, the first attempts failed.
After a renovation of the technology, autoguiding was possible from November 2019. However, it took until April 2021 for the present measurement to be successful after long periods of bad weather, the corona pandemic and a few failed attempts.
The Measurement
For April 23, 2021 [1] I had identified on the on the Swarthmore College internet page the transit of the exoplanet Qatar-1 b in front of its star as suitable. Swarthmore states a light reduction of 21.4 mmag.
Qatar-1 is a 12.843 mag star in the constellation Draco [2]. It is 0.8 times the size of our sun, 4800 K hot, 4.5 billion years old and 610 light years away. Its planet, Qatar-1 b, was discovered in 2010. It is about 1.2 times the size of Jupiter and orbits its parent star in 1.42 days at a distance of 3.5 million km. Therefore it has a surface temperature of approx. 1500 K and is referred to as a so-called "hot Jupiter".
Qatar-1 b was scheduled to step in front of Qatar-1 at 1:05 a.m. CEST, be in the middle at 1:55 a.m. CEST and step down again at 2:45 a.m. CEST. The 57% illuminated moon was 93° away, so it just couldn't shine into the telescope. Although it already brightened the sky, since it was a relative measurement, it would not affect the measurement apart from a slightly poorer signal-to-noise ratio. I started my measurements before the transit at 0:25 a.m. CEST so that I had a sufficiently large number of measurements of the brightness of the uncovered star. Measurements here means that individual images of the star field were made with and around Qatar-1. This field of view had a size of 16×10 arcmin² with the selected structure.
A total of 98 recordings were made between 12:25 a.m. and 3:43 a.m. 10 other shots could not be used because the stars were not sharp enough, which was probably caused by errors in the tracking. The exposure time was 105 s per picture with a sensitivity of ISO 3200. These recording parameters were chosen so that enough light from Qatar-1 falls on the light-sensitive pixels of the recording chip, but not too much, which would mean an overshoot and thus an invalid measurement. The correct exposure time was checked with the software IRIS:
In addition, 20 flatfield, darkfield and bias images each were created. The image processing and evaluation was carried out with the MuniWin 2.1 software, which enables the photometry of exposure series. Here, the brightness of the target star (Qatar-1) is compared with that of another star (Comparison Star).
It is important that both stars are of approximately the same spectral type. During the more than three-hour measurement, Qatar-1 moved from 40° to 58° in altitude. This means that the light paths through the earth's atmosphere are of different lengths, which would lead to a falsification of the light curve when comparing the brightness of stars with very different spectral types. The comparison star has the somewhat long designation 2MASS J20131440+6507000 [3]. With a color index of B-V = 0.983, it is similar to Qatar-1 with B-V = 1.06.
The Result
As a result, MuniWin outputs the above light curve as a diagram and as a data record. This data can also be uploaded to the Czech Astronomical Society in the so-called Exoplanet Transit Data Base (ETD). [4]It should be noted that the times are corrected to UTC. You get a light curve adjusted for possible trends and, in addition, a model-based fit [5]. In addition, the parameters of the time of the middle of the transit, its depth and its duration are determined from our own measurements. These values indicate properties of the exoplanet.
Values of my measurement (from ETD)
Middle of transit: 1:52 UTC +/- 0:02 min
Duration of transit: 88,0 +/- 6,0 min
Reduction of brightness: 25.3 +/- 2.7mmag
Literature Values (ETD, Swarthmore)
1:54 UTC, 1:55 UTC
96,7 min, 100 min
20,4 mmag, 21,4 mmag
Discussion
My readings are slightly outside of the ETD and Swarthmore College values listed above. At the midpoint of transit, there is agreement even within the errors. If you compare my values recorded with the DSLR Canon 6D with those of Astro colleague Thomas Grunge, which he created with a cooled CCD camera, it is noticeable that the signal from the DSLR is significantly noisier [6]. Obviously detection of exoplanets with a DSLR is a challenging task. But there is still room for improvement. The autoguiding was not yet optimal because only very weak guiding stars were available. Possibly the moon disturbed the measurements via reflections in the gap of the observatory dome. So it goes on.....
__________________________________________________________________________________________
[1] https://astro.swarthmore.edu/transits/transits.cgi
[5] Poddany S., Brat L., Pejcha O., New Astronomy 15 (2010), pp. 297-301,
Exoplanet Transit Database. Reduction and processing of the photometric data of exoplanet transits (arXiv:0909.2548v1)
