"Exoplanets and differential photometry"
 realized with graduate students of SupOptique
(in connection with our observatory)

 

The exoplanets are planets which orbit around other stars than the Sun. To detect them directly, it is necessary to be able to solve the 2 major following difficulties.
First, the angular separation between the star and its planet is very weak (< 1 arcsec).
Second, the contrast in brightness between the star and its planet is huge (109 in the visible and 106 in the thermal infrared, near 10 mm). The detector is thus dazzled by the star.

On the other hand, there are several methods of indirect detection .
One of them is based on the Doppler effect . The star and its planet turn around the common centre of gravity and the light of the star undergoes a periodic Doppler effect. Most of the known exoplanets were discovered by using this effect.
Another method, which is used here in this project, is based on the transit effect. During its periodic motion, the planet can transit between its star and the Earth. This leads to a periodic variation of the flux and thus of the magnitude of the star.

The star which is the subject of this work is HD209458 . It is located in the constellation Pegasus, at 150 light years away, and its spectral type is very close to that of our Sun. A few years ago, astronomers discovered by Doppler effect a Jupiter-type planet in its vicinity. In 1999, this exoplanet was rediscovered by the method of the transit : every three days and a half, this planet eclipses its star during approximately three hours, inducing a fall of its flux close to 1.6 % !!!

The goal of this project is to get the transit curve of this couple. We thus need to carry out extremely precise photometric measurements, that, to date, only few amateurs in the world succeeded in (see the site of the network transitsearch for more information).

After many attempts, we could finally image HD209458 on  december 1, 2004.
The predicted transit epochs (UT) were as follows : begin at 16h09, center at 17h50, end at 19h30.
The Sunset was at 16h00, therefore the twilight prevented us to see the beginning of the transit. We thus decided to concentrate on the end of the transit, by taking 70 images between 18h33 and 19h39 (UT).
Here is an example of preprocessed image of HD209458, surrounded by 3 comparison stars :


For all the images : C14, ST8, exposure time = 20 s, with green filter, total field : 12 X 8 ', very bad seeing.


HD209458 has a magnitude of 7.65. The comparison stars have magnitudes ranging between 11.2 and 11.6.

The photometric processing of the 70 images was made with IRIS . We used 2 methods : aperture photometry and the function called PSF, the latter having given us the best results.

Here is the curve of the relative variation of flux of HD209458, versus time :


HD209458b : End of transit (just after the twilight). The first exoplanet detected from our observatory.
Data from HST are drawn from the curve below.
The model is based on a small black disc which passes in front of the star. The latter is supposed to have a uniform brightness.

Please note that we plotted raw data (no averaging, no filtering).
It seems that the end of the transit is well detected.
The measured transit depth is approximately 1.6 %, close to the value measured by HST :


Transit of HD209458b seen by HST.


It would have been preferable to measure all the transit, like Hubble.
Nevertheless, this is a first step for us, even if it is really hard to measure a relative variation of flux smaller than 2 %, on an 8 mag star.
We learned a lot doing this first transit and we have already some ideas to get better results.
See you in a few weeks with TrES-1...Stay tuned !

To join us :
thierry dot lepine at institutoptique dot fr (anti-spam)