Last February, the European Southern Observatory (ESO) announced the discovery of a small exoplanet around the nearest star to the Solar System, Proxima Centauri. This discovery was made possible by ESPRESSO, an instrument installed a few years ago on the Very Large Telescope in Chile.
The team that made this announcement includes Romain Allart, a Trottier postdoctoral researcher at iREx. Here he answers our questions about ESPRESSO, this newly discovered exoplanet around Proxima Centauri, and his contribution, which consisted in developing a new method for correcting the data obtained with the instrument.
iREx: What’s special about the newly-found exoplanet around Proxima Centauri?
Romain: Proxima Centauri is the closest star to our Solar System. Other planets were already known around this star, but the one announced by our team, called Proxima Centauri d, is one of the smallest exoplanets ever discovered. Its mass could be as low as a quarter of the mass of the Earth, or about twice that of Mars. We suspected that these small planets existed, but it is a real technical feat to detect them, so we know very little!
This planet is about 35 times closer to its star than the Earth is to the Sun, and it goes around it every 5 days or so. It is too close to its host to be in the so-called habitable zone: it receives too much energy to be able to keep liquid water on its surface. It is therefore a very different planet from ours!
iREx: Tell us about ESPRESSO. Why is this instrument so effective in detecting exoplanets?
Romain: ESPRESSO, for Echelle SPectrograph for Rocky Exoplanets and Stable Spectroscopic Observations, is an instrument installed in 2018 on the Very Large Telescope in Paranal, Chile. The VLT has four telescopes with mirrors 8.2 meters in diameter. ESPRESSO can harness light from any of these four large telescopes or all four combined, allowing it to achieve unique performance.
Like its predecessor HARPS, also installed in Chile but at a different site, ESPRESSO is a spectrograph that uses the radial velocity method (or velocimetry). This method involves detecting the small periodic movements of a star generated by the presence of a planet in orbit around it. The less massive a planet is, the less important this oscillation is and the less the radial velocity (in our direction), which can be measured with a spectrograph, varies. To be able to identify a planet as little massive as Proxima Centauri d, we must have an instrument that can detect a variation in speed of only 40 centimeters per second (1.44 kilometers per hour), which is remarkable. We hope that in the future, ESPRESSO will be able to detect even smaller variations, of the order of 10 centimeters per second!
iREx: What is your contribution to this discovery?
Romain: I developed a method to improve the results obtained with ESPRESSO and other similar spectrographs. In order to reach an accuracy of 10 centimeters per second, it is necessary to have the ability to use all the information available by reducing the sources of contamination to a maximum. This method allows to reduce the contamination generated by the atmosphere of the Earth. It can be applied to the data obtained for Proxima Centauri by ESPRESSO.
iREx: Could you summarize the correction method you have developed?
Romain: It is while I was completing my PhD thesis in Geneva, Switzerland that I developed this method, which I have just presented in detail in a scientific paper to appear in the journal Astronomy & Astrophysics.
Our goal was to develop a simple, automatic method with few parameters to adjust to remove the signature of the Earth’s atmosphere in data from spectrographs like ESPRESSO. To do so, we built a simplified model of the Earth’s atmosphere by combining its physical properties with the meteorological conditions during observations. This model can then be fitted to and removed from the spectra observed by the ESPRESSO instrument.
By removing a large portion of the signal from our planet’s atmosphere, our method allows us to obtain more information on the variation of the radial velocity of the star, and thus, on the possible presence of planets around it. It can be used in future analyses with ESPRESSO and other similar instruments.
iREx: What work did you have to do to get these results? What would it look like if we saw you working on this?
Romain: This work, as is often the case in astronomy, consists of writing an algorithm in the Python programming language, a language that is widely used in the industry as well. If you saw me working, you would just see someone typing lines of code into a computer! It may seem a little boring or repulsive at first, but Python is very easy to use and there is a lot of help online. I like to think that this tool allows me to harness the power of modern computers to better understand the universe!
iREx: Why are you interested in this kind of project? Why should the public be interested?
Romain: This type of project, more technical, is essential for the good functioning of an astronomical instrument and are the necessary gears for great scientific discoveries. The correction method that we have developed will be included in the ESPRESSO data reduction and will be offered to the scientific community. It will improve the quality of all studies made with ESPRESSO, which is very motivating!
iREx: What are the next steps?
Romain: The results of my paper show a clear improvement in the quality of the results for Proxima Centauri. Combined with other analysis methods, we are confident that it will confirm the existence of the Proxima Centauri d. This kind of independent confirmation is a crucial step in any scientific discovery.
We are also adapting our method for another instrument to which iREx contributes, the NIRPS spectrograph. This instrument will have the same role as ESPRESSO but will observe in the near infrared, alongside its big brother HARPS, installed on the 3.6m telescope in La Silla, Chile.
So there is still a lot of work to do!
Note: This interview has been edited for clarity.
Link to the ESO Press Release on the discovery of Proxima Centauri d.