A Discussion with Ansgar Reiners on
Discovering Exoplanets

23rd February 2018

Dr. Ansgar Reiners is a researcher working on discovery of exoplanets using radial velocity. He was one of the co-authors of the paper that announced the discovery of Proxima Centauri b in June 2016, which has caused quite a lot of excitement in the astronomical community. Dr. Reiners has worked on various areas in astrophysics, including stellar astrophysics, in which he explored the characteristics of various types of stars, and on exoplanets around M-dwarfs. He is currently the managing director of the Institute of Astrophysics, University of Gottingen. He delivered a talk on his research on radial-velocity based search of exoplanets at the Indian Institute of Astrophysics, Bangalore on 23rd February 2018. Our junior researcher Suryoday Basak was able to catch up with him and ask him a few questions on the growing area of exoplanetary research.

Suryoday Basak: How were you drawn towards astronomy and astrophysics, and in exoplanets in particular? I have seen a some of your papers which are focused on computation, some which are focused on stars, and some on exoplanets. So how did you get here?

Ansgar Reiners: I really came from the physics side. I'm interested in physics. In general, of course, everyone is excited about the universe and astronomical objects and similar stuff, but that was not really my motivation to start with. In physics, I was doing stellar astrophysics, which I found exciting because you can learn a lot about physics by studying stars. There are a lot of big experiments set up in the sky: we just have to observe them. That's something I was fascinated by. I sort of moved in my career, the way down from hot stars, to sun-like stars, to cooler stars. Often, the research on stars is motivated by the question, "are we alone?" and "are there other planets out there?" So that sort of came together. We did the magnetic-field measurements in M-dwarfs, and I did a lot of research on spectroscopic investigations of low-mass stars. There I gained some experience in the spectroscopic work with M-stars.

Suryoday Basak: So presumably, because you were involved in research on low-mass stars, you came towards finding planets around them?

Ansgar Reiners: Yeah, at the time that I worked on those stars, they became more popular! They were completely unpopular for a long time. I worked on them together with some colleagues and we collected a lot of information, which was also at the time when they were becoming popular to look for exoplanets around them. We had the experience of how they are looking spectroscopically and this is what's needed to find these exoplanets. So this came together, and it is not very difficult to be drawn into this business because it's really exciting.

Suryoday Basak: Could you tell us about the main challenges faced in the pipeline? I'm not enquiring only about the instrumentation aspect of it, because, sure, there's a really big technological challenge and the general opinion is that we're all improving, but tell us more about handling the machinery and getting the computation done.

Ansgar Reiners: All the observations are done at the observatory. There's a quick-look pipeline at the mountain, which is similar to the pipeline we have, but it's just for a quick look by the observers. So there are actually staff observers in Spain. The data is transferred to our institute and we have all the data on a local machine. Then there's a pipeline reducing the data. There's one person looking at the pipeline every day but it is running automatically. So we're observing essentially every night. This cannot be done by hand. If that person is on vacation, you're lost, and it's just way too boring for him. With lots of data, you also need to have a consistency with your data analysis, and this is sort of a 'machine' thing. So this runs automatically and there's a person looking at it, just checking whether everything is great. When it works out, it's fine, we store the data.

Pretty much every two months, we do a re-analysis of the radial velocities. The analysis is a two-step process. First, you have these 2D spectra and from that, you have to make a 1D spectra: basically representing the magnitude as a time series. So you have these spectra - that's the first step, and then from that, you have to have these radial velocities based on the Doppler shift: that's the second part of the pipeline. All this needs to be done consistently for the entire dataset every two months.

Then the data goes into archives. People download the data and look at it. The team looks at the individual objects and if we find something interesting going on, it goes in a list. We then assign people to individual targets who then look at them in detail, and find out what is going on, and whether there's really an exoplanet.

Suryoday Basak: After having gained some insights from your own research on how exoplanets are, can you comment on their characterization? I know that at this point in time that's a dicey question but I'd like to know how we can reasonably characterize or group a lot of exoplanets from different solar systems into categories.

Ansgar Reiners: There are generally two different approaches. One is that you understand the planets as a group, which means that you look at the entire crowd of planets that are known and you do statistical analysis. You ask the question of how many planets of a certain mass exist around a star of a certain mass; add a distance, eccentricity and all the parameters that you have. By now we have more than a few thousand planets so this already gives nice answers! There's also a lot of stuff coming out about formation, evolution and all these kinds of things.

That's one side. It doesn't tell you a lot about the planet itself, it just gives you some information about the parent stars and their existence. About the exoplanets themselves: there are mainly the spectroscopy-based approaches. So if you have a transiting planet, then from it you can learn more. From this, you can get the radius and the atmospheric characterization. And that is one of the reasons why these missions are so successful. We've had Kepler and Corot, and we'll have CHEOPS, TESS, and Plato in the future. These are all satellite missions and there's a huge investment in this high number of missions for this one particular target. All this is for finding transiting planets because that's the way towards characterizing them.

Suryoday Basak: I suppose that I have one last question for you. What you see to as future of this field, maybe five to ten years down the line?

Ansgar Reiners: Well, we will have a lot more planets known. The whole picture will be much better drawn, much better painted.

Suryoday Basak: And I suppose that at that point of time a very robust characterization scheme would be necessary?

Ansgar Reiners: Yeah, the characterization is one thing. Though, the study of exoplanet atmospheres is also becoming strong; there are missions for exoplanet atmospheres and all the elements that you can find. This is all super exciting: to know what they do, and whether the atmospheres are expanding and how they react to different environments! All this will be figured out.

Two specific questions that I find exciting that will be addressed in the future are, first, there's a prediction from planet-formation models that around one solar mass stars there's a maximum number of planets. We've found a lot of Jupiters, we know that M-dwarfs don't have hot Jupiters, but there is a lot of M-dwarfs, and they have a lot of Earth-like planets. That's what we know. And they are `supposed' to have a lot of those! We hope to find many of those planets with different missions and see if this prediction is actually true regarding planet formation. The second thing is that many radial-velocity surveys are now running long enough to find sizable samples of planets that are similar to the planets in our solar system and ask the question whether the planets in our solar system are some weirdos or just planets like all the others. There are some people that say that our system is really an exception. I'm not sure I buy that because there's a huge bias in the entire idea and in all the observations that we have done. If we go on observing for a few more years, maybe ten years or so, several surveys would have run long enough to find these Jupiters and Saturns. And if we find those, then why not look for copies of our solar system? And how many of such systems exist? This is a really important question.

Suryoday Basak with Dr. Ansgar Reiners