It’s already September and the summer internships at iREx have come to an end. After having presented their main results to their iREx colleagues and a few guests on August 24, it’s time for our 10 interns to take a look back on their summer before returning to their undergraduate studies or embarking on their graduate adventures!
Before leaving us, Luc Bazinet, Laurie Dauplaise, Emina Hafiz, Pierrot Lamontagne, Arnaud R. Larochelle, Érika Le Bourdais, Daniella Morrone, Philip Richard, Salma Salhi, and Vincent Savignac agreed to answer a few questions.
Intern from the University of Ottawa who worked with Björn Benneke at the Université de Montréal
My internship was to analyse data of the exoplanet WASP-121 b captured by the CRIRES+ instrument. With this data, it is possible to detect which molecules are present in WASP-121b’s atmosphere and what their abundances are. It’s also possible to find the temperature of the atmosphere at different altitudes.
Finding the abundance of molecules in an exoplanet’s atmosphere tells us a lot about the formation and evolution of the exoplanet. Moreover, CRIRES+ is a new state-of-the-art instrument. The analyses I have done are among the first in the world with data from this instrument!
The analysis allowed me to confirm the presence of carbon monoxide and water in the atmosphere of WASP-121b. I also determined that this exoplanet has a stratosphere (a layer of the atmosphere that increases in temperature with altitude).
I learned a lot about analysing telescope data to extract useful information. Observational data is never “clean” when received, there is a lot of work to get it into a state where it is possible to draw meaningful conclusions.
My biggest challenge was taking my time. Looking at the data for the first time, I had promising results, but the data needed to be cleaned up to get a better result. This cleaning was a long process filled with trial and error.
I enjoyed being surrounded by amazing people. The professors and graduate students that I have been around are very intelligent people who know a lot about exoplanets. They were always willing to help me when I was having trouble. Also, I made good friends among the summer interns.
Trottier Intern from the Université Laval who worked with Björn Benneke at the Université de Montréal
My internship focused on the improvement of the SCARLET code, developed by Prof. Björn Benneke, which allows to generate models of exoplanet atmospheres. These models are then compared with observations. I modified the way the thermal emission spectrum of these planets were calculated. I also added an absorption source due to collisions between molecules.
It is always interesting to develop models that better reflect the real physical and chemical processes that occur in the atmospheres of exoplanets! It is by comparing models and observations that we can know what exoplanets are made of. So what better way to understand exoplanets than to improve the models?
I discovered that, in the calculation of an exoplanet’s emission spectrum, it’s important to consider the thermal emission coming from the layers of the atmosphere at the periphery of the planet. This is a detail that is not often taken into account, but it causes important discrepancies with the spectra that we can observe during secondary eclipses, i.e. when exoplanets pass behind their host star.
I learned a lot about exoplanets in general: how they are detected, how their composition can be determined. Since I had the chance to work with an exoplanet atmosphere simulation code, I discovered the complexity of simulating the multitude of physical and chemical phenomena that occur in a planet’s atmosphere, which we totally take for granted in our daily lives! Working with this code also allowed me to improve my programming skills.
The biggest challenge was to add the source of absorption from collisions of some molecules. The code had to read data files available on the internet, but the structure of each file was totally different. It was difficult to adapt the code so that it could read all the structures corresponding to each molecule without errors.
I loved being able to live the life of an astrophysics researcher for a summer! We had the chance to attend weekly iREx Cafés, to do training courses at the Mont-Mégantic Observatory, and even to participate in a celebration event when the first images taken by the James Webb Space Telescope were released. This summer has been memorable, equally for all the new knowledge I have acquired about exoplanets than for the wonderful encounters I had!
Trottier Intern from the University of Calgary who worked with Jason Rowe at Bishop’s University
Solving the N-body problem for the eight planet compact system of KOI 2433 by implementing Jacobi coordinates using the Julia programming language.
It was interesting to learn how to use MCMC (Markov Chain Monte Carlo) to take estimates we have of certain planet characteristics such as mass and period to calculate a series of best fits for the data that allow us to recalculate those values more accurately. It’s also very interesting and exciting to know that these values at some point will be uploaded to the NASA Exoplanet Archive.
I found the masses, periods, centre of transit, and eccentricities for the planets in the KOI 2433 system and created a visual of how the orbits looked. This data included a newly discovered planet for this system. I believe the most important result was the orbits of the exoplanets because it provides a lot of information about how the exoplanets interact with one another, which in turn affects their individual characteristics.
I learned how to code in Julia and how to use MCMC. Before this internship, I only knew how to code in Python which is quite similar to Julia but isn’t as computationally powerful. In terms of statistics, I have not taken too many courses on them or related to them so it was very useful to be able to learn it through my project.
My biggest challenge was understanding the significance and purpose of the statistical analysis of the data since, as I mentioned before, my knowledge in statistics is still quite rudimentary. Thus, it sometimes took me longer than I had hoped to piece together the information and process the outputs of the code.
My favourite part of the internship was being able to create the graphs and tables of the results to really see how the data came together and understand the significance of it. I also really enjoyed being able to do the last month of the internship in person and, although it happened very late, I was very glad to meet all the other interns and people at iREx.
Trottier Intern from the Université de Montréal who worked with David Lafrenière at the Université de Montréal
My internship was divided between two topics. First, I worked on the simulator of the single object slitless spectroscopy (SOSS) mode of the NIRISS instrument on the James Webb Space Telescope. I added two distinct noise sources to the simulator: cosmic rays and field stars. I also implemented the possibility of simulating secondary eclipses with the simulator. I also analysed data from the Dragonfly telescopes in order to obtain transit signals that could potentially confirm TOIs (Tess Objects of Interest).
The most interesting part about my internship is that I got to work with very complex and impressive algorithms and add features to them. It felt like working on a video game!
My favorite result during my internship had to do with the implementation of the field stars in the SOSS simulator. In order to confirm that the algorithm adding field stars worked properly, I had to compare the simulation’s result with actual JWST data. When I did, I found out that the two images matched almost perfectly. This was really satisfying!
I learned a lot of coding and software skills that I will use all my professional life. It was also my first experience in a research group, and I got to learn how to plan my working time and how to share ideas. I also learned a lot about the details of data analysis and reduction in astronomy. I also had the chance to give a public talk at the ASTROLab du Mont-Mégantic which was very instructive.
My biggest challenge was to keep my motivation up when a problem persisted for a long time. In these situations, you have to change approaches often and give up some routes on which you’ve been working on for a long time. It can be hard to keep coming up with new ideas, but when the holy grail arrives, it’s very satisfying!
The fact that I got to work with the greatest telescope ever created blows my mind! It is a dream come true. I also got to work with great scientists that I’ve always admired. I loved the creative process of finding intricate solutions to problems and I’ve learned so many skills from it. Big thanks to David Lafrenière, Loïc Albert, and Chris Mann for making my experience unforgettable!
Intern from the Université de Montréal who worked with Jonathan Gagné at the Université de Montréal and the Planétarium Rio Tinto Alcan d’Espace pour la vie
My topic was the search for planemos (planetary-mass objects) in the Volans-Carina Association, which is a young moving group with an age of approximately 89 million years. Its newly discovered corona spreads much closer to us and thus allows for the survey of objects that are not as bright as stars.
The corona of this association had just been discovered when I started working on it, which made it all the more exciting.
I have two promising candidates that come from a catalogue of over 330 000 entries. To confirm their membership to the association, we would need further telescope observations.
A lot! Although I have always been interested in astrophysics, it was my first academic/professional experience in this field. I learned a lot about brown dwarfs, moving groups, and the numerous techniques used to discard field objects, but also how much the scientific community works together.
My lack of experience in the field made me feel a bit overwhelmed at times, but I am so glad now that I feel I gained a lot of knowledge and skill.
The feeling I was working on something new. I was searching something no one had searched before, and to me it was very inspiring.
Sureau Intern from the Université de Montréal who worked with Nathalie Ouellette at the Université de Montréal
Education and Public Outreach
I had the opportunity to work on various projects related to public outreach during events, writing for iREx’s website and social medias as well as video production!
I discovered a lot of ways to do public outreach and how different initiatives allows our research to shine in various environments!
I mostly learned about the difference techniques used to simplify science in a way that is comprehensible for the various target audiences we have to communicate to. I also acquired a lot of astronomy knowledge that will help me to better understand my research during my graduate studies.
My biggest challenge was probably that I had to learn about different aspects of exoplanets really quickly and other topics so that I could understand it well enough to properly answer the public’s questions during many events that I attended as a host.
I adored my collaboration with iREx’s communication team! All of our tasks were complementary and it made it so easy to work! The work climate established by my supervisor made me feel really secure, and it was easy for me to get feedback almost instantly and to quickly improve.
Trottier Intern from the University of Toronto who worked with Jonathan Gagné at the Université de Montréal and the Planétarium Rio Tinto Alcan d’Espace pour la vie
My internship was focused on finding exoplanet candidates using astrometric data from the Gaia Data Release 3 (DR3). The goal of this work was to identify moving stars on the sky and filter through the parameters in Gaia DR3 to create a list of potential exoplanet host stars.
It was interesting to learn the significance of different parameters in Gaia DR3, how some implied there was likely a binary star system and others detailed the confidence of the measurements. Further, applying all these to the project goal made pinpointing exoplanet hosts in the data even more gratifying.
The result of this work was identifying 284 candidate exoplanet hosts in Gaia DR3 within 500 parsecs. Of these, 34 were identified as young stars and we have started the process of directly imaging these candidates.
This summer, I learned so much about exoplanet discovery and the research being done in this field. Coming from a different field, I started by learning the basics of exoplanet science (methods of detection and classification, etc.) and then refined my knowledge towards my project. Further, this was an incredible opportunity that allowed me to develop my coding and scientific communication skills!
The biggest challenge this summer was moving to and navigating Montreal as a non-French speaker. Although tedious at times, the language barrier was a wonderful opportunity to practice French, and everyone at UdeM and iREx was understanding and helpful throughout the summer!
While learning about exoplanets and working on my project were both wonderful aspects of my internship, my favourite part of this summer was meeting and getting to know the other interns and members of iREx. Everyone was so welcoming; it made me feel included in the community right from the very start.
Intern from McGill University who worked with Eve Lee at McGill University
This summer, I investigated how Super-Puff planets that are rapidly losing mass can destabilise the orbits of their system. I did this by modelling the planets’ atmospheres and then implementing the result into a simulation of the host system’s orbits using N-body integration in Rebound.
On their own, Super-Puffs in general are really interesting. For example, a good number of them have been around for billions of years, but current theories can’t seem to explain how they managed to survive this long and how they can still exist in their current state. My project attempted to explore one hypothesis that tries to explain this fact.
Though there is still a lot that remains to be learned, our simulations so far seem to suggest that the rapid mass loss experienced by the Super-Puffs we were investigating doesn’t destabilise the orbits of its system as we might have thought, which might be a promising insight into explaining why we still see some of these strange exoplanets today.
This summer, I learned a lot about fluid mechanics and what it can teach us about exoplanets and their atmosphere. Also, I learned plenty about interesting orbital mechanic phenomena such as Mean Motion Resonance, where planets “”synchronise”” and mutually stabilise each other’s orbits.
One downside of N-body simulations is that they can take a while to complete. This means that when experimenting with our models or modifying parts of it, we had to be careful when considering what to test and how, because each run could take up to a couple of hours (or even days!). Figuring out how to make things more efficient thus became a difficult yet important aspect of my project.
I really enjoyed learning about exoplanets and how they form. Also, I found the act of trying to answer questions to which no one knows the full answer quite exciting, and it was rewarding to make progress and figuring out ways to solve problems that we faced.
Trottier Intern from the University of Calgary who worked with René Doyon at the Université de Montréal
This summer, I worked as a member of Prof. René Doyon’s team analysing new JWST data as it came in. My project was entirely data-driven; I designed codes to examine anomalies in the data such as flux variations and stellar variability. I also did some white light curve Bayesian fits as a first step to getting a full transit spectrum.
It was such an honour to be one of the first people to access and analyse the JWST exoplanet data. I felt like I was making history as I was downloading the files. I really enjoyed conducting analyses and designing code that would uncover new results or avenues of investigation.
During commissioning, we discovered what we later coined “tilt events”, which is when the flux of stars we were observing jumped at some point in the time series. When we started working with TRAPPIST-1 data in July, we discovered that there was a lot of stellar and perhaps instrumental variability occurring out-of-transit. It is important to correct for these effects before we can produce a good transit spectrum, which is our ultimate goal!
Over the course of this summer, I learned how to efficiently design code, how to collaborate within a team on time-sensitive projects, and how the process of research works. I learned that there are often unexpected results and that it is important to be continuously adaptable to new changes.
The biggest challenge was the speed at which things were developing and new information was coming in. Data reduction codes would change every day, and I had to learn how to use them quickly, as well as to take into account new results that were coming in from the analyses of other people in the team.
I loved the unique opportunity to analyse the best new JWST data, and to do so as part of a collaborative research team. I was able to contribute to our project in a meaningful way by conducting modes of analysis that helped us uncover some interesting results. It was truly a revolutionary experience and I am incredibly grateful to have been given the chance to work on something so historically significant.
Trottier Intern from McGill University who worked with Eve Lee at McGill University
During the formation of mini-Neptunes, the gas envelope surrounding the rocky core gains mass as the core accretes particles from the surrounding protoplanetary disc made of gas and dust. Studies have described exchanges of gas particles between the disc and the envelope which we call “atmospheric recycling”. My goal was to include this process into simulations of the envelope formation in order to see whether it significantly slows down the accretion of gas from the disc.
The rate of the accretion of gas determines the final mass of the envelope when the disc dissipates. Studying this mechanism thus tells us whether we end up with a gas giant or a mini-Neptune. The problem is that, although we observe a large proportion of mini-Neptunes, current models predict that the accretion should be fast enough to produce gas giants. This is why we are considering different scenarios like atmospheric recycling to delay the formation of the envelope.
After successfully including recycling flows of gas into my model of the envelope formation, I found that they do delay the accretion of gas from the gas disk. Although the effect of this atmospheric recycling is significant when the planet forms at a short orbital distance from its star, it never delayed the system enough to successfully produce mini-Neptunes in my simulations. This suggests that including only recycling is not sufficient to make mini-Neptunes.
This summer was my first introduction to theoretical astrophysics. I got to learn a lot about the thermodynamics and fluid dynamics behind the formation of planets while also discovering numerical methods used to simulate physical phenomena. Furthermore, I had the opportunity to develop my communication skills by learning how to present my research to my peers.
My biggest challenge this summer was to write the code for my simulations and get multiple different pieces of code to work together. In this kind of project, it can be easy to get stuck on some problem for a long time. I definitely got to learn a lot from these setbacks and develop my patience over the summer!
Aside from the research experience I developed this summer, I particularly enjoyed the weekly iREx Cafés where I got to discover other interesting research in astrophysics. I really enjoyed working with my supervisor, Prof. Eve Lee. She explains the theory behind all the numerical work very clearly and her guidance greatly contributed to the success of my internship. Overall, I had a lot of fun this summer!
Note: These interviews were slightly edited for clarity.