Decoding starlight to learn the secrets of exoplanets
As told by David Charbonneau

I grew up in Nepean, Ontario, Canada, with my younger sister, Christine. Both our parents were scientists working for the federal government: my dad, Brian Charbonneau was a geologist who would return from summers in the Northwest Territories with a long beard and stories of close encounters with bears, and my mom Sylvia Charbonneau had completed medical school and was working as a research toxicologist to determine safe levels of various environmental toxins. My mom had fled Latvia during World War II as a child refugee, and grandpa Kazimirs was always very proud of my education and interest in science.

I can’t remember a time when I wasn’t passionate about astrophysics and Star Trek. My parents had agreed to let me choose the wallpaper for my room, and so I picked an enormous image of the Space Shuttle set against the blue marble of the Earth. I remember with pride when I used a star chart to see the faint smudge of the Andromeda Galaxy, the only object outside our own galaxy that can be seen without a telescope! Scouts Canada, and my scout leader Bill Bowman had an enormous influence on me growing up, both to develop a confidence in exploring the outdoors, and in developing leadership skills. And I always took a star chart along on the canoe trips in the Canadian wilderness.

A detour through biology
When I was twelve years old, my family traveled to the west coast of Canada for Expo’86 and we camped alongside the ocean at Pacific Rim National Park. I was captivated by the incredible diversity of life in the tidepools! So, from that age until senior year of high school, I pivoted to a love of biology. I was certain I was going to be a marine biologist!

"At my mother’s urging, I switched my major to physics at the last minute."

In the Canadian university system, you generally have to choose your major when you apply and so when senior year arrived I indicated biology. Biology seemed to be full of exciting and relevant questions, whereas physics seemed mostly to consist of force diagrams of frictionless pulleys. But in my last year of high school, I discovered a profound joy in physics. I remember sitting in the driveway during the brief, hot Canadian summer and reading A Brief History of Time by Stephen Hawking and being spellbound by the ideas of general relativity and quantum mechanics. Why had these ideas been withheld from me in physics classes? At my mother’s urging, I switched my major to physics at the last minute. It turned out to be good advice! (It isn’t lost on me that I’ve somehow managed to combine both interests, namely through the astronomical search for inhabited worlds.)

My first exploratory experiments
At the University of Toronto I eagerly enrolled in as much physics and mathematics as I could, and I stuffed in astronomy courses wherever possible. I won a research internship after my first summer to work in an inorganic chemistry lab, and the professor told me of a puzzling chemical behavior his team had noticed and set me free to do whatever experiments I wanted. Whenever I wanted to use new chemicals or equipment I just had to ask the grad students and postdocs how they worked. No one ever said no to my requests. It was a great summer and my first exposure to true exploratory experimental work, which seemed so fundamentally different from the canned labs I had done in my courses.

While in Toronto, my enjoyment of the outdoors led me to become active in a club organizing hiking and canoeing trips, and on one of those I met another undergraduate who was also studying mathematics and physics: Sara Seager, my Kavli co-laureate. Of course neither of us knew at the time that we were going to become professional astronomers (or at least I didn’t!). Canadians are raised to be naturally skeptical of the United States, and so when I decided at the end of my time in Toronto to apply to graduate school in astronomy, I assumed I would go to a Canadian school. I remember calling one program and they said the competition was very stiff so they suggested I not bother. But Sara had gone off to Harvard and was studying cosmology, and encouraged me to do the same. I was shocked when I found a voicemail telling me I had been admitted, but off I went a couple months later. Like most Canadians, I thought the US was crazy, and I would never end up living there permanently.

I arrived in Cambridge, Massachusetts in late August 1996 and settled in at the Harvard-Smithsonian Center for Astrophysics, which contained Harvard’sDepartment of Astronomy. I was certain I was going to be a theoretical astrophysicist and I would work on cosmology, which I now recognize was likely due to the fact that my training at Toronto emphasized mathematical methods over experimental design. Ten months prior, Michel Mayor and Didier Queloz had announced their discovery using the Doppler method of 51 Pegasi b, the first planet known to orbit another Sun-like star. But I had not heard the news until I got to Harvard.

In the fall of 1996, I went to an informal talk there by Bob Noyes. He described that a debate was raging as to whether the signal seen by Mayor & Queloz, and recently found for a handful of other stars, was really due to an orbiting planet, or instead was merely a previously unidentified form of stellar pulsation that had only now become detectable due to new instrumentation. Bob described an experiment to test this idea, which was to gather high resolution spectra of the star and to look for a faint copy of the stellar spectrum due to the light reflecting from the planet. Nowadays we use this method as a means to study exoplanet atmospheres, but at the time the goal was much simpler, namely to prove whether or not such planets even existed.

"It certainly wasn’t a prestigious field"

I was hooked: It seemed that cosmology had worked out the fundamental ideas and now the active questions were refinements pursued by large teams, whereas the questions of exoplanets were basic and straightforward: Were there planets orbiting other stars? Were they similar or different to those in our own solar system? It certainly wasn’t a prestigious field, and there wasn’t much literature, probably a grand total of 10 relevant papers at the time. In early 1997, I jumped into this project.

Meeting my soulmate
That year was also the year I decided to move into the Harvard Dudley Coop as a resident tutor, to save money on housing and because I had lived in coops in Toronto and I was a big fan of the notion of communal housing. I would often be working late at night and drop into the large shared kitchen for a break and conversation, as there was always a crowd of students.

It was there that I met a stunning redhead who I learned was not a resident, but instead was sneaking all of her meals as a means to save money while working at the Boston Medical Center. This was Margaret Bourdeaux, now Prof. Margaret Bourdeaux of the Harvard medical school, my wife and soulmate.

A slow start
I spent two years toiling on the dataset from the Keck Observatory attempting to detect the light reflected from the planet Tau Boo b. Each day I would return to the co-op and my housemates would joke “did you find the planet today?” and I would respond no, and we would all have a laugh. But I began to worry. Some of my PhD classmates had already written several papers, and I had nothing to show.

“But DAVID, you will NEVER be able to detect such a small signal!”

My first international conference meeting was a NATO workshop in Corsica and I’ll always remember the first talk: Here we were, a group of students in their early 20s in a field that barely existed, and Dr. Martin Harwit warns us that whenever two civilizations came into contact for the first time, it always ended very badly for one of them. I was struck by how far-fetched, almost silly, such a worry seemed, while also impressed that it was apparently OK to already be thinking about the radical, long-term impacts of this work! Later when I presented my ongoing work to detect reflected light, the organizer laughed and blurted out “But DAVID, you will NEVER be able to detect such a small signal!”.

I wasn’t put off by this, but my point is that even at a meeting on exoplanets, there was a great deal of skepticism of what we could ever hope to know about these worlds. My reflected light work eventually resulted in only an upper limit, but I had developed a sturdy sense of self-reliance and a deep ability to understand imperfect and incomplete data. My time had been well spent. And at least I had a paper!

"Tim pointed me to a small wooden shed in a parking lot, which housed the experiment."

For my thesis work, I was encouraged by my advisor Bob Noyes to not work on the Doppler method, which many others were already doing, but instead to join forces with a colleague of his in Boulder, Colorado: Tim Brown, at the High Altitude Observatory. Tim had built a 4-inch telescope and was planning to search large patches of the sky for eclipses (now called transits) of exoplanets. No one had detected a transit, but the claim of the existence of hot Jupiters (still debated!) implied they might occasionally exist. I bought a used Ford and drove to Boulder. Tim pointed me to a small wooden shed in a parking lot, which housed the experiment. Could we really discover planets with such a modest setup? Some of my fellow students were analyzing data from the Hubble Space Telescope, and here I was facing a telescope smaller than that employed by most amateurs.

After about a month, I got an angry call from the director of graduate studies at Harvard. “Where the hell are you?” he barked. I had not cleared the plan with them, and so I had to fly back to Cambridge to explain myself and seek their approval.

During my visit, they were dubious, and so I had to write a lengthy thesis proposal describing all the research one could do with the vast data on variable stars that the survey would generate (if it found no planets). The no-planets option was thoroughly boring, but at least, I assured them, there would be papers. And I had to submit weekly reports.

A glorious 1 % bump
I had never done photometry before, so as a starter project in Boulder I proposed to Tim that I look at a star that a colleague at the Center for Astrophysics, Dr. David Latham, had been studying with the Doppler method. Latham and his team had evidence that it was orbited by a hot Jupiter, but maybe only the star was to blame. In August, most nights in Boulder are cloudy, but by September most were clear and we managed to get several dozen nights on the target with Tim’s STARE telescope. I first analyzed several nights when no event was expected, and indeed the light curves were flat. I then analyzed the data from a memorable date: 9/9/99. And it showed a 1% bump. A beautiful, glorious, 1% bump!

The discovery process is not at all like I imagined. It was deeply uncomfortable. Maybe I was fooling myself? But I analyzed more random dates, and they were also flat. And then the one other date when a transit was expected. The bump was back! I raced to tell Tim at his house. By this point, my Harvard committee wasn’t asking for weekly reports anymore! The transit proved beyond a shadow of a doubt that there was an exoplanet, and moreover Tim and I quickly calculated its density and the implied composition. It was indeed a hot Jupiter.

The paper presenting the discovery of the transits of HD209458 is only 4 pages, and the last page is devoted to ideas of what could be done once transits of exoplanets were known. In many ways I’ve spent much of my time since then pursuing this list of ideas. The first of these was to study the atmosphere in transmission. By December 1999 we had submitted a special request to the Hubble Space Telescope to do just that.

"We had made the first detection of an exoplanet atmosphere!"

We elected to target a feature due to sodium, guided by the vanguard PhD work of Sara Seager and her predictions of the properties of hot Jupiter atmospheres. The data arrived in mid-2000, and it would take me a year to extract the sodium signal from the Hubble data. The skills I had acquired in my earlier reflected light challenge served me well. We had made the first detection of an exoplanet atmosphere!

Immediately our minds jumped ahead: If we can do sodium in a Jovian atmosphere, why not a biomarker like oxygen in the atmosphere of a terrestrial exoplanet?

Settling down in Massachusetts
After I defended my PhD at Harvard, I chose to head to Caltech as an R. A. Millikan fellow as Caltech and JPL had opened up funding for me to build an automated survey telescope at Palomar Mountain. I still have this Sleuth telescope, displayed in my office. Guests often assume it is a matter of pride or historical significance, but actually I keep it as a reminder to myself of how not to manage a project. Because I had tried to save money on hardware, I spent all of my time driving to and from Palomar fixing it. Margaret took to referring to it as “that bitch” as it always called me away in the middle of the night.

At this point, Margaret and I were running the ultimate long-distance relationship, as I was often at the observatories in Hawaii, and she was in Kosovo treating the children of communities that had been devastated by the war there. We married in North Carolina in a Quaker ceremony in October 2003. Margaret had already begun a medical residency in Boston, and I was still a postdoc in Pasadena 3000 miles away. So the officiant asked that those gathered at our wedding pray for a physics job in the Boston area.

As luck would have it, MIT posted a search for a professor of astronomy that year, and I was overjoyed when I was told that I had been selected as the top candidate by the astronomers there! Would Margaret and I finally get to live in the same city? Several weeks later, I was devastated to learn that the department of physics had overturned the plan, as the field of exoplanets was apparently not physics. Fortunately Harvard later decided to run a search and I did receive a bona fide offer.

Overjoyed, Margaret and I rented an 800-square foot condo in Brookline, Massachusetts, walking distance from the hospitals where she was working 100 hours a week. Our first three daughters, Stella, Aurora, and Selena, were born within 3.5 years. We moved into a house and welcomed our fourth daughter, Terra, in 2013. With both Margaret and I working full time and four school-age daughters, life has been hectic, messy, and wonderful.

Working with the Spitzer Space Telescope team
After I started my professorship at Harvard in 2004, my first new collaboration was with Lori Allen and Tom Megeath, who were at the Center for Astrophysics as part of the Spitzer Space Telescope team. They wondered if Spitzer might be useful for exoplanet studies? As a student I had gotten time on ground-based telescopes to try and directly detect the thermal emission from exoplanets near the time of secondary eclipse (when the planet would pass behind its star), but we were swamped by the infrared photons of our own atmosphere. Spitzer, free from such worries, proved to be an unparalleled platform for such work, yielding not only the first detections of light emitted by exoplanets, but the first
emitted spectra, and ultimately allowing us to reconstruct maps of the exoplanet surfaces. These orbs had moved from mere spheres to worlds with personalities.

An unexpected discovery about M-dwarfs
During these first years as a professor I also became obsessed with small stars, so called M-dwarfs, and the opportunities they would provide for discovering small planets. I was fortunate to receive a Packard fellowship, and I’m very grateful for the injection of a large amount of unrestricted research funds, which allowed me to build the MEarth project. MEarth was an array of 8 robotic telescopes at Mount Hopkins in Arizona, later expanded to a second set of 8 telescopes at the Cerro Tololo Observatory in Chile. Unlike other surveys, MEarth would be pointed, surveying only one nearby M-dwarf at a time and jumping from one star to another every minute or so, but returning with sufficient frequency so as not to miss a transit.

"We found that most M-dwarfs rotated extremely slowly"

It was through MEarth that I began a long and very fruitful collaboration with Dr. Jonathan Irwin, who had just finished his PhD at Cambridge. Quite simply, Jonathan made MEarth work, and over the following years I’m proud that we found some of the most spectroscopically accessible small exoplanets. But because we were looking at these stars in a novel way, namely monitoring them nightly for years, we made a completely unexpected discovery: In contrast to the widely held belief at the time, we found that most M-dwarfs rotated extremely slowly, with typical periods of 130 days that no one had previously been able to measure. This meant, in turn, that they were magnetically quiet and produced very little in the way of high energy emission. This had enormous implications for the ability of their attendant worlds to retain atmospheres.

The purposefulness of each day
I have been profoundly blessed to work with so many great graduate students and to pursue all these adventures together with them. I’ve still largely been able to work one-on-one with them on their thesis projects, in the same manner that my advisors Bob and Tim did with me. Now that I’m just completed my 20th year as a professor, I’m at the point of meeting the students of my students. Two years ago I received a shocking diagnosis of stage 4 colon cancer completely out of the blue. Apparently the number of such instances in patients like myself who are under 50 and have no risk factors is increasing dramatically each year. My goal here is not to end on a down note.

"I want to spend as much of my professional time as possible working on the next exciting explorations in exoplanets with those students."

Our community in Brookline has been unbelievably supportive (and I’m pretty sure that everyone within a 5-mile radius has now gotten their colonoscopies). Rather my point is that I feel in many ways I’ve been given a very special gift, a clarity of mind that possibly trades the number of days that I have for the purposefulness of each of those days. And after a great deal of reflection I’ve decided that I want to spend as much of my professional time as possible working on the next exciting explorations in exoplanets with those students.

We can now point to a handful of nearby stars and state with confidence that each hosts a rocky planet with the same size and temperature as the Earth. And we now know the methods that we will use to probe their atmospheres for biomarker gasses. It will take bold new observatories, and students ready for the greatest scientific adventure.