Exoplanet Innovators Interview: Sara Seager Interviews Dave Charbonneau

Sara Seager: Welcome. It’s really exciting to have you as part of the Exoplanet Innovators interview series. Thanks for joining. Let’s start with how we met.

Dave Charbonneau: Okay, well, my recollection is that we were both undergraduates studying math and physics at the University of Toronto. You were a few years ahead, and I had always loved the outdoors and found a community at the University of Toronto Outing Club. And I met you through that, and in particular I remember a hiking trip that we did in Frontenac Provincial Park in Ontario.

Sara Seager: I remember that. It was late fall, and the world was kind of shutting down for the winter.

Dave Charbonneau: Yes, and I’ve always loved Frontenac because it’s this extrusion of the Canadian Shield. This is a very kind of rocky landscape that you have to sort of hike up and over all of these ridges. So, it has always been one of my favorite Provincial Parks, and there we were.

Sara Seager: Yes, I remember it being off season, because in the winter people would cross-country ski. Summer was for canoeing. I was just browsing the selections of Outing Club trips, and I saw one that was to be led the upcoming weekend. I remember playing phone tag, because in those days we only had landlines, and finally getting in touch with you, and I was absolutely astonished that the leader of this trip was also studying math and physics. It was a very underpopulated major, the joint math and physics specialist program, and then, I remember doling out some advice on that trip, because I usually gave advice to people slightly younger.

Dave Charbonneau: Yes, actually, you doled out more than advice. [Sound of chair moving.]

Sara Seager: Oh, my gosh, okay, this is funny [laughs]. We are in Dave’s office at the Harvard-Smithsonian Center for Astrophysics right now, and he’s going to his bookshelf.

Oh, wow! You have my book, my old textbook from 1992. Jeez, that’s crazy.

Dave Charbonneau: [Laughs]. I’m showing Sara her book, “Electricity and Magnetism by Purcell,” and I’m sure you handed this over when you were packing up, heading out to Harvard.

Sara Seager: Well, let’s move on to that. I remember applying for graduate school, because at some point—and maybe you had this realization, too—I realized that if you find something you love doing that you’re good at, you should pursue it. According to my father, I was supposed to be a doctor and do something more practical. So, I remember applying to grad school and surprisingly being admitted to Harvard and going to Harvard, and you also went to Harvard a bit later.

Dave Charbonneau: Yes, and the reason I applied to Harvard was because you encouraged me to. I had assumed I would go to grad school in Canada and the Canadian application deadlines were later. So, in the end I ended up never applying to any schools in Canada, because I had already found out I’d gotten into Harvard by that time.

But I remember being a little unsure whether I could really make it as an astronomer, as that seemed sort of niche compared to physics, and so I had thought I would only apply to grad schools in physics. But then you had mentioned that the program was very good and encouraged me to apply, and so I was really delighted when I got in, and of course, here at Harvard it isn’t just Harvard, it’s the Harvard Smithsonian Center for Astrophysics (CfA). It’s hundreds of research scientists, and so any feeling that astronomy is a little niche activity immediately goes away—once you walk in the building, you realize this is a massive enterprise.

Sara Seager: Right. One of the reasons I chose to come here. Okay, admittedly, I didn’t have too many choices, but one of the reasons I chose to come to the CfA is because I didn’t know which part of astronomy I wanted to study, and knowing that there were a large number of topics to choose from was definitely a bonus.

Dave Charbonneau: Yes and I think we both started in cosmology right?

Sara Seager: Yes, and oh by the way—just as an aside—I recently bumped into Dick Bond, and he was actually extroverted and really nice, which is not the authoritative Dick Bond we remember of our youth. [Both laugh]

Dave Charbonneau: Yes, Dick Bond was my advisor in my undergraduate senior year project and my summer research, and he kept what I thought were very strange hours. He sort of came in toward the end of the day and then worked all night, and usually that’s what you expect the students’ hours to be and the professor has more 9 to 5 h. But I was the guy coming in at 8 or 9 in the morning and trying to leave at 5 PM; and I remember often meeting Dick in the hallway as he was coming up the elevator, and I was going down, and he asked me once if I was working enough. I had to assure him I was. I just was trying to end my day around 5 or 6 PM.

Sara Seager: I’m pretty sure you’re not allowed to ask that question nowadays. [Both laugh]. But what was your undergraduate cosmology project? Can you let tell us a bit about that project?

Dave Charbonneau: So, well, I don’t think I had ever heard of exoplanets while I was at Toronto. I now know that the first planet orbiting a Sun-like star, 51 Peg that had been announced in October of 1995, which would have been my last year, I would have been a senior, but I don’t think I’d really heard much about it. And for me astronomy was cosmology; that was obviously what everybody in astronomy should be studying, because the questions were profound. You know, how old was the Universe? What is the Universe made of? How is it going to evolve in the future?

And so, I found my first-year study plan, which everyone has to file when they’re here at Harvard, and it’s all about doing theoretical cosmology. I also had really assumed I would do theory because I had studied mathematical physics as an undergraduate, and, I think, the program at Toronto at the time really emphasized mathematical methods and theoretical physics as opposed to laboratory physics.

Sara Seager: Can you give a two or three sentence summary of the cosmology project you worked on as an undergrad?

Dave Charbonneau: As an undergraduate, I worked with Dick Bond on the cosmic microwave background, and it was right at that time where physicists were seeing the first peak in the power spectrum. [The power spectrum of the cosmic microwave background temperature anisoptropy on the sky.]

So now, of course, the power spectrum is exquisite, and we can see all of the wonderful structure in the power spectrum. But at the time there were just these data sets, and in particular, there was a data set that was being gathered by a balloon experiment and one had to very carefully measure the beam of the experiment. So, to actually understand what parts of the sky were being observed and being correlated at any particular pointing. And so, I worked up details of that and what the impact would be on the measurements, and it wasn’t just Dick Bond; there was a group of postdocs as well. But honestly, I feel like I didn’t really make much progress. So later that was published as part of a larger group effort.

Sara Seager: Yes, not to worry. I was telling my son, who’s an undergrad that undergrads usually don’t contribute too much to research. So, this is really interesting. And you switched to exoplanets in grad school?

Dave Charbonneau: Yes, once I moved to Harvard as a graduate student, I changed topics within a few months. That’s because Bob Noyes, who was a professor here and a Smithsonian scientist, gave a talk about how these few—at that time it was just a handful—planets had been found and there was a debate about whether they really were planets or whether it was some misdiagnosed form of stellar pulsation, because all that was measured was the radial velocity of the star.

Bob proposed a very straightforward project in a sense to answer that question. If you could see the light reflected from a planet then that would prove there really was an object there that was a planet, as opposed to being just a stellar pulsation.

Sara Seager: Yes, let’s just pause for a second, because I’d also started working on exoplanets back then, and Bob had told me that he did believe deep down exoplanets were real.

Dave Charbonneau: Yes, I think that he certainly felt it was somewhat contrived to come up with a model that could generate a stellar pulsation. The individual who was leading that proposal was David Gray, who was an astronomer at The University of Western Ontario, in Canada, and it was physically plausible but maybe unlikely. But there was nothing wrong with the model. And so, this exoplanet reflected light project just seemed so practical to me compared to joining cosmology, which was a very mature field. There were a lot of senior people in cosmology, and I felt that maybe my contributions would be incremental.

Sara Seager: I’m impressed! Did you have that level of maturity to think through that?

Dave Charbonneau: I think it was more the practical aspect that really impressed me, that there was this very straightforward question: Is there really a planet orbiting the star or are we being fooled? And it wasn’t hard to explain to anyone what that question was. I liked the fact that I could describe my research to nonastronomers, nonscientists. And, nobody else was doing it. Even though exoplanets are a huge field now, you have to remember that at the time there were really very, very few people involved. Most institutions didn’t have anyone working on exoplanets.

Sara Seager: Well, exoplanets was certainly not grant funded back then, as you know. And, Harvard, let’s just face it, is a great place for doing new things, because there are financial resources to fund people to work on new topics.

Dave Charbonneau: Yes, and Bob’s group had a spectrograph called the AFOE spectrograph. They were thinking of trying to measure asteroseismic signals. So, they were thinking about trying to measure the pulsations of stars with precise radial velocities. And of course, lo and behold! They suddenly found they had a machine for actually discovering planets through orbital motion. And so, that’s why they were ready to pivot, and they were experts in terms of developing and guiding me in terms of understanding the really nitty, gritty details of stellar spectra, which is what was required to really push down and try to see this reflected light. So, I pivoted by the late fall of 1996, roughly, a year after the announcement of 51 Peg b, to try to do this project. I left cosmology behind, and I never really looked back.

Sara Seager: So, again, it’s really astonishing that a graduate student would take on the new reflected light project, while not really knowing anything about exoplanets and having to understand the instrument to some exquisite level of detail to try to find what presumably would be a very tiny signal amid data full of systematic noise.

Can you outline some of your efforts and how this reflected light project ended up?

Dave Charbonneau: Yes. The data had been gathered with the Keck high-resolution spectrograph, which was a workhorse instrument for trying to discover exoplanets by radial velocities by Geoff Marcy and Paul Butler, by that team. But here, of course, the project was very different. What we were trying to see was whether there was a copy of the stellar spectrum Doppler-shifted by maybe 100 kilometers a second sitting elsewhere in the data. And the trick of the spectrograph is that it’s a slit spectrograph. So, it’s not a fiber-fed spectrograph, which at the time, you know, was a distant technology or at least not widely in use. And what that meant was that the shape, the instrument response function, the shape of the beam of the starlight that’s being passed into the entrance aperture of the spectrograph is constantly changing, as the telescope does not point perfectly, and there is a lot of jitter and there are seeing changes. And so, every spectrum is different. Every single spectrum that we had gathered—that they had gathered—was bouncing around. And there were hundreds of these spectra. And because we had to see that that’s the spectrum of the planet move and we had to wait for that to elapse. And so, these data spanned many, many different times under many different conditions. So, I would say that was definitely a year and a half of work.

We never found the reflected light. Okay, so we never discovered, we never proved that there was a planet orbiting the star Tau Boo by this method. Everybody now is pretty confident there really is such a planet, and this is one of the original hot Jupiters. But that was the object I was looking at. But what I know now, that I didn’t appreciate at the time, was that this real confidence in just developing my own methods to really understand every little part of a data set and to not be put off by data sets that were flawed and gappy and imperfect and understand that every feature really could be understood. And it doesn’t mean you can make the data better than they are. But you can really understand the uncertainties. And so, when you really do see something, and then later down the road, for me, for example, I would see things in spectra that I knew I could establish—not just for myself, but for the community—such that the effect was real.

Sara Seager: Let’s pause and let this land. I think it’s incredibly helpful for the readers out there, because in today’s world we’re used to just downloading python code. Downloading some data that’s already calibrated to some level. I’m thinking of the James Webb Space Telescope (JWST) and Transiting Exoplanet Survey Satellite (TESS), and other things, and there are very few people who practice the craft you described, of really understanding every last detail. Do you have any thoughts on that?

Dave Charbonneau: Well, I mean, in many ways those codes are great because they’re incredibly sophisticated. You know the statistical methods that astronomers are using now to understand data sets are very, very advanced compared to what I was coming up with and trying to apply.

Sara Seager: But what about the lack of understanding for individuals today?

Dave Charbonneau: Yes, one element when it comes to exoplanet atmospheres that I really do worry about for students is pacing. Back then, I wasn’t under any time pressure, because nobody else was working on this.

Sara Seager: There was the Scottish group.

Dave Charbonneau: Well, I found out later that another group was working on it, but I didn’t know that at the time. But also, my advisors were really focused on getting it right, which is also a great message. And now, because a lot of the data perhaps go public immediately, there is a bit of time pressure, and the danger is there are these codes that will take your data and produce the spectrum. And then there’s another set of codes that will take the spectrum and tell you what molecules are present, in what abundance. And I worry that students are taking those codes and getting results, and the results sort of make sense, but they aren’t really approaching it with a very suspicious and critical heart, which they really must do. Otherwise, we’re all just going to reach the same conclusions. But it doesn’t necessarily mean they are the right ones.

Sara Seager: Yes, I’d like to add that we are seeing things like people will present multiple data analyses that differ by their treatment of offsets and not knowing how to reconcile them. We’re seeing what might be insidious flaws in retrieval, but we could leave that for later if we have time to unpack it.

Switching gears now, back to the past, like finishing high school, undergrad, or this first part of graduate school we have been talking about. Who were some of the most influential people in your life? Could you pick a mentor or just someone who is worth calling out?

Dave Charbonneau: First of all, I’m really thankful for my summer undergraduate research opportunities. They were very, very different opportunities. My first year I got to work in an inorganic chemistry lab. My third year I worked at the TRIUMF Nuclear Research Facility, studying an astrophysically relevant nuclear reaction. And then in my last year I worked on the cosmic microwave background. And they were not all great, but I learned a lot from each of them, and so I think it is really, really important that we continue to offer undergraduates messy, somewhat open-ended projects where there’s a lot of mentoring available, but maybe things aren’t so neatly tied up with a bow. I’m a little worried sometimes we give undergraduates projects that can’t fail and we’re sort of not really teaching them a really important part of science. Something which is really important when you get to grad school, because grad school is all about dead ends.

Once I was in grad school my two main mentors were Bob Noyes, who was my academic advisor, and guided me on the reflected light project, and then one of the really tremendous things he did was, he pointed me towards Tim Brown. Tim Brown was not at the CfA but at the High-Altitude Observatory in Boulder Colorado. And as I finished up my reflected light project, Bob said, “Don’t do radial velocities. Everybody else is doing that, and you’re not going to catch up to those big teams. Instead, go work on transits. There is this guy named Tim. He’s built a small telescope and is hoping to go and look for transits.” If there really are this many hot Jupiters, then the idea of looking for transits of exoplanets was no longer crazy. If all the Jupiters were out in 12-year orbital periods, then that would be a terrible thesis project. [Because the transit probability drops linearly with increase in the planet-star separation.]

Sara Seager: That was really sage advice and very prescient of Bob to have been that forward looking.

Dave Charbonneau: Right. And so, I went out and I bought a car, a cheap car, and I drove out in probably late July or August of 1999. And then I set up shop with Tim. So, the point is, in terms of mentors, Tim Brown and Bob Noyes. And then the third mentor really was Dave Latham. Dave Latham is another astronomer here at the CfA, and he would host wine tastings in this office. But of course, really what was going on was networking. Everybody came to those. A lot of very, very senior professors and a lot of very junior folks. And I got a chance to really feel like I was part of this community and also present some of these ideas that I was thinking about with some other graduate students in exoplanets, well with one other guy Saurab Jha, who’s a very close friend in exoplanets. And then later, Dave Latham, of course, was the connection to HD 209458. He was the one who suggested … I said, I want to pick one star and look for transits, and I asked him if he had a good candidate, and he said, yes, I would look at that one. So that was also obviously very, very impactful.

Sara Seager: Yes, that was amazing. So, I’d also started working on exoplanets in the summer of 1996, because I had worked on cosmology, and I’d solved my research problem. I got great advice from both Bob Noyes and my PhD thesis advisor Dimitar Sasselov that my niche cosmology field was a dead end. I mean, it’s amazing they had very strategic thinking here, looking back. Dimitar asked if I wanted to work on these new exoplanets, specifically their atmospheres, models. And so, I jumped on that in the summer of 1996. When I was rushing to finish my thesis in the summer of 1999, and I knew, well, it seemed kind of obvious that a transiting planet should be discovered any day, because we had a handful of hot Jupiters. There were more being discovered, although we only had about 30 planets overall, and every day I bumped into Dave Latham, because his office was around here somewhere, I think, somewhere around this corner. Well, my office was down the hall in that graduate student corridor, and his was right kitty corner to that grad student area.

So, every day I bugged him about it because I knew he had four radial velocity planet candidates. And I weakly tried to find the first transit myself. Actually, I asked my old advisor at my Alma Mater, and they didn’t have good enough photometric precision. As soon as I moved to be postdoc in 1999 at the Institute for Advanced Study in Princeton, I found out that Princeton has only about one photometric night a year. And, by the way, on that one photometric night I was camping in New Jersey, not wilderness, but whatever you might call it, and saw the Andromeda Galaxy with my naked eye, something that must be incredibly rare in New Jersey. Anyway, I just didn’t push it too hard, but knew the transit had to be there soon. But I didn’t know how all this unfolded. Later, in January 2000, at a conference, Dave Latham came up to me and gave me a giant hug. I’m basically not a huggable person. I do not hug. But he was really thrilled how this whole thing played out because I’d been bugging him every day about a transit that was going to appear. You wanted a target star, and it all just worked out.

So, tell us about this. I mean, this is so huge—the first ever transiting planet. Can you briefly encapsulate that story for us?

Dave Charbonneau: Yes. So, it was late summer, and that’s important, because you have to understand the weather patterns for astronomers. In Colorado, there’s the monsoon. So basically, it’s very, very cloudy. And so, you can’t do observing. And so, I went out to Boulder, and Tim described this transit survey he wanted to do where you would look blindly at 10,000 stars, and that’s why he built this very special telescope because of the wide field of view. And I said, “Well, that sounds great. But why don’t I start with doing a follow-up study on a candidate hot Jupiter?” And the reason is that I had never done photometry before, I’d always done spectroscopy, which is actually harder.

Sara Seager: Oh. I did photometry as part of one of those undergraduate research programs.

Dave Charbonneau: So, I talked to Dave and he suggested HD 209458. Now, very importantly it would be a mistake to say OK we knew there was a planet around 209458, and I went and measured the transit. At the time the group that was studying HD 209458 was not at all certain that there really was a planet. There was a big debate internally, and all the emails that I was part of …

Sara Seager: Like a signal-to-noise problem with the radial velocity?

Dave Charbonneau: No, an activity problem. We know that, for stars that are active, you can get a pseudo sinusoidal radial velocity signal, and so there wasn’t at all certainty. I think Dave was pretty confident.

Sara Seager: But four days, though? I thought the activity signals on four days were rare.

Dave Charbonneau: No, no, you definitely have signals that can be like that.

Sara Seager: So, I’m just wondering why? Because I swear since in the summer of 1999 I was talking to Dave Latham every day and bugging him, I remember he had at least four candidate planets. Did he give you all four, or did he just give you this one?

Dave Charbonneau: Well, no, this was the one….

Sara Seager: That’s crazy that you were so lucky.

Dave Charbonneau: … this was the one that was up visible in the sky in the fall. So, then we started trying to observe it, and Tim showed me how to use the observatory. But, of course, it was cloudy for weeks. And then finally, in early September, we started getting data, and the first night when I calculated there would be a transit is very easy to remember, because that was 9, 9, 1999.

Sara Seager: Your lucky number must be 9.

Dave Charbonneau: Yes, that’s right. And then we got some more data. Of course, equally important to get lots of data on nights when you wouldn’t expect a transit. But then I didn’t analyze those data right away, because I had also joined a project with Ron Gilliland at Space Telescope and Tim Brown to do a transit survey with the Hubble Space Telescope of the Globular Cluster 47 Tuc. And the thinking again, there were an enormous number of stars, tens of thousands of stars. Compared to looking at this one star where having a transiting planet was unlikely. And so, I worked very, very hard on figuring out the sensitivity. How many planets would we have found given the quality of the data. Now, of course, in the end, we now know that there are no transiting hot Jupiters in 47 Tuc, and there may be very interesting reasons for that.

But then, finally, in late October, early November, I was able to shake free and return and analyze the HD 209458 data. And lo and behold, there was this transit.

Sara Seager: Describe that moment.

Dave Charbonneau: It was very … So, I first pushed through the analysis on the night when there wasn’t a transit, and it was flat, which was a little rumbly. And then I looked at night when there was a transit, and I remember very clearly first plotting the data and not seeing a transit and then remembering oh, I was using IDL, which is software, and some software packages use index 0 and some use index 1. So, I realized, oh, I have to change the number of the star by one. I was looking at the wrong star, and I changed it. And then there was this very clear bump, and I remember thinking, well, that’s a great experiment, because I suddenly could see this transit. Now the transit was inverted because I was using magnitude. So, it was a positive bump. You might think if you’re hunting for something and if the entire purpose of the experiment is to find something, and you sort of see it, that it’s very comforting. Actually, it’s very discomforting, because if nobody’s ever seen something before, you become very worried that you’re being fooled and it really is very important to not delude yourself.

And so, I analyzed a few other nights. And then what really did it was there was one other night where we had gathered some data. The quality was not as good, and that was a week later, that was on the 16th of September, and when I analyzed those data, I saw the transit as well at a lower signal to noise. And so then at that point I was very, very, very confident. It was incredibly exciting. I had no idea of the impact it would have on me professionally, but what I definitely knew was that, for the first time, we could calculate the planet’s density. And so, Tim and I both excitedly calculated the density of the planet and learned that it was indeed a Jupiter, in fact, somewhat puffy.

Sara Seager: Well, that’s an amazing story. It’s just astonishing that a grad student could detect the first transit ever. I mean, it’s just wow.

Dave Charbonneau: Yes, it was just really, really special.

Sara Seager: Do you want to talk about the competing team or leave that out?

Dave Charbonneau: It is very, very important to note that the transit was independently measured by Greg Henry at Tennessee State. And he observed it in November.

Sara Seager: Weird thing is, in a way, we’ve lived a bit of a parallel professional life, which is … I love it. So, because I was so sure a transit would be discovered any day while I was finishing my thesis, the first thing that happened once I graduated from Harvard and I moved to the Institute for Advanced Study in Princeton … Well, I mean, you know what a good small talk question is, if you’re bad at small talk, which a lot of astronomers and physicists are [both laugh]? It’s, “What’s the next big thing?” So, I arrived at the Institute. No one there had ever worked on exoplanets. And everyone is just asking me, “What’s the next big thing?” And I say, “A transit.”

So, I started working on transit transmission spectroscopy. I actually also invited Dimitar to chip in, mostly so I could have a sounding board. And I really worked hard on it. Then I remember, one day I thought, “You know what? I’m going to have work life balance.” It was a November fall weekend with leaves on the ground, and I remember leisurely walking my dog. That may have been the first weekend I ever did that while home (no smart phones back then, so on outings no email). Sunday night I check my email, and I have a message from you.

Dave Charbonneau: Oh. [Laughs].

Sara Seager: And my reaction is Whaaaa?!! Transit!!! So, I dropped everything. Such as I was running a reading group on planet formation and I dropped that. I dropped everything and busted my butt to get my paper done. And I got this paper done. Got it submitted, you know. A bit more behind that story. But wow! That was an exciting time.

Dave Charbonneau: Yes.

Sara Seager: And we can continue the story, what had happened was while I was finishing my PhD thesis, I thought I had a bug in my code because the sodium gas signal was so strong in my simulated spectrum. This isn’t just about transmission spectra, just in general. I honestly thought something was wrong. But I went away for my requisite one month offline for my annual Northern Canada canoe trip. When I got back, I saw a paper in the literature on brown dwarfs, with a giant sodium feature. So, then I knew. Wow, okay, my code is probably okay.

Dave Charbonneau: That’s interesting.

Sara Seager: I knew that was really a key part of the puzzle, because when I did the transmission spectroscopy sodium was also strong.

Dave Charbonneau: Our decision to pick a band to include sodium was, of course, driven by your paper, that and the brown dwarf study and we felt like this was the best bet.

Sara Seager: Well, you can say, logically speaking, sodium has an incredibly strong oscillator strength for its ground to first excited state transition. There’s some logic behind it. You can have a code that always helps, but just this fact that a gas at the temperature of the hot Jupiters will have sodium. It sounds so unusual because in stars sodium is ionized and in cooler objects sodium is in molecular form with another atom. So, we had a lot of reason to believe that sodium would be really strong.

Dave Charbonneau: Yes, and in terms of the observation of sodium. I also want to make sure to mention Ron Gilliland. So, Ron was working at the time at Space Telescope and Ron contacted Tim and I, and said that basically, you could use an instrument on HST called the STIS instrument in a mode that nobody ever thought about using, which is basically, you would saturate the pixels. And he had done studies showing that, even though you saturated the pixels, the photo electrons would bleed over into adjacent pixels, but no information would be lost. And so, you had to create a custom aperture where you could co-add that light back up again, but that you could then get much, much higher signal to noise. And I think we’re really thinking of the transit. Not just the atmosphere or the wavelength-dependent study, but just getting an exquisite measurement of the transit.

And then all of a sudden, we could get a much, much, much better measurement of the ingress and egress and look for moons and rings, and so that’s what led to the proposal to try to observe it with Directors Discretionary time with the Hubble Space Telescope. It was Ron really who pointed out the special mode to us.

Sara Seager: Well, I really like this conversation a lot, because it’s not only giving credit where it’s due, but it’s showcasing the deep technical understanding that was required to make these early discoveries.

Dave Charbonneau: Yes, I think that’s absolutely right.

Sara Seager: I sometimes feel like it’s lost from the conversation today, because NASA, in a good way, wants to make the data and the science more accessible to people. For example, as you know, I helped lead the TESS team, and our job was to help make sure that planet candidates got out there. Although people could download and use a more basic data product, NASA also wanted people to have a more digested version.

Dave Charbonneau: Yes, I mean, you certainly know this, Sara, but I think the thing to convey to the readers, too, is that, of course, the Hubble Space Telescope was designed and launched long before exoplanets were in anybody’s thinking. And certainly, long before transmission spectroscopy was in anybody’s thinking.

So, knowing how a general-purpose observatory behaves under these very, very tough conditions, and then being able to turn and exploit that. And that was also true with Spitzer. So then, later, when we made the first detection of the secondary eclipse with Spitzer that was really developed in conjunction with two scientists who had really developed mastery of their understanding of the camera on Spitzer, which is called IRAC. That’s Lori Allen and Tom Megeath. And they were both on the IRAC team here at the CfA, and they had the technical know-how so that we could go and try to measure the secondary eclipse.

Sara Seager: Let’s reminisce for a moment. It was a pretty interesting time I was working on Drake Deming’s team, the only other team doing this. Can you imagine they’re only being two teams?! As compared to today.

Dave Charbonneau: I know and I have a great story about how Drake and I found out about each other’s work.

Sara Seager: Let’s hear it.

Dave Charbonneau: Well, so I was working on IRAC, and I didn’t know that Drake and Sara were working on MIPS, which was the name of the instrument. So that’s a different wavelength. And we were studying a planet that we had just found with the Wide Field Survey.

Sara Seager: TrES 2, I think.

Dave Charbonneau: TrES1, the first.

Sara Seager: And we were doing our old favorite HD 209458.

Dave Charbonneau: Yes, HD 209458 a great choice. Love it. And I again seen the secondary eclipse. And I was again going through that moment of do I really believe it? It looks compelling. But am I kidding myself? And I certainly didn’t want to go and announce anything. And, I showed Drake the measurements at an Aspen meeting. There was a meeting in Aspen, CO, about exoplanets, and I remember showing it to him, and I said I was nervous about whether the signal was real. And he said, “David, I can see it from here,” and he was sitting far away from my laptop.

Sara Seager: Aww, that’s so sweet.

Dave Charbonneau: Yes, and I believe at that time he told me about the MIPS observation. So, we realized that we really were seeing the secondary eclipses. And, of course, that’s just tremendously complementary in terms of the information it gives about the atmosphere [as compared to the primary transit]. So that would have been late 2004, early 2005.

Sara Seager: Right, yes. Okay, so let’s just switch gears and get to some personal/professional questions. Was being a professor of astronomy something you always dreamed of or if not, what steps in life led you to take this path?

Dave Charbonneau: No, it was not something I always dreamed of. I grew up in Ottawa, and both my parents were scientists. They were government scientists. My dad was a geologist working with the Geological Survey of Canada. My mother had completed medical school but was working as a toxicologist at Health and Welfare Canada. And so, I definitely had science and a love of science and a little of astronomy since I was a kid. But no, I had no concept. There were no professors in my life. I had no idea that one could work as a professor where one would teach and do research.

At each step of my professional development, I sort of just asked myself, “Is this the next thing I want to do? And this is the right step?” So, when I went to grad school the fact that I got paid made the decision easy. I really thought, “Okay, well, this is a pretty good paying job, and it has stability, and it is something I truly love.” But I wasn’t doing it because I felt like, okay, this is my next step to becoming a professor. I probably never really thought I would make it and become a professor. I don’t think I really had those thoughts until I became a postdoc. And of course, by that time, though, it seems it had become much more normalized to me that one could be a professor. So, I was a postdoc at Caltech, and again I took that job because they had offered me some research funds. I could build an observatory, and it was a fantastic institution in terms of supporting my research. But not because it was somehow strategically the next right step.

Sara Seager: Cool. So, throughout your career you’ve achieved many great things. What’s your recipe for success? Or how have you overcome the weight of expectations? Or have there been significant challenges that contributed to your recipe for success?

Dave Charbonneau: Those are great questions. I always worry that looking retrospectively I don’t create a narrative that wasn’t there. I’m not saying I would, but it’s if there’s sort of simpler narratives, and I’m sort of wondering if they’re really true.

Sara Seager: But one recipe for success out of your story, in my viewpoint, was the detailed, deep technical work. And knowing when your signal is right and what uncertainty belongs to that signal.

Dave Charbonneau: So really, really understanding the guts. So definitely one important part of success, I would say, is, you know, building your own instrument and really tuning it and understanding it and using it for this one application. Or if you’re going to use a general-purpose instrument to really understand every little detail about that instrument. And so later, I was building my own telescopes at Palomar, and then the MEarth project, which I am very, very proud of. Really, really, really understanding the details and tuning the experiment just right, certainly that’s part of the recipe for success. I think also, doing something obviously that’s new. I really have enjoyed developing the transit method, developing the transmission spectroscopy, developing secondary eclipses and mapping, developing looking for transits around small M dwarf stars.

I recognize that there were other people who were better at taking those methods and doing them many, many times. And it’s really, very important that we do them many times, because it’s only in the population that we develop this understanding. But I know now that there are other people who really excel at studying the atmospheres of 20 hot Jupiters and doing comparative studies. But for me, the development of the method was more exciting, and so I’ve always tried to do that. So, for example, a real tension I face with the JWST is that I’m incredibly interested to know the results of those atmospheric studies, but I find myself not wanting to do them. I find myself wanting to try to develop new ideas.

Sara Seager: Is this a challenge to your current career? Having invented a good fraction of the way we study exoplanet atmospheres today observationally, and now the current situation where people are just applying those methods. I’m also thinking about myself from the theory and modeling side of exoplanet atmospheres, having developed brand new ideas and so laid down the foundation, and now faced with today’s changed world. Let’s explore this tension a bit more, because if you’re good at creating new things, you’re not necessarily wanting to, or being good at just churning through lots of stuff.

Dave Charbonneau: So, there is a real benefit of tenure. I would say, my job stability allows me to go and try new things, and there isn’t FOMO. There isn’t the fear of I’m going to miss out on all the JWST work.

Sara Seager: So, you don’t have FOMO about that? I think I do have a bit of FOMO.

Dave Charbonneau: I have worked hard to remind myself why I’m doing it. Why I’m not doing it. I definitely think at moments I felt like ooohh…should I have done, should I have put more time …

Sara Seager: … should you have capitalized in your initial investment?

Dave Charbonneau: No. No, I think that these are community efforts. And if you look at how JWST observations are being done, you know there are hundreds, arguably 1,000 folks in terms of some of those very large community-led proposals and…

Sara Seager: … and too many. Sometimes, if you’re on one of these, there’s not enough work to go around.

Dave Charbonneau: Yes, and the other tension is that at the end of the day the thing I enjoy most is advising students, and I really didn’t want to advise students on those projects. I’m not saying that one shouldn’t. I’m just saying that wasn’t what I wanted to do.

Sara Seager: Well, why not?

Dave Charbonneau: Because I don’t think students would emerge with something that they really had ownership of, intellectual ownership. I felt like, if one does something that other students at top institutions can do, and they’re all interchangeable, then those people are not going to get set up to innovate.

Sara Seager: But how many new things are there to innovate? How many new things are there left to do?

Dave Charbonneau: The other challenge, of course. Now, we really need these big, powerful observatories, and those just aren’t going to be led with one professor and a couple of students the way that the TrES survey was, or the way that MEarth was. You know, we really are building these giant space telescopes that cost billions of dollars. And so, there’s a tension as well. So, I don’t think I have completely reconciled this, but I do think there are very, very interesting questions where a professor working with a really good student can do things that are really new and innovative. And it is certainly not true that the only game in town is very big enterprises like TESS and the JWST studies. And I certainly have done a lot of work. I was very involved in TESS, and I really support and believe that that was essential. But at the end of the day, I do feel this tension.

Sara Seager: Alright, so we have to be mindful on time, but we’ve missed major chunks because your career spans so much. Let’s pick two more scientific topics, then we’ll wrap up with a few personal-type questions. We definitely want to hear about MEarth.

Dave Charbonneau: So, I think the trick with MEarth was we had this ongoing successful ground-based survey called TrES.

Sara Seager: For planets transiting Sun-like stars?

Dave Charbonneau: Yes, these were small aperture telescopes for Sun-like stars. And I remember really making the decision that I wanted to develop something, just to look at M dwarf stars. And it was quite different because it was a pointed survey. So, you weren’t just looking at a field. You had to build telescopes that could look rapidly at many different stars, and you needed lots and lots of telescopes. But it was really the decision that I had to turn off TrES, so that I would have enough time to make MEarth work.

Sara Seager: But couldn’t Tim Brown or others keep TrES going?

Dave Charbonneau: No. Well, maybe, but I didn’t want to do that, and I had built one of the nodes and TrES means three. And what is TrES without each of the nodes? So, no. And also, I felt like, once you have a few hot Jupiters or ten hot Jupiters, I didn’t want to find the 50^th or the 100^th hot Jupiter, because other people were building better instruments to do that.

I wanted to find the first true planets transiting M dwarfs. Anyway, so important in terms of life lessons, knowing when to turn the project off. If the project is successful and it’s getting grant funding. That’s tough. That’s really tough to say I am going to turn this off, even though by many external metrics it’s doing great. It’s not like turning something off when you can’t get funding for 5 years, that’s a different question.

Sara Seager: But here we’re in your office right now at 60 Garden Street [Cambridge USA]. What is this telescope in your office?

Dave Charbonneau: This is Sleuth, which was one of the nodes of TrES, and that was the one that I built at Palomar.

Sara Seager: Nice.

Dave Charbonneau: Many people think I keep it as a memento to great success. But actually, the reason I keep that is a reminder to myself about how not to build projects, because I built it on the cheap. I was very cheap in terms of hardware, and so I spent all of my time driving back and forth to Palomar to fix it.

And of course, labor is much more expensive than hardware. So then, when we built MEarth I bought the very best that I could afford, the very best telescopes, the best cameras, the best mounts. So, it’s really kept as a memento about experimental design, but not the particular success of finding some of the first transiting objects.

Sara Seager: I love it. So, give us the top-level summary of MEarth and what in your mind you’re proudest of what MEarth has done.

Dave Charbonneau: MEarth had 8 telescopes in Arizona, and then later, 8 more in Chile. I really want to highlight the Packard fellowship. I think that’s tremendous that basically was an injection of cash. I didn’t slowly spend Packard over the five years that you can. I spent it all in the first six months to buy all the hardware for MEarth. I knew if I built the hardware, the National Science Foundation would probably support the operations, but they wouldn’t actually give the hardware, because it was too crazy, too innovative, too different.

I’m very proud that we found a number of planets, rocky planets transiting very small nearby stars. So, these are the most spectroscopically accessible terrestrial planets. They were all Cycle 1 JWST targets, every single planet.

Sara Seager: Tell us the planet names.

Dave Charbonneau: The first one we found was GJ 1214 b, and that one’s different, because it actually does appear to have a hydrogen envelope. And then, GJ 1132 b was a hot, rocky world. Then LHS 1140 b, which is my favorite, because that’s I think one that is a habitable zone terrestrial planet that might retain its atmosphere. And then later, another planet in that system, LHS 1140 c.

And then the other big result from MEarth that I’m very proud of is in stellar physics. We learned that most nearby, low-mass stars spin very, very slowly, and are very inactive, and at the time the wisdom was that they were all rapidly rotating and active. But it was just that the active ones were much easier to find. But actually, most of them are very slow, and that means that they emit very little amounts of X-rays and UV light. And so, their planet atmospheres should now survive if they survive the initial tumultuous period. So, a very different picture for the radiation environment for these rocky worlds.

But then I turned off MEarth a year or two ago. With the same idea that, if I want to do new things, I really have to take a sober look and say …

Sara Seager: I did not know MEarth had been turned off. What is going to happen to all the telescopes?

Dave Charbonneau: Well, that’s a great question. That’s something we’re still figuring out. But you know, definitely, with the success of TESS and with a decade of MEarth observations the question is not what is MEarth’s success in the past? It is, what is the discovery space in 2024, and is it better to take my time and energy and put it into different things?

Sara Seager: Well, what are you most excited about in the future of exoplanets?

Dave Charbonneau: I think some of the big immediate questions are, do terrestrial planets orbiting M dwarfs have atmospheres? We have studied a few of them, and we know that the hottest ones, the ones that are the least likely to have atmospheres, do not have atmospheres. We can see that. Do any of them have atmospheres? Or have they all been stripped of their atmospheres because of this very prolonged period of a very active radiation environment. That’s exciting. That’s timely. That’s a profound question in terms of what do we do next.

I am very, very excited to try to study ancient planets. So, I think with Gaia we now understand that there are an enormous number of stars in the Galaxy that are thick disk stars. These are ancient stars. They’re probably 8 to 10 billion years old. They’re mixed in among us. They walk among us, but we don’t recognize them, because the M dwarfs of course don’t evolve. But those planets were born in a time when the Galaxy had a very different amount of iron and magnesium, because they’ve been enriched with different supernovae. So, seeing planets, which have different core mass fractions and the dynamics of those planets, I think that would be a whole new field of study.

Sara Seager: Wow. And are any of these stars bright enough for us to do studies on the planet atmosphere to get mass–radius measurements?

Dave Charbonneau: Yes, and nobody is doing this.

Sara Seager: This is a great new idea.

Dave Charbonneau: The planets would have a different core mass fraction. So, the dynamics of the planet would be totally different. The cooling history be different, the planet’s ability to outgas an atmosphere would be different and of course we would see planets that have undergone 10 billion years of evolution.

And then another idea is definitely to go after satellites, to go after moons. If you look at the Solar System, so much of what we know about the actual history of the formation of the Solar System comes from small bodies. If we didn’t know about the moons in the Solar System, I think we’d really get it wrong.

Sara Seager: Too many ideas, too little time.

Dave Charbonneau: Well, there’s a guy named David Kipping who has pushed the moon idea for a long time, so I really wanted to make sure to mention his name. I think now, we can actually look for moons that really are the same size as the moons that are in the Solar System, and not just hope for moons that are much, much bigger. So, these are the projects, the directions I want to go.

Sara Seager: Have you had any major setbacks? And if so, how have you overcome them?

Dave Charbonneau: That’s always a tough question. Yes … I mean no, no, I feel very blessed. I definitely don’t feel like I’ve had major setbacks.

I definitely feel, though at the time I had moments of profound doubt. Okay, so after I’d been looking for reflected light for almost two years, and I didn’t find it, and I saw these great discoveries in other fields of astronomy, I really thought maybe this wasn’t going to work out, and I was not going to advance. Certainly, I did not have any papers. All my classmates had lots of papers. I didn’t have any papers, and it was the third year of graduate school.

Sara Seager: Yes, I remember the lack of papers for me too.

Dave Charbonneau: So, when I think about when I went to Caltech, I’m very, very appreciative of the resources they gave me. A challenge was that some of the faculty at Caltech were combative and not supportive, although some of them were very supportive. But I definitely think that some individuals were not operating in my best interests. And then also, when I got married, and I was trying to live in the same city as my wife, it was a challenge and I was really worried that we wouldn’t get to live together. And we didn’t want to do that. And in the end that worked out. So, I don’t think I’ve had a setback that derailed me, because here I am. But I just want to point out that I definitely had moments of real doubt and worry for sure.

Sara Seager: Well, thanks for sharing that. Despite the doubts, it’s truly remarkable to go through a career without a major setback of some kind. I often wonder if anyone makes it through life unscathed.

So let me ask a harder question, if you feel comfortable talking about this. I understand you have a very major setback at the current time.

Dave Charbonneau: Yes, well, definitely. In late July of 2022, two years ago, I was diagnosed completely out of the blue with stage IV colon cancer. That’s very, very serious. So definitely here I am two years later. The two-year survival rate is 37%. So, it’s already unlikely that I would have been here, and so definitely, I have learned to stay very positive. There are some very specific things I do to stay very positive. And I really am not worried now about the cancer. I really enjoy time with my students doing research, and especially time with my family. My daughter just graduated high school. When I first was diagnosed, I thought statistically I wouldn’t even get to see that. So, I feel very, very fortunate for the time I have, and I think it makes me very intentional about each day.

Sara Seager: I just want to say it takes a lot of courage to face the unknown.

Dave Charbonneau: Yes, and it’s also very surreal. I mean, it’s very surreal to have this illness where it’s Stage IV.

Sara Seager: In case anyone doesn’t know what Stage IV cancer is, please explain.

Dave Charbonneau: Stage IV means the cancer has spread to other parts of the body. The cases of people who have been cured of this are anecdotal at best. But I do everything the doctors recommend, and I have a lot of good days when I’m not under the haze of chemo, or radiation, or recovery from surgery, and I really enjoy the time I have.

Sara Seager: Well, I’m really proud of you for being so bold and brave and courageous and positive under this extreme circumstance.

Dave Charbonneau: Well, thanks, I appreciate that. We have known each other a long time.

Sara Seager: Let’s finish off with some exciting news. You and I, we are to share the 2024 Kavli Prize in Astrophysics for our early work on exoplanets.

Dave Charbonneau: Yes, [laughs]. Definitely for me, it seems the events of the past year are almost fictional nature with this very serious illness, and then to be able to share this incredibly prominent prize with a good friend and colleague, with you, Sara. So, it really brings great joy to do this. And I’m really looking forward to the Kavli award events in the fall.

Sara Seager: Same. Well, could you share a final piece of advice that you’d give to your younger self or a student?

Dave Charbonneau: Yes, I think it is very important to ask yourself before you do a project, could this result in a major discovery? I think what we all do in astronomy with most of our time is incremental. We’re all doing it. We’re all writing papers, which are small advances and there’s nothing wrong with that. But I worry sometimes students are asked to do projects, where even if things go perfectly, the outcome can only be incremental. And I think it is very important to not do a project unless it has the potential, even if it’s unlikely, that something really wonderful and magical can emerge from it. And that is a good question to ask oneself before one starts any endeavor.

Sara Seager: Very well said, thank you. And, thank you for a wonderful interview.

Dave Charbonneau: My pleasure.