Reading Starlight: In Conversation with Michelle Kunimoto
The Blue Hour: Reading Starlight and Inferring Other Worlds
My second interview of the spring season of The Blue Hour is now available as a podcast.
In this conversation, I speak with Michelle Kunimoto about exoplanets, planetary populations, the Kepler and TESS missions, and the rapidly evolving science of worlds beyond our solar system.
CiTR 101.9 FM / The Blue Hour
Recorded live on May 19, 2026
Listen to the interview on CiTR
The Blue Hour airs live every Tuesday at 2 p.m. on CiTR 101.9 FM and citr.ca.
Transcript lightly edited for clarity while preserving the natural rhythm of the live conversation.
Transcript
Farha Guerrero: For most of my lifetime —— when we looked at the night sky, and certainly when I was young, I believed we could only understand was what was in our solar system. I remember looking at books and seeing the names of the planets that became very familiar, and of course our beautiful star, the Sun. But in the last 30 years, a lot has changed.
The first confirmed exoplanet around a sun-like star was announced around the mid-90s. And since then, many scientists around the world collaborating together have confirmed the discovery of exoplanets in the thousands, as well as many thousands of candidates, which we’ll talk about.
So within a single human lifetime, astronomy has moved to worlds really beyond our solar system, and that’s extraordinary. So welcome, Michelle, to the show, and thank you so much for coming today.
Michelle Kunimoto: Yeah, thank you so much for having me.
Farha Guerrero: Can you define for us the difference between our solar system and what astronomers mean by an exoplanetary system?
Michelle Kunimoto: Right, so an exoplanet is really a planet that is outside the solar system and going to be orbiting another star that looks a lot like our Sun.
There’s actually some maybe disagreements within our own community about what exactly you have to be in order to be an exoplanet. For example, there are some objects we’ve found that look very much like planets in our own solar system — they’re planetary mass objects — but they don’t orbit any star at all. So because they’re not in a system that looks just like our own, does that make them an exoplanet or some other class of object?
So there’s a lot of definition debates that we have, but most simply, it’s going to be like a planet that’s outside our solar system.
Farha Guerrero: So when we look up at the night sky with our naked eyes, we can sometimes see planets in our own solar system, right? Like Venus, Mars, Jupiter orbiting our own star. But the stars themselves that you are studying are outside of this solar system entirely. But we’re also, with our naked eyes, looking at stars.
So it is interesting — we are seeing them but we’re not seeing them — and these planets are orbiting those stars, and that’s what you are out to find. Am I right?
Michelle Kunimoto: Right. So the next time you go out — and I know in Vancouver it can be very overcast and rainy — but if you see the night sky, you can just think about every single one of those stars probably has a planetary system around it that looks very much like our own.
Now, in terms of whether our solar system appears to be an outlier — like how similar the exoplanets we’ve found are in terms of their properties compared to the planets in our solar system — this is something we’re still trying to understand. Is our solar system unique? Is it particularly rare? Or is it a really common outcome of planet formation?
So a lot of the planets that we’ve found so far tend to be a bit closer to their star than many of the planets in our solar system, but that’s really a result of the limitations of surveys right now. So the longer we continue to look at the sky, the better we’ll have the ability to find other systems that look just like our solar system.
Farha Guerrero: Now, I brought something personal into the studio today, which I quickly showed you. It was a handwritten science question from my older son from 2015 where, when he was in potentially Grade 3, the question was: “How similar to Earth is Kepler-29?”
And to me, this speaks to something about how quickly exoplanets have entered the public imagination during what we could call maybe the Kepler era that he was born into.
Now, did the Kepler mission fundamentally change not only astronomy, but how ordinary people — even children — imagine the universe?
Michelle Kunimoto: Yeah, absolutely. And this is one of the great legacies of the Kepler mission, it's doing exactly that.
So imagine pre-2009, pre-Kepler mission. A lot of the exoplanets that we’d found so far were all really big planets, like Jupiter-sized planets. And many of them, somewhat surprisingly, orbited really close to their stars — they orbited within a few days.
Now compare that to our solar system. Mercury is our innermost planet, and it has an orbital period of, I think, about 89 days. So these are fundamentally different from the planets that we’re familiar with.
And the big open question was: where are all the other Earths? We weren’t finding any Earth-sized planets in year-long orbits. And the answer to that is, we don’t actually know, because we don’t have the technology or the missions to find them.
So Kepler was designed to be the first exoplanet-finding mission with the capability to find other Earths, and in doing so, it completely revolutionized how we saw exoplanet science. Because before we had, about 90% of all the planets were bigger than Neptune, and now we think about 90% of all planets are smaller than Neptune.
So it just completely turned the world on its head in terms of exoplanet science, and we now believe that other Earths are some of the most common planets that we might be able to find. Farha Guerrero: So this is a mission started in 2009. It’s the Kepler space telescope, but it lasted a few years?
Michelle Kunimoto: About four years.
Farha Guerrero: And there were some failures on the equipment side of things. Is that right?
Michelle Kunimoto: Right. So essentially it has these four little what are called reaction wheels that are what control the telescope’s pointing so that it can always look at the same section of the sky to look for exoplanets. And then one of them unfortunately broke, a second one broke not long after, and you need at least three to really be able to point stably.
So the mission ended early. It was actually supposed to last for several more years, but even with the four years of data we have now, people are still finding new things in it, and it’s been an incredibly valuable resource.
Farha Guerrero: Now I have a guest here in the studio. His name is Isa. He’s come from Whistler Secondary School. He’s missing school to be here in the studio and he’s signalling me to ask a question.
But I will just say: this is a young man that was born in 2009.
Isa: Yes. My question is, did Kepler stay in Earth’s orbit while it was studying the stars, or did it go somewhere farther?
Michelle Kunimoto: Great question. So Kepler was launched into what we call an Earth-trailing orbit. So what that means is it was basically launched just behind the Earth.
So Earth orbits around the Sun every 365 days, and Kepler, I believe the period of Kepler was 375 days, so it was really close to the Earth. It’s always following Earth.
It is still there. It hasn’t gone anywhere else. It was officially shut down a few years ago — but it is still in an Earth-trailing orbit.
Farha Guerrero: And you’ve said that the Kepler telescope had very high precision, so it took really good quality images. We’ll also later talk about, TESS, which is a little bit different, more about quantity.
But let’s stick with Kepler a bit, because Kepler not only monitored roughly — was it 150,000 stars, right? So tiny changes in brightness every 30 minutes over those four years. It’s revolutionary, [because] we went from, if I’m not mistaken, from just a few exoplanets, and then Kepler comes and [finds] thousands.
And four of which, you had discovered, when you were doing your undergrad, and then you discovered many more potential candidates during your PhD — many now since confirmed, I’m sure. We’ll talk about that process, but, it certainly was an extraordinary mission, in your opinion?
Michelle Kunimoto: Yeah, absolutely. Overall, Kepler found 4,700 planets and planet candidates. Planets being things that we believe are confirmed planets. Candidate planets are those that we think have a very high likelihood of being planets, but there might be something about them that we’re still not completely sure, we can rule out something else, like maybe it’s caused by a star or just the star itself getting variable randomly.
So I think Kepler’s confirmed over 2,000 planets at this point, which is about a third of all of the planets that have been confirmed from every mission across all 30 years of searching.
Farha Guerrero: So even though more than a decade has passed, the legacy of Kepler is still very much important, especially to your research, as I understand.
So the incredible archive that Kepler has given scientists like you is still being used and reevaluated and understood and being poked at constantly. So it is, without doubt, really extraordinary.
But now let’s talk about something that you work a lot on, and it’s something called demographics. It’s, I believe, an attempt to understand planetary populations statistically.
But let’s bring this word demographics into the astronomy world. What does it actually mean?
Michelle Kunimoto: Right, so demographics is basically a population study, very much like how you can do a demographic study of people.
Like let’s say if you are testing how often people of different races or sexes or education levels — any different dimension — you can kind of see how that changes over time in a various city.
So what I’m doing is a population study to understand how common different types of planets are in our galaxy.
And one of the things that we can do most simply is look at the total number of planets we’ve found so far. And we’ll notice we have a lot of really big planets. We have a lot of small planets. But it’s not as simple as just saying,, 50% of planets are bigger than Jupiter, therefore 50% of planets in our galaxy are bigger than Jupiter, because some planets are a lot easier to find than others, which means they’ll be overrepresented in our data set.
An analogy would be if you go to, let’s say, a sports game and you ask, “Are you a sports fan?” you’re going to have a lot of people say yes I am, but that’s an overrepresentation in your population.
So you have to be really careful about thinking about what types of planets are easier to find than others, and how might that affect our understanding of the underlying planet population.
So a lot of what I do with the statistics is try to understand those biases — and understand what planets actually look like in our galaxy.
Farha Guerrero: And biases is really the word because it’s this idea in your work that exoplanet data sets are intrinsically biased, like the example you gave.
Can you get into the kinds of biases that are really specific to, say, what you found in the Kepler data set?
Michelle Kunimoto: Sure. To understand biases, we first have to understand how do we actually find exoplanets with something like the Kepler telescope.
So what Kepler did is what is called the transit method. And this is actually the method that I continue to use, even at University of British Columbia, to find exoplanets.
And the idea behind the transit method is, as you’re looking at a star, if there’s a planet in orbit around that star and that planet happens to pass directly in front of the star as it orbits, some of the light will get blocked from reaching us.
So if we’re measuring the star’s brightness over time, we’ll see just for a few hours the brightness of that star will decrease because something is simply blocking the light.
So you can imagine that the shorter the period — so the smaller the orbit of a planet — the more frequently these dips are going to happen, and they’re going to be really easy to spot.
Whereas something like the Earth, it’s only going to transit once every year. So it’s going to be a lot harder to see that than something like a three-day planet.
A bigger planet is going to block a lot of light, whereas a tiny planet like the Earth will block only a little bit of light. So big planets are going to be a lot easier to see.
So essentially the transit method and missions like Kepler and TESS are biased because it’s easier to see these really big planets in really short-period orbits than the really tiny long-period planets like our Earth.
Farha Guerrero: They’re often called hot Jupiters.
Michelle Kunimoto: Exactly, yeah — the hot Jupiters. So those are kind of some of the earliest classes of planets we found just because they’re so easy to find compared to small ones like the Earth.
And yes, they’re fascinating because when you look at our solar system, all of the giants are super far away, and we designed all of our planet formation theories off of that configuration, and then suddenly these planets show up and we didn’t even think that they could possibly exist.
And even today, we still have open questions about how they formed.
Farha Guerrero: And one thing that I learned was that in this kind of detection, sometimes asteroids would get in the way and they could give you false positives.
Can you give us a picture of what that actually might even look like?
Michelle Kunimoto: Right, so different detection methods have different kinds of false positives, so things that mimic the sign of a planet but aren’t actually a planet.
For the transit method, probably the most problematic one is that it’s not actually a planet, but it’s another star entirely. So if a star is passing in front of the target you’re looking at, it will block some of the star’s light and cause something that might look a lot like a planet transit.
So that’s an example. As I mentioned before about how stars are just intrinsically variable, that means that they naturally get brighter or fainter with time. And so if this happens periodically, it could look a lot like a transit, just causing the star to get fainter every now and then.
The asteroids are an issue for a different method called microlensing, which I’ve also used for my work.
Farha Guerrero: And then there’s also other things, right? Cosmic rays and other stellar activity that could affect?
Michelle Kunimoto: There’s a lot.
Farha Guerrero: I want to get into now something that is really quite interesting because you have this distinction between a planet candidate and an actually confirmed exoplanet.
So I can imagine there’s a whole process — other people who are also part of making the confirmation.
Can you speak about how this process actually works? So if someone like you finds an exoplanet, what happens after the discovery? And how does the planet actually get confirmed?
Michelle Kunimoto: Yeah, so there’s a few ways you can do this, but for transiting planets, the most powerful way is to follow it up with a different technique called the radial velocity follow-up.
And what this technique gives you is essentially the mass of the object that’s causing these transits.
So if you have a radius from the transit method — right, you can measure the radius of the object because of how much light is being blocked.
Especially once you get to something about eight times the size of the Earth, that’s the regime where it could be a planet — a gas giant planet — or it could be a brown dwarf, or it could be a low-mass star. They all have the same radius, but they’ll have very, very different masses.
And so if we measure the mass with the radial velocity method and we find that it’s planetary, we can confirm it as a real planet.
Isa: I was just born when Kepler was launched — but I do remember, I think it was either Kepler-29 or Kepler-22b that got very famous online for being considered to maybe contain certain life.
And I remember it was disproven, that they found that it wasn’t similar to our Earth. I don’t know if you know anything about it. I just remember it got a lot of attention online— that it was more like a false promise.
Michelle Kunimoto: I can’t say I remember the story of Kepler-29 explicitly.
There was a probable planet from Kepler that looks most like the Earth in terms of its basic properties — Kepler-452b.
This is a planet that’s about 10% larger than the size of Earth. It orbits a star that is the exact same type of star as our Sun — a G2 dwarf star — and its orbit is 385 days instead of the Earth’s 365.
So based off all of these basic properties, we consider it part of landing in the star’s habitable zone. So this is the range of distances from a star where a planet could potentially support liquid water on its surface. It’s just the right temperature for liquid water, which we obviously think is really important for life on Earth.
But you did touch on how there was some controversy. So this particular object was reanalyzed later after Kepler had considered it a confirmed planet, and astronomers were worried that the transits were not actually caused by another astrophysical object, but rather were just noise in the data.
So one of the issues with finding these tiny planets is that their transit depths are so small, it can be really hard to pick them out compared to other events in the light curve.
And this was a case where people cast doubt on it being an actual planet.
So even today, I think we still consider it part of our confirmed planets. If you look at the total number of about 6,300 planets at this point, it is still listed there.
But it gets into the challenge of being able to be so certain when you’re looking for these and trying to confirm them.
Farha Guerrero: And so when does the rubber stamp actually happen? The names of planets are interesting to someone like me who grew up with very concrete names of planets. What is the rhyme and reason for the naming? And then when do they get that rubber stamp? When are they added to the table to make them official?
Michelle Kunimoto: So basically the rubber stamp approval is somebody writes a paper that goes through peer review, and the referee for your paper reviews it and says, yes, you’ve given a convincing enough argument to say that this is a confirmed planet.
And once it’s in the published domain, then the NASA Exoplanet Archive has people that are scouring publications all day and add it as soon as they see a new paper has announced the discovery.
So that’s pretty much how these planets get added.
The names of the planets are based off of the star that they were observed around, and then they get a little letter right after to indicate which order the planet was discovered in.
So for Kepler-452b, that’s an example of a planet that was found around the star Kepler-452, and the “b” indicates it was the first planet.
In the case of multiple planets, we try to list them by increasing orbital distance, so “b” would be the innermost planet, “c” would be the second planet, “d,” etcetera.
Although sometimes we might find a planet — let’s say we call this one Kepler-76b — and then we find another planet that’s even inner to that one. Now the letters are all mixed up. We don’t rename them, we just kind of live with the fact that the names are a little bit chaotic.
But essentially we just do the best we can.
Farha Guerrero: That’s great. And so in a timeline, when would it finally be — depending on all these variables? And how many people are working on that kind of confirmation?
Michelle Kunimoto: It’s not uncommon for these papers to have 60 people on them. It’s a lot of people.
Because let’s go back to the confirmation we’ve kind of established. We want to measure the mass of this planet, which means we need to get data using the radial velocity technique.
Now in order to write a telescope proposal to do that, we need to have enough evidence beforehand before we ask for that final step of the confirmation process, to make sure that we’ve ruled out everything else possible.
So for example, when we’re finding a planet around a star, there’s a chance that it doesn’t even orbit the star that we expect. It’s on a nearby star that was close enough that when we were measuring the brightness of the star, it basically was contaminating the measurements we were making.
And so one of the very first steps is to just confirm: do we see this planet transiting the star that we expect, or is it on something else?
Some stars are not amenable to doing radial velocity follow-up. There might be some issues with them — like maybe they’re in a binary star system. There’s a lot of reasons.
So we typically do what’s called reconnaissance to make sure that the star is even ready to propose this data.
One person has to write the proposal, one person is taking the observations, a whole team is analyzing the data. You have all these teams internationally working together to do this.
And so it ends up being about 60 people on average. That’s my anecdotal experience from writing several of these papers at this point.
Farha Guerrero: That’s amazing, and that the ecosystem of that kind of collaboration is real, and it goes around the world, and there are no borders, which I think is really cool.
I want to talk about TESS because TESS is on right now as we speak.
But let’s just take a little diversion because this was very interesting to me — that you’re also understanding how planets form, how they migrate, how they lose atmospheres, how they evolve over time.
So how much can astronomers actually infer about planetary evolution from this type of population-level data?
Michelle Kunimoto: Yeah, that’s a great question, and it’s one of the things that fascinates me so much about doing demographics.
Because what we can do is look at the properties of the planets that we see. We can see how that maybe changes with time. So we can compare populations of young planets versus old planets.
And if we see there’s a significant difference, then that means there’s some kind of evolutionary process that is causing these changes.
And then there’s typically a lot of collaboration between observers like myself who are taking these observations, finding planets, understanding their properties, and then theorists who come up with the theories to explain why we see all of these trends.
And typically we try to collaborate so that if the theorists come up with an explanation, either from the planet formation stage or subsequent evolution, that can explain our trends, we might be able to test those theories with the observations.
Farha Guerrero: That’s fascinating. And then you are teetering on even other areas of research that are beyond the physics and astronomy department. It feels like that to me, at least from an evolutionary side of things, right? Because you’re entering other worlds in some way by getting this data.
But let’s talk about TESS.
The Transiting Exoplanet Survey Satellite — it was launched in 2018 — and its approaches to exoplanet detection are different from Kepler.
And it looks like TESS really is about scale and surveying enormous regions of the sky and observing millions of stars.
What can you tell us about TESS? Because you had a relationship with it at Massachusetts Institute of Technology.
Michelle Kunimoto: Yeah, so I worked on the TESS mission as a postdoctoral associate and then later as a postdoctoral fellow, and this was what I worked on right out of my PhD at UBC.
So for my PhD I had focused on finding new planet candidates with the Kepler mission, and TESS was hiring for someone with exactly my skill set to basically do the same thing with TESS, and I was very excited.
So the TESS mission is based out of MIT. That’s where the principal investigator lives, and it collaborates — it’s a NASA-funded mission. So I worked with a lot of MIT researchers as well as NASA researchers on this mission.
So TESS, as you mentioned, kind of took a very different approach to planet finding.
Kepler was really interested in pinning down the demographics of small planets, so that’s why it had to look for so long. It looked for four years at the exact same stars to try to see just a few transits from Earth-sized objects in year-long orbits.
Whereas TESS was much more about: let’s do an all-sky survey. Let’s survey all the nearest and brightest stars.
And the reason for that choice was because it’s the nearest and brightest stars that are most amenable to that radial velocity follow-up.
So a lot of the Kepler stars are very faint. We’re never actually going to be able to measure a mass for a lot of these planets and therefore confirm them.
Whereas TESS, by searching for planets around really bright stars, they’re a lot easier to follow up and confirm, and potentially follow them up with things like the James Webb Space Telescope to do atmospheric characterization and all sorts of other science.
Continue to Part II — Reading Starlight: In Conversation with Michelle Kunimoto.