How Do Scientists Make Discoveries?

Transcript

ERIC: From the Museum of Science in Boston, this is Pulsar, a podcast where we discover answers to the most common questions we've ever gotten from our visitors. I'm your host, Eric. And sometimes the questions that were asked in our exhibit halls are very general and high level such as: How do scientists make discoveries? It turns out that a straightforward path from question to experiment to answer is pretty rare. And most discoveries happen unexpectedly. This was the case for Dr. Kishalay De, an astrophysicist at MIT, whose recent discovery was the first of its kind. He came to the museum to share his research with our visitors, and I was able to ask him about the process of making this unique discovery. Kishalay, thank you for joining me on our podcast.

KISHALAY: Thank you so much for having me.

ERIC: Our question today is, how do scientists make discoveries? And I love your story of the discovery of the paper you just came out with, because it's not what you were looking for. You started out looking for something else. So why don't we start from the beginning. What was, in this discovery, your aim? What were you looking for?

KISHALAY: Right, so this discovery was totally accidental. I mean, I'll say it out loud up front. What I was looking for, I work on binary stars, and a lot of binary stars, they undergo explosions during their lives. Some of these explosions, they're called novae, a dead remnant of a star like the Sun, which is what we call a white dwarf, the white dwarf is slowly pulling away matter from its companion because it has such a strong gravitational field. And as this matter accumulates on the surface of the star, every now and then it becomes unstable, and essentially explodes like a nuclear bomb on the surface of the star. And that's what powers this brightening that we see in novae. And you can confirm whether it's a novae or not by getting more data on the stars to confirm whether you see the chemical signatures that you'd expect from these novae. There are about four dozen novae that go off in our own galaxy every year. And the way you find novae is that you keep taking images of the sky every night. And every now and then you'll see a star that brightens by a factor of thousands, sometimes millions over the course of a week or so. And when you find a star that changes brightness like that, we get more data on those stars with bigger telescopes to try and confirm whether it's a nova.

ERIC: So the first idea is you look at a wide area of the sky. It doesn't make sense to look at one star at a time. And then when you see that brightening, then you say: quick, let's look at that one closer.

KISHALAY: Yes, yes. So we keep taking images of pretty much the entire sky every single night just looking for the stars to brighten. And it was during one of these nights when I was looking through data taken using the Zwicky Transient Facility Survey, which is a wide field camera at Palomar Observatory in California. When I noticed this one particular star that sort of seemed like a nova. It seemed to be brightening very fast. It brightened by a factor of a few hundred over the course of a couple of weeks. And my first suspicion was, okay, this is another nova, we should go get more data on this source.

ERIC: So you're looking for the stars getting much brighter, because like you said, there's a huge amount of energy coming in from the partner star. You see them fairly frequently. But you found one that didn't look quite normal, you can tell, like you said, with the chemical signature, what they're supposed to look like. And this one didn't look like that.

KISHALAY: Yes. So as you would imagine, if you ignite a nuclear bomb on the star, it should really heat up the gas around it, should be extremely hot gas, hundreds of thousands of Kelvin. And when we looked at this particular star, we got more data, there was absolutely no sign of hot gas on this star. It appeared to be cold gas all around it. For some reason the star had brightened, but it didn't have any hot gas around it. So that's really what started the detective story: why is the star brightening, but does not have any hot gas around it? It's not a nova, for sure. But what else is it?

ERIC: So you have kind of like a mystery. So at that point, you can come up with an idea of what it is. Do you need more observations? Like, what's the next step?

KISHALAY: When you see something that's surrounded by cold gas, you sort of go back to the basics and ask: what does cold gas do in the universe. And the first thing that comes to your mind is that cold gas emits light in the infrared wavebands, which is not light that our eyes can see. These are longer wavelengths that our eyes can't see. And in fact, most of it doesn't even come to the Earth's surface because the atmosphere is opaque to it. When you see cold gas, the next step is can we get infrared observations to see if is this shining in the infrared bands. And what does that tell us about where this cold gas is, what its temperature is and so on. So that was our next step. We got infrared data from the ground, whatever it is possible from ground based telescopes, and also looked at space-based telescopes that NASA has been operating for the last few years to try to understand what this cold gas was doing.

ERIC: So is this like, you call up those observatories and say, hey, quick, can you point at this star? Are you looking through data that's already existing, because they are the kind of those wide fields looking at the whole sky? How does it happen?

KISHALAY: A combination of both. So the ground-based data, because, you know, we are on the ground, it's actually very easy. To get more data, I can just use the telescopes on the ground and just point at this part of the sky, we don't even need wide-field cameras because we know where it is in the sky, we found it. So the process is to just take a ground-based telescope and point it at that part of the sky to get the infrared observations from the ground. In the case of the space-based observations, it's usually a lot harder to coordinate and request observations. But we were lucky in that in this particular case. NASA's WISE mission was actually just looking at that part of the sky as part of a survey. So we didn't even have to call anyone. The data was out there in the public for anyone to look at. And that's what we did in this particular case.

ERIC: That's great when it works out like that. So what did that data tell you? What was next?

KISHALAY: So what it told us was that in addition to the cold gas that we were seeing in the chemical signatures, this thing was slowly producing a ton of dust around it. Usually, when you see dust around a system like this, what that comes from is that you have some sort of a system that is ejecting a bunch of material that was initially in the binary, but it's now being essentially lost to space, because of whatever the binary is trying to do. And when you sort of take a step back and ask, okay, if I put all of these things just together, cold gas, dust, sudden brightening, what does it look like? And the only things that we've seen do this in the past are these phenomena that we call stellar mergers. Stellar mergers have nothing to do with novae. You get stellar mergers when you have two stars that were born as a binary, they were born together. But at some point in their life, one of the stars decided to die, they ran out of fuel in its core, and it starts to expand. And as it does that, it engulfs its companion during that process, and that's what produces the brightening of the star. We know of dozens of these stellar images that we've seen on our own galaxy and in other galaxies. So that was the first piece that we tried to figure out.

ERIC: Stellar merger is, like, the most boring name for the most epic thing in all of science. It's two stars circling and then colliding, and one consumes the other. And stellar merger sounds like a bank transaction.

KISHALAY: It does. Yes, it is spectacular, they completely changed the lives of stars. And they're spectacular in terms of how they behave in the sky as well.

ERIC: So you thought that might be what you were seeing? Did you need more data to figure out whether it was that or not?

KISHALAY: The signatures of this event all smelled like a stellar merger, they all consistent with that. But there was one piece that really struck us and kept us confused for half a month, which is that, regardless of what kind of stellar merger you invoke, in this case, it turns out that it just wasn't bright enough. It was, compared to any other stellar merger that we had seen in the past, it was something like a factor of a thousand fainter than anything we had seen. So then you ask, how do I keep the properties of this eruption basically the same as a stellar merger, but just tune down the energy? What knob can I turn? Maybe if you have a star that's not running into another star that next to it, maybe there's something that's a thousand times less massive than the star? And if you ask what is a thousand times less massive than the star? Jupiter. Jupiter, it is about a thousandth of the mass of the Sun. So that's when, really, the first connection started to become apparent, which is that maybe what it was trying to consume, as it was dying was not another star, but a planet that was next to it very much like Jupiter.

ERIC: You thought we might have just seen a star consume and destroy a planet in real time, which has never happened before. We never observed that. How do you go about saying whether that's actually what you saw? Could you confirm it any other way?

KISHALAY: Yes. So that's really started off the months of more physical modeling of trying to understand, you know, this is something that's been predicted to happen in the universe for decades, since the first days, we've known how stars live their lives, we know that this must happen in the universe. The technology never existed in the past, for us to be able to get these observational signatures until now. And that's why you know, we are living in a really exciting era for astronomy. What came after this was really taking all of the data that we had at hand and trying to make a physically consistent model for what would the mass of the planet have to be? What would the mass of the star have to be? How do we explain all of this data together? So it was really not a case of getting more data, it was more of how do we make it physically consistent with this picture? And what does that tell us about the planet and the star?

ERIC: So you're able to kind of come up with a way to explain the observations each step and try to say, okay, we have new observations that don't fit with our explanation, what do we have to change and tweak about it? And then eventually getting towards something that explains everything.

KISHALAY: Yes, exactly. In this case, because it's a phenomenon that's been predicted to occur for many decades, scientists had models for what this might look like. But these observations require you to revise your models because you know, you the universe always works in more creative ways than you can ever imagine. So the way to make things consistent, you have to sometimes tweak your model and say, what else could be different that helps me match these observations. And that's really where the beautiful physics comes into play.

ERIC: I love the sheer different amount of observations, different types of observations. I've been thinking about how we've never had that technology, like you said, it's great that we have all that at our fingertips to be able to use.

KISHALAY: Yes, suddenly, I think one the key pieces of evidence in this particular case was that we had infrared data available. Astronomy started in the optical bands. So it's not surprising that we've made the furthest progress in optical astronomy in the last century. But infrared astronomy is hard. In fact, infrared astronomy didn't even exist 30 years ago, it's a new field. And part of that is because infrared observations, you can't most of them, you can't do it from the ground, because the atmosphere is opaque to infrared light. And even the detectors for infrared astronomy are much more expensive than optical detectors. So it's very hard to create a camera that is sensitive to infrared light. And it's really these technological drivers in the last decade or so that have allowed us to make make these connections make these discoveries that are possible today.

ERIC: My favorite part about all of this is the model that came out and listeners, you have to go and actually watch this because we think of stellar things, the Galaxy, a solar system evolution, stars being born, they take much longer than a human lifetime. But this process that you observed, happened over the course of days. I mean, like you said, the brightening goes up by a thousand. Very quickly, can you paint a picture of what it looks like in that last little bit with the planet getting sucked into the star? I always kind of assumed that when stars expand and they've reached the end of their lives and a planet goes in, it was like a bug hitting a zapper. And it just doesn't affect the star at all. But it completely changed the star. Can you talk about what it is that happens?

KISHALAY: So the planet that we think got consumed was something that looked like Jupiter, around the mass of Jupiter. But unlike Jupiter, which is very far out in our solar system, this Jupiter was in a very short orbit around its star. The orbital period, the time it takes to go around the star, was probably less than a day for sure, at its very last moments. And what happens in this case is that the star is slowly expanding as it's beginning to die. And the first thing that happens is that the planet begins to feel the outer atmosphere of the star is suddenly feeling this frictional force that's coming from this hot gas, from the star. The planet is trying to survive, the only thing you can do is to try to rip out the outer layers of the star as the star approaches its surface. But you know, the planet is only one thousandth of the mass of the star, it basically can't put up a fight at all. So it's only a matter of time before as it's trying to rip out the outer surface of the star, the star is going to be like, I don't care about you, you're eventually, it's going to make the plunge into the surface of the star. And what happens in that case is that in the very final moments, it spews out some of the outer layers of the star. That's what we see in the infrared light that's glowing in this cold gas around the system. And in the very final moments as the frictional force gets stronger and stronger because of the surface of the star, the planet basically succumbs and plunges into the surface of the star. And all of that energy from the gravitational energy of the planet just suddenly gets injected into the star itself. And that's what causes it to balloon up into this, you know, thing that is about a factor of 10 bigger than what it was. And in terms of the brightness of the star, it changes by a factor of a hundred over the course of a week.

ERIC: It's crazy to think that a planet one thousandth the size of a star going in could just make it completely expand huge and go way brighter. Like that's, like I said, I always thought of it being a star doesn't care what a planet does to it.

KISHALAY: Yes, so that's actually going to be very close to the situation for our own Solar System planets. Because as far as we can tell planets like Mercury and Venus, which are very close to the sun, they are even tinier compared to Jupiter. So the picture that you have in your head is actually going to be exactly true for small planets. As the sun expands as it reaches Mercury, it's not even going to notice that mercury is just going to keep going, Mercury is going to make a small blip to this energy output of the star because it's so tiny compared to what the Sun is. But in this particular case, the difference was that it was a much more massive planet, something that's about, you know, a percent of the mass of the star itself. It's really in that ballpark that it makes a difference that we can observe at this level.

ERIC: Well, this is an awesome story of discovery that came about accidentally in a detective story. Thanks for telling us all about it, Kishalay.

KISHALAY: Thank you so much.

ERIC: Our Mars spotlight is heating up this summer. Visit mos.org/mars to plan your next visit to the museum around exploring the Red Planet. There's also plenty of content there to enjoy while you're home including other podcast episodes, Kahoot trivia, videos, and even our Roblox game. Until next time, keep asking questions.

Theme song by Destin Heilman