Sitting at Home, Searching for Stardust

MADELEINE BRAND, host:

This is DAY TO DAY. I'm Madeleine Brand, and on the line with me is my colleague Alex Chadwick. He's usually sitting right next to me here in the studio, but he's up in San Francisco now. He's going to co-host the program tomorrow from member station KQED, but now, Alex you're here with one of your stories. What do you have?

ALEX CHADWICK, host:

Madeleine, I've been over at Berkeley yesterday, talking to some scientists there about a outer space search story for people here on earth. This is a big volunteer project with anyone, really, able to participate from home over the internet. It's fun science that's also very useful to these researchers and astronomers for studying questions about the origins of the universe and how stars and planets are born, things like that.

Okay, so I spoke with Dr. Andrew Westphal. He's at Berkeley, he's a Senior Fellow, and he's Associate Director of the Space Sciences Lab there. And he's working with NASA on a way to find and analyze tiny particles of space dust that were collected by a space craft called Stardust. It went out between Mars and Jupiter, ran through the tail of a comet, and then came back to earth. It landed in Utah last month, there were stories about that, because it had this cargo of space dust collected by a sort of space sponge it was carrying. And it actually collected two kinds of dust, the comet dust it ran through; it's very very old, maybe four and a half billion years old.

And this other dust, called interstellar dust, that they're very interested in. Now, these are little particles that are floating around the vast region of space that the Stardust flew through. The scientists knew that they would get the comet dust. The interstellar dust is trickier. They guess that they may have picked up dozens of these little things somewhere on their collector, but they don't know where it is, and this is the search that they're organizing with internet volunteers. They've already got more than a hundred thousand people signed up. We're going to explain to you all in a little while how to sign up, but first here is Andrew Westphal from Berkeley.

Doctor ANDREW WESTPHAL (Senior Fellow, Associate Director of Space Sciences Lab, Berkeley): The project really was born out of desperation. We had been thinking about how to find these tiny interstellar dust particles in the aerogel collectors from the Stardust space craft. And we had originally assumed that we would use some sort of automated image recognition technique to find these particles, but we couldn't really convince ourselves that this was a reliable way to go. So, we were thinking about this problem, and talking with our colleagues down the hall here at the Space Sciences Lab who run the Seti at home project.

CHADWICK: So Seti at home, this is the search for extra terrestrial intelligence which is going on. It basically looks through a huge amount of data collected by radio telescopes so huge that it takes a vast amount of computer power to search through all this, and the scientists at Berkeley came up with this idea of asking people to link their home computers to the search, and allow researchers to use these home computers in their down time, basically, and they were amazed at the response.

Five million people have signed up for this Seti search already. In some ways, this new Stardust search that they're starting is going to be even more engaging, because people at home on this project will be looking at pictures on their home screens. Tiny sections of this space sponge I keep referring to, it's actually something called aerogel, and trying to find, in these pictures, signs of these particles. They're really really small, but when they hit the aerogel, they leave a little track, and Andrew Westphal told me that's how people, like anyone, like you and me, are going to be able to recognize them.

Dr. WESTPHAL: You know, if we had started this project twenty years ago, what we would do is hire a small army of professional microscopists, people who spend their careers looking through microscopes. And they would be working eight hours a day, every day, going from one field of view to the next, peering through microscopes, and focusing up and down, looking for these tiny tracks. And they would have to look through one and a half million fields of view to find the interstellar dust particle tracks. If we had started this twenty years ago, we probably would still be doing it today using that technique. But we think we can do all of this in just a few months with the assistance of volunteers from the public.

CHADWICK: Why do you think people are better than computers at this?

Dr. WESTPHAL: Part of the reason we think that's true is that people are, if anything, more naïve. They don't really know what to look for. And we're actually relying on that, because we have our own ideas, our own preconceptions of what these particles should look like in the aerogel. But the fact is that we can't actually simulate these impacts in the laboratory, so we don't know what they're going to look like. It may be that people who have never looked at aerogel before, and have never looked at tracks before, will actually be better at it than we are.

CHADWICK: If I volunteer for this program, or someone listening to our program volunteers for it, when these images are up, what, next month?

Dr. WESTPHAL: We think it's going to be next month, yeah. Mm-hmm.

CHADWICK: When they're up, you'll go click on an image, and what do you see? A blue screen?

Dr. WESTPHAL: What you will see is a field of view from the aerogel, from the surface of the aerogel. And it looks a little bit like looking at a piece of glass, except that it has some imperfections; it has scratches, it has pieces of dust that came off of the spacecraft when it was in space for seven years. It has some little pieces of dust and other things that were incorporated into the aerogel, inevitably, when the aerogel was made. And that's part of the problem, is that it's hard to train a computer to distinguish between those things.

What we noticed when we were doing this was that our eyes are exceptionally good at recognizing these tracks. When a computer is rather hard to train to recognize these, and we found that even people who were quite naïve and could walk into our lab and look, just for a couple of minutes, we give them a few examples of tracks, and they recognize them really easily. But we just didn't know how to write a computer algorithm which would do the same thing. And we found that people are really good at it. So, it's was a...

CHADWICK: What are you going to get from this? What do you get from a little tiny particle of interstellar dust?

Dr. WESTPHAL: Well, these interstellar dust particles are extremely small. They're only about a micron in size, or smaller.

CHADWICK: And how big is a micron?

Dr. WESTPHAL: A micron is about a hundredth the diameter of a human hair. So, these are very tiny particles, and they're very precious, because we'll only have a few dozen of them in the entire collector. So, we'll look at a few of these particles, but we really want to let the analytical community be driven by the inspiration of analyzing these tiny, tiny things.

CHADWICK: But, are you saying that you're not quite sure what, what data you will get from these, these particles?

Dr. WESTPHAL: Absolutely right. We don't. We don't know what we're going to find. That's why we're doing it. If we knew, we wouldn't bother.

CHADWICK: Madeleine, what I liked about this story is that they actually don't know what they're going to find. You can infer there'll be certain information about astronomical phenomena, but they don't know exactly what they're going to find, but people are already signing up for the search, and it's going to get under way in about a month.

BRAND: So does that mean you, or me, or anyone could just sign up?

CHADWICK: Anyone can actually do this. You go to a website. We have a link at our website, NPR.org. You go to a website at Berkeley, you take a little, kind of five minute training thing, and then you're on. You're a part of the team.

BRAND: All right, DAY TO DAY'S Alex Chadwick, and I am looking forward to co-hosting with you tomorrow from San Francisco.

CHADWICK: Talk to you then, Madeleine.

BRAND: And NPR's DAY TO DAY continues. I'm Madeleine Brand. Transcript provided by NPR, Copyright NPR.