It was instantly obvious that the Oschin was the telescope to use to find the largest Kuiper belt objects out there. The amount of sky that had been searched for these objects was insignificant. And the way to find these objects was to get a telescope that you could have access to a lot of the time and survey the whole sky.
I spent three years doing a huge survey using the 48-inch telescope and the type of photographic plates used for the Palomar sky survey. Each plate was a 14-inch square of glass with photographic emulsion painted on the back. You'd take it up to the telescope in the dark, load it with the lights out, expose it to the sky for about an hour, take it out, and drop it into a dumbwaiter that went down to the darkroom, where someone would do the developing. After all that, you'd have one picture of the sky on a piece of glass. Then the plates needed to be scanned and digitized so that the computer could look for things that were moving, and we no longer had to look by eye.
This led to three years of very intense effort. We found absolutely nothing, but it didn't matter. I knew that we had the chance to find something really big and significant out there.
The only reason we didn't find anything is that photographic plates can't pick up things as faint as what we can see today with new technology—and we got unlucky. Where we looked there was nothing, but if we had gone just five degrees south, we would've found Xena five years ago on those photographic plates.
In some ways, I'm glad we didn't. It would have been exciting, but it's been kind of fun to do the progression. It was so clear that there had to be large objects out there because people kept finding things a couple hundred kilometers across, and you can extrapolate. You never know if extrapolation is going to work, but we could extrapolate that there'd be a couple of things the size of Pluto or bigger.
We were obviously disappointed that the first three years didn't work out. Apparently, my tenure committee was a little worried about that too. But I wasn't. Not finding something is not a problem—it is still good science. What you need to do is go back, do very careful calibrations, and write a paper about not finding anything so that it's useful.
At that point we had already started working towards putting in the new system that we have now, which is a CCD camera that's very much like the small digital cameras everyone uses these days. The same objects that the photographic plates needed an hour to record could be seen in minutes by the digital camera. That's a huge difference and enabled us to cover a lot of sky. To give you an idea of the difference, three years of work with photographic plates could be done with the CCD in the course of roughly a month. And I could see things that were one-tenth as bright.
I realized that I had to make a big decision. I could either spend my time doing the calibrations of the old survey and write about why I didn't find anything, or I could put the old survey in the trash can and do it again with the new equipment.
"You have to write the old survey up," I was told, and I understood that. That's the obvious advice. I would give that same advice to almost anybody. But I knew it was the wrong advice. So I ignored it. Now you do that at your own peril.
We restarted the survey using CCDs just as the tenure process was moving ahead. By the time the committee had to make its decision, we hadn't found anything at all—well, a few small things, but nothing big.
Luckily, this wasn't the only thing I spent my time on. I'd done a few other things the committee was happy about. And as my division chair explained afterwards: "When it comes to tenure, everyone is looking for home runs. I had to argue that you had hit a lot of singles and doubles, which added up to one or two home runs."
A week after I was given tenure, we found Quaoar.
Quaoar is about half the size of Pluto. Everybody was really excited and wanted to hear about it. This was June of 2002. Now when I look back, it's "Hmmmm, Quaoar was big, but not that big compared to what came afterward."
Sedna was completely unexpected. It's 8 billion miles from the sun—Pluto is 3.6 billion—and in 2004 we had no idea that things in that very outer region of the solar system existed. The fact that they do is going to tell us an incredible amount about the birth of the sun and the earliest history of the solar system.
Sedna shouldn't be there. There's no way to put Sedna where it is. It never comes close enough to be affected by the sun, but it never goes far enough away from the sun to be affected by other stars, which is the case with comets that have been observed in the Kuiper belt. Sedna is stuck, frozen in place; there's no way to move it. And if there's no way to move it, basically there's no way to put it there—unless it formed there. But it's in a very elliptical orbit, and there's no way to form anything in an elliptical orbit like that. It simply can't be there. There's no possible way—except it is. So how, then?
I'm thinking it was placed there in the earliest history of the solar system. I'm thinking it could have gotten there if there used to be stars a lot closer than they are now and those stars affected Sedna on the outer part of its orbit and then later on moved away. So I call Sedna a fossil record of the earliest solar system. Eventually, when other fossil records are found, Sedna will help tell us how the sun formed and the number of stars that were close to the sun when it formed.
Sedna is incredibly far away, and we never would have seen it if it weren't as close as it gets on its orbit. In fact, there's about a 200-year period when we can see it, and it has a 12,000-year orbit. So what does that mean? If we see it for 200 years out of 12,000, that means there's only a 1 in 60 chance that we could've seen it, which means to me that there may be 60 of these things out there. And if there are 60 of these things, then there are probably 20 of these things just a little bigger and maybe a couple the size of Mercury or Mars. We're trying very hard to find the whole population. Once it's done, we'll be able to read the entire fossil record and learn incredible things.
Even though we went on to discover Xena, which is bigger than Pluto and could be called a planet, that is not particularly profound in and of itself. We've known all along that there was likely to be something bigger than Pluto out there, and we finally found it. Scientifically, without question, the most important object we've discovered is Sedna.
Clyde Tombaugh, who found Pluto in 1930, spent a decade or more going out to the telescope at night, taking these photographic plates, developing the plates in daytime, and looking through them. I've never really seen any of the things I find. By "see" I mean looking through a telescope and having photons actually hitting the eye. I don't even have to go to the telescope and do observations. The telescope takes pictures, and I see the pictures on a computer screen in my office. It's abstract and at the same time robotic.
The computer churns through most of the data, and I look through it for about 15 minutes every day. Now it's not like I don't do anything—to automate it like that took years of effort. But that's why it works and why I can actually have a wife and a life.
The very first time I saw Xena on my screen, I thought that there was something wrong. It was too big and too bright. I had to double-check where it was in the sky. Then I did a calculation of how big it was and how far away it was. Xena is the most distant object ever seen in orbit around the sun—10 billion miles away. And it turned out to be 1,800 miles in diameter, about 400 more than Pluto.
I grabbed the phone and called my wife. "I just found a planet," I said. She was pregnant at the time, and she replied: "That's nice, honey. Can you pick up some milk on your way home?"