 I know. It's beautiful. It's just spectacular. We thought we'd be alive at this point. Well, I mean, the fact that these go right up like that, and then down, you know, is just astonishing. Look at how steep that is. I'd love to take about a year or two down tonight. I know. It's beautiful. It's just spectacular. Block flew. Good afternoon. Welcome to NASA Headquarters. Today's Space Science Update. We're very pleased today to have a panel of astronomers here to make a significant announcement on research funded by NASA and the National Science Foundation. And here to tell us about that, and to introduce our panel to tell us about that, is Dr. Ann Kenney, who's the Director of NASA's Origin Programs here at NASA Headquarters. Ann? We have some very exciting science to talk about today, which has to do with planet searches. When you look up in the night sky now, you know with a certainty that there are planets around some of the stars in that sky. But this has only been the case since 1994. Before then, it was simply not known. People thought that there were probably planets out there, but it wasn't known. Since 1994, we've discovered about 30 planets. We very loosely hear, since this was not my personal research, but we've discovered about 30 planets of those. 20 of those planets were discovered by the team that is here to talk with us today. I would just like to say something about those planets to put this in perspective. Those 30 planets have all been Jupiter mass or larger. That means 300 Earth mass-sized planets. The planets we're going to talk about today have more to do with Saturn-sized planets, which is 100 Earth mass, so a very different order of magnitude. As I said, of the approximately 30 planets that we know of, 20 of them were discovered by this team. This is a group that has been working on this field for approximately 15 years. This is very hard work. It's very exacting work. I'm really pleased to have this group to talk to us today, the scientists as well as our science experts who will comment on it. Let me introduce today's panel. Jeff Marcy and Paul Butler are people that I've known for 15 years myself, and that makes it a double pleasure. Jeff Marcy is from University of California in Berkeley. Paul Butler is from Carnegie Institution at Washington, D.C., here in town. Allen Boss is also from the Carnegie Institution of D.C. and Heidi Hamill from the Space Science Institute in Boulder, Colorado. So, Jeff, would you like to start out and tell us what you've found? Using the world's largest telescope, the Keck in Hawaii, and using some new improvements to our search technique, we have discovered the first Saturn-sized planets ever found outside our solar system. In fact, we've found two Saturns, and I'll be showing you the data for these in just a moment. One of the Saturns orbits the star called 79 SETI in the Southern Hemisphere, and one orbits the star called HD 46375, both of these stars being old, sun-like stars. An artist's rendering is shown in the animation, if I could have it, please. This shows a star field and then the star itself, 79 SETI, and what the Saturn-sized planet may well look like. The artist's rendering shows the size and orbit of the planet that we actually measure, and it also shows rings and moons drawn speculatively to mimic our own Saturn in our solar system. Our search technique is well known to many people now. It involves monitoring the wobble of these host stars shown in the next animation. What we detect, of course, is the wobble of the star, the motion of the star as it responds to the gravitational pull on the star exerted by the planet. This wobbling motion of the star tells us a number of facts about the planet. It tells us the minimum mass of the planet, and it tells us the orbital size and the orbital shape within which the planet resides. Let me turn now to the data itself, what's really precious to us. The data are shown in the next graphic. The data reveal these Saturns, and these that you see here on this graphic are the data for the star 79 SETI. What you see are the velocity measurements, the dots, versus time on the horizontal axis taken during the last two years at the Keck telescope. You can see that the velocity of the star varies back and forth like a buoy on water waves, indicating that 79 SETI indeed wobbles back and forth. It has a Saturn mass planet. One can use Newton's laws of physics to deduce the mass and orbit of the planet, and that the planet goes around the star 79 SETI in about 75 days. The planet's orbit around 79 SETI is similar in shape and size to Mercury's orbit around the Sun. Same distance and even more eccentric for this planet, a more elongated orbit. And then finally, the last graphic I'd like to show indicates the data we've acquired for the other star, HD 46375. Here you're only seeing three days of data compressed down to three days, and you can see that the planet orbits the star in only three days, one full wobble in three days. And so this planet is very close to its host star. The planet indeed is about 25 times closer to its host star than our Earth is to the Sun. So it's scorchingly hot on this planet. We estimate the temperature on the planet to be something like 2,000 degrees Fahrenheit due to the proximity of the planet to the star. And I think now I'd like to turn to Paul Butler who can tell us a little more about the interpretations of these data. Thanks, Jeff. I'm really excited about these discoveries because they mark two important mile posts in our exploration of planetary systems because of the way that these planets are different from the ones that have been found up until now. But first I'll talk about how these planets are similar to what have been found up until now and how they relate to the planets that have been found. So far all of the planets that have been found fall in one of two classes. They're either these 51 peg-like orbits, four-day orbits where the planet is roasting hot or they're more distant orbits but eccentric, not circular orbits like we find in our own solar system. And these two planets actually fall one in each class. HD 46375 is one of these 51 peg roasting planets and 79 SETI is eccentric. So they fall in the two classes that have been found up until now. These classes are very different from solar system planets. Solar system planets are in beautifully nested concentric circular orbits. And so this has caused speculation that the planets that have been found to date are somehow different than solar system planets. The way we can address this issue is to look at the mass distribution. So if I can have the mass distribution graphic. What this shows is the number of planets that have been detected versus how much mass each planet has. So you see the scale goes from about zero to about 15 Jupiter masses. And the thing that initially was quite startling to us is that we're finding almost nothing that's bigger than about 10 Jupiter masses or so. Such planets or such objects would be very easy to find. By virtue of their huge gravitational field, we get enormous wobbles and yet we're finding none. Instead what we're finding is just a trickle of planets sort of between five and 10 Jupiter masses. And then as we move to smaller mass regimes, one and two Jupiter masses, we're finding just a torrent of discoveries that are coming through. Now the critical thing here is that these planets at the smallest end, the one and two Jupiter mass planets, are the hardest to detect and the ones that we're most likely to miss in our surveys. So this suggests that there's going to be many, many more small planets than there are large planets. And of course the discoveries that we have today push us to even lower masses for the first time going under 100 Earth masses. Now I'd like to talk about why these two discoveries are so special, why they mark two important mile posts. First is simply that up until now the planets, as has been stated, the planets that have been found are more like Jupiter mass, sort of 200 or more Earth masses. And the one enormous milestone now is that these things are about 70 to 80 Earth masses. We've passed 100 Earth mass limit and pushing down even further. And the second thing is this points out sort of the future. So if I can have the 79 SETI graphic again. Up until now we've only been able to find planets that induce really huge wobbles in their star, wobbles that are about four times bigger than Jupiter induces in the Sun. And with 79 SETI for the first time we have a wobble here, which is plus or minus about 11 meters a second, which is actually smaller than the wobble induced on the Sun by Jupiter. So for the first time we are now demonstrating that we can detect solar system-like planets, a real-life Jupiter-like object and a real-life Jupiter-like orbital distance, things that would remind us of ourselves. This is in fact now the biggest remaining long-term goal of this program to maintain this kind of precision for another decade so that we can answer the huge looming question, what fraction of these planetary systems are going to be like our own? Is our own system rare or is it common? It's a completely open question that we hope to be able to provide preliminary answers for within 10 years. That's basically my set of comments. I'll turn it over to Anne now. Thanks Paul and Jeff. I think this represents an incredible achievement and very impressive science. So I'd like to turn now to Alan and Heidi and ask for comments and context on these results, Alan. I'd like to comment on the importance of these discoveries within the context of what's been discovered so far as well as what will be discovered in the next few years. If I could have the discovery space image, please. This plot shows you not only the masses of the objects which have been discovered so far, but also their separations from their stars. You can see that there is at the top of the field there are a few possible brown dwarf companions to some of these nearby solar-type stars, but they're quite rare, surprisingly rare, and there's a rather large gap between the least massive brown dwarf star and what we think is the most massive planet at around 10 Jupiter masses. Then below 10 Jupiter masses, there's this grouping of objects that we believe are gas giant planets very similar to our own Jupiter, and there are roughly 30 or so of those objects now. Now what you see is, first of all, the two new discoveries in pink, HD 46375 and 79 SETI, those are the two newly discovered planets, and as Paul pointed out, they map out a new area of discovery space. They're pushing down the region where we can find extra solar planets to considerably lower masses than what we've seen before. The first point I want to make about this, though, is what you don't see on this plot, which is that you do not see another gap. You see there's a gap up at the top of the plot down to our stars from gas giant planets, but there's no such gap evident in the distribution of what we believe are planets, implying that we really are seeing a continuous distribution in masses of objects that are gas giant planets and going into the range, perhaps, of ice giant planets eventually. And so what we're seeing is really just the tip of an iceberg, that there is an incredible number of other planets that we'll be finding as our search techniques improve. If I could have the graphic back again, I want to make the second point, which is actually mentioned by Paul again, and you'll notice that 79 SETI is sitting out in a region of parameter space that is untouched by previous observations. All the previous observations lie above that top-dashed line called old detection limit, which tells you basically that with a certain amount of precision of your radio-velocity search, you can find everything above that line. 79 SETI, for the first time, defines a new line. The line which says new detection limit pushes it down into a regime where we can expect to find, around other stars, Jupiter analogs. That is truly Jupiter-mass objects with 12-year period orbits, like Jupiter and our solar system. And that sort of object is really, in many ways, the holy grail of what we're after. Now, because such an object could actually also be a signpost for finding an Earth-like planet, which is really what we're trying to find. And holy grail is sort of a term that gets used and overused too much. In terms of our analogy, I thought of perhaps the Golden Fleece, but of course that has some rather unfortunate connotations because of Senator William Proxmire. But you get the idea we're looking for something which is very valuable and very hard to find. And the real goal of much of this long-term program is to try to find a habitable Earth. If we can find a long-period Jupiter around a nearby star, then that will tell us that might be a very good candidate for looking for another Earth. And my colleague Heidi will be telling us a bit more about what we can expect to find in the next few years. That's right. I'd like to actually go back to that graphic again. It's a very complicated graphic, but it has so much information. It really tells us a great deal about what we're talking about today. I also want to highlight something that's not on this graphic, and that is other planets in our solar system. To really put this in context for you, you see the two points there of the new planets that have been found, and you see Jupiter, which lies within the space that we can detect. But Saturn is also not in the realm of detectability by this technique. With the kinds of work that this team is doing, they're pushing and pushing and pushing, and they probably will be able to get to, well, they have gotten to Saturn masses, but they haven't gotten yet to Saturn orbits, and they'll probably be able to push down to Uranus masses, too. And I expected a year, or maybe even less than that, we'll be hearing from them again about Uranus-sized planets. But what about Earth-type planets? Where would they fall on that graph that we're talking about? The answer is Earth doesn't fall on that graph. It's so far below. It's so small that you wouldn't even see it on that figure. And so we're really... When we're looking to the future, we're really looking at even different techniques. We're looking at space-based emissions to find terrestrial planets. And that's, of course, another holy grail, if you will, or golden fleece. We're really looking hard to find terrestrial-type planets. It's a key goal of what NASA is trying to do. And for that, we're going to need to go into outer space, and we're going to be needing slightly newer techniques. We're going to be needing to use techniques called interferometry, for example, the space interferometer mission. We'll also be looking at the terrestrial-planet finder mission. These are going to be very important things that we're going to do in the future. The important news today, the reason we're here talking to you, is that the discovery of Saturn-sized objects is an extremely critical step towards finding terrestrial-type planets. We are now, as we've said, pushing boundaries that we have not pushed before. And so that is why we are all very excited about today's results. Thanks, Heidi. Before we go back to Don, I'd just like to ask Jeff and Paul what these discoveries tell us about our own solar system. I think one of the remarkable aspects of these discoveries and our previous ones is that for the first time in human history we can compare the characteristics of our own solar system with the exquisite nine planets, one of which we cherish the Earth, all of which reside in circular orbits. Some of them are massive like Jupiter and Saturn. Some of them are rocky like the Earth, Mars, Venus. And we have the opportunity now, finally, to compare our solar system to planetary systems as a generic class in our Milky Way galaxy. And the early signs are frightening, actually. And it is in brief that the planets we are finding, as Paul alluded, typically are in elongated eccentric orbits rather than circular orbits that characterize the nine planets in our own solar system. And so the early returns in this poll of planets suggests that our solar system is somewhat unusual. This is an entirely preliminary thought because we have not yet discovered true Jupiter analogs that Alan and Heidi were referring to. And when we can find Jupiter analogs that orbit as far from their star as our Jupiter orbits the Sun, then we will be able to do a direct comparison. But the early suggestion is that the orbits of planets around other stars are more eccentric than those in our own solar system. And this may have something to do with the fact that biology and life has flourished here on the Earth in our solar system. I'd like to go back to the mass distribution diagram for just a moment. One of the extraordinary things about the distribution, of course, is that as you go to lower masses you're seeing more and more objects although they're harder to detect. And if we are to maybe somewhat wildly extrapolate that diagram down to Earth masses it's probably extremely common in the galaxy. The galaxy might just be littered with Earth mass planets. The second point is Jeff was referring to was, of course, the habitability of these potential Earths. And the eccentricity of the orbits of the planets we're finding right now probably rule out nice, stable Earth-like orbital positions where you get liquid water. We can have the graphic of the orbital distribution In this diagram here you see where the sub-saturn that we're announcing today, 79 SETI how it fits in relative to the solar system. A planet like this it's about a border of 100 Earth masses would make stable orbits in the inner solar system impossible. So if such a planet were in our own solar system we wouldn't be here. It turns out that in addition to the fact that our orbit is circular which makes for a nice stable climate and allows liquid water to exist year round and life to flourish. The other planet that's critical to our life is Jupiter and the circularity of Jupiter's orbit because Jupiter is the big boy on the block more massive than all the other planets combined. Whatever it does it enforces on everybody else and by virtue of the fact that Jupiter is in a circular orbit it enforces circularity on the rest of us. Jupiter does a second extremely critical thing which George Wether will point it out that the gravitational vacuum cleaner sweeping up all of the comets and asteroids which might otherwise smash into us and so instead of having one of these huge dinosaur killing impacts every few weeks we only have them every 10 or 30 million years. So Jupiter has these two critical things to make our existence possible and that's why it's so important for us to find solar system analogs Jupiter like planets orbiting other stars. Jeff just mentioned that early returns don't look good and he said that we're finding planets around about 5 to 10% of the stars right now and what's going on with the other 90 to 95% of the stars simply isn't known at this point we need another 10 years of data to make any speculation whatsoever so the field is completely open at this point whether it's every other star that has a solar system like objects or every 10th star or every 100th or every 1000th star completely unknown at this point and we hope to be able to provide some empirical limits within the next 10 years. Ann? Jeff, that's a beautiful description of our solar system as a sort of family that protects where the different members protect each other for developing. Don? Thank you very much. We'll start with questions here at NASA headquarters and then check our centers for go ahead and give you your name and affiliation. Dr. Marcy, can you tell us exactly what you did to refine your technique to find it this way on this graphic it talks about new detection limits. Is there a discernible detection limit that you see given the existing technology? Is this it or how much further can you be back here a year or two years from now? What has made these Saturn discoveries possible is indeed some new breakthroughs technically in our work at the Keck telescope and I should say that our other collaborator Steve Vogt has led the way in this. He built a spectrometer which is at the back end of the Keck telescope $5 million spectrometer without which we wouldn't detect any of these planets and what we've done in the last two years, Paul and I and Steve is to learn some of the nuances the idiosyncrasies if you will of the behavior of this spectrometer it's a complicated optical device nobody could have predicted its actual performance ahead of time and we've done a large number of tests indeed Paul and I have spent the majority of our time carrying out tests of the performance of the spectrometer and developing software indeed Paul has led the way in this, writing software that accommodates the idiosyncrasies of the spectrometer we account for all this odd behavior subtracted away from our measurements leaving us with just the clean data the clean wobble of the star an enormous boost forward making these ever smaller planet detections possible the ultimate limit is indeed ahead of us we are measuring the velocities of stars to plus or minus three meters per second which is sort of bicycle speed it is remarkable if I may reflect on that for a moment these stars are enormous spheres a million times bigger than the earth they are two or three hundred light years away and these globes of gas can be measured to plus or minus bicycling speed but we are at that point and what we'd like to do is to measure the speeds of stars even more accurately to human walking speed about one meter per second and I think Paul and I feel that that's a realistic goal given the known vagaries of the equipment the light gathering power we think with the new improvements that we have coming down the line we will be able to detect not just Saturn's but Neptune's which are about the third the mass of Saturn's we probably will not be able to go further frankly we'll be out of business we think in ten years when we found the Saturn's as well as the Neptune's and then the Jupiter's being cleaned up we will need indeed to turn to the new technology that NASA is developing space-born telescopes without which we will never detect Earth-like planets so you might say this is somewhat unusual and that NASA is supporting you know ground-based astronomy but for us this is a very long-term commitment we we need the planets that these guys are finding and I really would like to emphasize that this is not something they developed the last fifteen minutes of my work on this for fifteen years when I first knew Jeff fifteen years ago I went to visit their setup at Lick Observatory and you know they were hard on work on the zero generation of this technology it's extraordinarily difficult work that is very precise and you have to really be intrepid to succeed at it Leonard? Leonard David with space.com a little bit on the same wavelength what about the complementary nature of the techniques you're using for space-based are you coming up with techniques that could be applied to space-based instrumentation I get some sense here there could be a race between breakthroughs on the ground with technology and what NASA is going to be funding to spend big dollars to put in space Shall I try it? Go ahead Actually I see it a little differently I see the ground-based efforts with Keck and other large telescopes as complementary and early reconnaissance for the space-born work that NASA will be doing and which will blow our minds away I think when it happens and the reason I say this is that our technique is just limited we really will not be able to detect the rocky earth mass planets the space-born techniques both the space interferometry mission called SIM being developed here at NASA and also the terrestrial planet finder mission also developed here at NASA those two are targeted specifically to find earth-type planets we can't touch that so in a way we're operating in parallel the reconnaissance we do now will tell us which stars have the big bullies the giant planets they may be SIM posts for the earth-like planets that NASA will come through to detect in the coming decade or two there's also another element of competition in that NASA has recently approved another space mission called FAME for the full-sky astrometric mapping explorer which finds planets not by looking for the Doppler shift of the star but looking for the positional wobble of the star on the sky and FAME will actually have the ability to find these long-period Jupiter-like objects that Jeff and Paul can find from the ground so unfortunately Jeff and Paul have a several-year head start FAME won't be launched for a few more years but once it gets up in orbit they will also be looking for planets in that same area of parameter space where Jupiter-like objects exist so in this sense it's really very good because competition we all know leads to people to make sure they're doing the best possible job but it's also very important because it will mean that there will be at least a few velocity confirmers can confirm astrometry and vice versa which is of course something we really have to not forget about in any very difficult astronomy, you really want to make sure that any dubious measurement can be confirmed by another measurement and so you really are finding something which is reproducible. But this is a very exciting time in a field, I mean to only know 34 with today's announcement I believe the number is 34 and it's outside of our solar system and this is brand new we're going to learn what animals are in that zoo and we have no idea today, it's going to be very exciting. Bob? I have a question I want to mention that Neptune, when do you mention Uranus which is the outer limit of what you might do from ground-based things? Well actually he said they're going to be looking at Neptune but I think that Uranus is probably going to be easier for them because they're comparable masses but Uranus is closer in so it has a shorter period and so the closer it is to the star the easier it is for them to find. Did your 34 population include those so-called free-floating planets that the Royal Academy Society talked about last week? I don't think so. Those are brown dwarfs I believe I believe they aren't what we would call planets. We call them and this isn't meant to be derogatory they're called failed stars. Randy Schostack reported with E.O. Snooze, at what point in your research of these planets did you say okay we've got some other planets and these are different and what were your initial reactions? These discoveries both sort of came on slowly these were not the sort of eureka moment discoveries we've had those also in the case of 79 SETI we actually had a 75 day orbit for that a year ago and we sat on it for an additional year to make sure that all the data absolutely fitted exquisitely otherwise we would never have announced in the other case 46 375 that's got a slightly larger amplitude the amplitude for that is about 35 meters a second it's about three times larger amplitude which means it's three times easier to detect so again that one didn't really push the system as much it was 79 SETI that we really had to sit on for a long long time to make sure it was absolutely right. Yeah I'll just follow up to confirm what Paul said I was going back through my emails to remind myself of the history of the discovery of the planet around 79 SETI and I have an email from Paul that's dated August of 1999 so I guess that's eight or nine months ago Paul lists four or five hot candidates that he had sifted through and discovered in our existing database and 79 SETI is in there so back in August the email from Paul said it looks like it's a 75 day period and the orbital radius is blah and the mass is blah and what we normally do at this stage is to wait a year typically gather more data and indeed the data for the past year has simply confirmed that we had known back in August of 99. But let me repeat the question he asked, how did you feel when you found these sub-saturns? I would have been more shocked if we'd not found them. We build a technique which can find Saturn mass objects within one Earth orbital distance and if we hadn't found them that would have been infinitely more shocking because that would have been there was a gap in the mass and maybe the objects we had found today weren't related to solar system planets but in fact we're seeing no gap and it's better we're finding smaller and smaller planets and that's exactly what we expect if these things are in fact planets as our own planets are so it would have been shocking not to find them. I'll add another comment to that with regard to our emotional reactions it's very interesting when Paul and I and I can speak I think for both of us here when we see a potential planet coming down our conveyor belt of data we are indeed excited we're happy we're very hard for those moments but frankly the elation is usually tempered by the fright that we've made some mistake somewhere and indeed that's why we spend an extra year following up doing some calculations, some theoretical work we talk with our colleagues and we try to ascertain whether there's some way that we could have gone wrong and Paul and I I think it's fair to say are proudest of the fact that of the 24 planets we've discovered we've never made a false claim in the past four and a half years or so and it's a track record that comes from fear and concern about making mistakes and so our excitement always is tempered by sort of the objectivity of going back to the blackboard and trying to figure out what we can do even better to confirm what we suspect. We'll take two more questions and go to the first and go to Ames and then come back to headquarters. Kurt, go ahead. You explained to us just a little bit more what precisely breakthrough it is that you're talking about in the detection. There's no new detector the resolution is the same the instrument is the same so what is the technological breakthrough specifically that you're describing? We've modeled the spectrometer better now as the bottom line these detections are really incredibly minute literally at the point of Olympic sprinting speed and it turns out very very teeny changes in the spectrometer very very teeny changes in how the light is focused by the spectrometer can dwarf our signal and we need to actually build a physical model of our spectrometer in the computer and we've made significant improvements in our modeling within the last year. We'll take one more question here. Sorry. For us that's the key question actually that you've asked and I'll just say somewhat more quantitatively up till a year ago the precision with which we could measure the speeds of stars was plus or minus eight meters per second and only in the last six to twelve months have these improvements that Paul mentioned building a computer model of our spectrometer allowed us to achieve a precision of plus or minus three meters per second in fact Paul and I think that our precision is even a little better than three meters a second so we have demonstrable improvements both in our results and in the approach that we took to get to these results. Paul. How many other objects do you have on that hot list of candidates that you discussed from August of 1999 and when not we learn of those? It's reminds me a little bit of the when you ask how many potential Jupiters are stored on our hard disk that we haven't yet revealed it reminds me a little bit of the I Love Lucy episode where Lucy says to herself I can have a job and make money and she goes to be a quality checker at a candy factory and the candies come down the conveyor belt and she sees them coming down and there's more and more candies and she can't check their quality to just handle the flow. Paul and I are in a very lucky era which I think we're both just ecstatic about actually and that is we can indeed see down the line there are something like 6 to 10 planets that we know of that we've emailed each other about that we talk about when we're at the telescope that look promising we are waiting for another half a year or year or maybe year and a half of data to confirm these planets some of them are Jupiters sized one I think it's fair to say that it's about a third of a Jupiters sized very close to a Saturn we decided not to announce it today but it's a dead ringer for something about very close to a Saturn sized just a little above so we have a lot of them the Keck telescope is just this marvelous machine for finding planets with the spectrometer and we feel very lucky to have been in the right place with the right telescope to have this conveyor belt of planets occur. I have two questions from the Ames Research Center in California and please give us your name and affiliation. Hi this is Glenda Chiu from the San Jose Mercury. I have two questions first is is there any indication that either of these planets could harbor liquid water? It's very unlikely that these planets would have liquid water. HD 46375 is in one of these three-day orbits and so it's going to be heated to like about 2,000 degrees Fahrenheit. The other planet the 79th SETI is in a 75-day orbital period. The orbital distance is in fact very similar to Mercury in our own system so it's also going to be far far too hot so sadly no liquid water on these systems. Question is can you say something about the status of these space-based missions that would look for Earth-sized planets? How far along are they in the conveyor belt of NASA? Well let me address that. There's really a sort of line of these projects that are very interwoven and we start on the ground actually with the Keck interferometer to try to really get a grasp on optical interferometry from the ground and then we do a technology test in space. It's called Space Technology 3 and that is to get a grasp on formation flying from space because that's what we eventually have to be able to do to find the Earth-sized planets. We have to do formation flying and then we get a grasp on doing interferometry in space so that's the space interferometry mission and then after that and only after that after we have proven each brick of technology in the wall only then do we go on to the terrestrial planet finder where these different technologies are all utilized. This is a challenge nobody says this is easy and I think if it were easy and trivial we probably wouldn't be all that interested in doing it. It's a challenge and there is a very concrete plan in how to proceed. More questions from Aims? Yeah, I'm from a Brazilian daily newspaper we started in Sao Paulo and my question relates to the orbits of these new systems you said they are most of them are eccentric and our own solar system has concentric orbits I wonder if you have a good explanation for that and if it's related to the age of each system? The stars which we have discovered having these planets in eccentric elongated orbits have all types of ages some of the stars are very young in astronomy terms that means the stars are only a billion years old and some of the stars are quite old ten billion years old and stars of either great youth or great age both seem to have these eccentric oval orbits we think but it's in theoretical realm that the oval orbits of the planets stem from gravitational slingshotting of one planet off of another but we're not really sure this is in a realm where we need to gather more data and turn to the theorists who are generating We could try turning to our theorists Alan Boss may have some ideas about where the eccentric orbits come from That's actually an excellent question I think trying to understand the eccentric orbits of these new extra solar planets is one of the major theoretical problems that's outstanding at this point it's really very much a surprise that they have their own solar system and so theorists had spent several decades trying to explain circular orbits now suddenly they realize they've got to explain eccentric orbits as well and one possible explanation is what Jeff just pointed out that perhaps when you make several planets they end up so close together that they gravitationally perturb each other even if they start out on circular orbits they kick each other into eccentric orbits that's one possible way of doing it you have to worry about making them a particular mechanism there is another possibility though which is that if you make gas giant planets through a disk instability which is not the preferred way of making it if you make it through a disk which is cold enough and massive enough to clump together into large balls of gas through its own self-gravity some recent simulations that I've been working on at least imply that when those clumps first form they start out on eccentric orbits so it could be that that particular mechanism actually produces clumps with eccentric orbits which is really pretty much premature to make that claim at this point because these are calculations that need to be reproduced by other workers in terms of the conventional way of making Jupiter where you build up a solid clump and then pull gas onto it those models pretty much imply that you start off on a circular orbit so we've got a real puzzle here well let me address this also one thing we need to stress is that we actually have been finding Jupiter sized planets but we haven't yet seen a true Jupiter analog in other words one that is at Jupiter's distance from our Sun we have not yet found a solar system like our solar system and that's because you guys haven't been observing long enough you should be back at the telescope for another ten years and then after that then we might be able to find solar systems that are like our solar system at this point we just don't have enough data yet okay we'll come back here to NASA headquarters and take more questions go ahead Seth for Dr. Marcy just to continue on that line you mentioned that perhaps the more normalized circular concentric orbits of our solar system are rare is how I guess I know that's more theory how confident are you that that just aren't that a lot of those out there that you're not seeing that they are rare I'm trying to understand why you're leaning toward that way yeah it's a good question here's the situation Paul and I have been using the Keck telescope for the past four years and I should say by the way that we're greatly indebted both to NASA and the National Science Foundation for the funding to let us do this and also to the University of California that also operates the Keck telescope and I bring this up because during the past four years we've been gathering data which allows us to detect planets in orbits that take about four years for the planet to go around if a planet took more than four years to go around its star we haven't been watching long enough and so with NASA and the University of California's help as well as NSF we hope to continue this project for another five or ten years and only then will we be sensitive to planets that orbit farther away namely the orbits that take longer to go around their host star so the bottom line is absolutely right as I think all of us now have touched on we are very excited about the prospect for the future which is to detect planets that take as long to go around their star a decade or two decades as the giant planets in our solar system take to go around our sun but if scientists have 30 objects they're going to start making conclusions about 30 objects who can resist that temptation Dan Bergano with USA Today I had two questions really on the wobble charts you showed the data points I'm curious about how many data points does it take to for you to say looks like one we get early hints literally by the fourth or fifth observation but we can't really get a good orbit until we've got about 15 or 20 observations and the observations have to be over a time baseline that's longer than the orbital period any other questions have you ruled out any stars about how many have you ruled out host for these hot roasters the hot roasters if you pick a mass limit like a Saturn mass limit it's easy to rule out my guess is right now probably only about one star out of every 200 it's got one of these hot roaster planets in sort of the three to five day orbits the maybe the more exciting thing to rule out is the presence of solar system analogs and while I wouldn't go so far to say we can rule any out we do see a lot of stars that are just flat line dead stable that show no velocity variations above our measurement error of three meters a second and we're seeing this now over four years five years or so promising let me just ask how many meters a second does earth tug the sun earth tugs the sun by an imperceptible amount it's about one-tenth of a meter per second no technology in the world can come close to that right now Jupiter tugs the sun by 12.5 meters a second so with 79 SETI for the first time we're measuring things that have a smaller tug than that so jupiter's are not detectable after a jupiter orbital period leveled by order 12 meters a second and we have a number of stars that after sort of four or five years are just dead at the three meter per second level and so I would say of order a quarter of our stars don't look real promising for having anything that we can detect but it will be fun to do this for another ten years and more properly answer your question we have several more questions to go and Stuart Stu Magnuson with Space News I just want to clarify something you kind of answered two questions ago but are the wobbles for eccentric orbits easier to spot than the circular orbits and how many circular orbits have you found so far Fanny? the the best way for me to answer the question is this it's a little bit of a complicated question I would say the following almost all of the planets that have these elongated eccentric orbits we would have detected them just as easily if they had been in circular orbits and the way I'm answering it is I think the important way because your question is a good one one might well worry that we have some bias toward detecting the elongated eccentric orbits over our ability to detect circular ones and that's not the case the orbits that we've detected elongated eccentric these wacky orbits if the planet had been in a circular orbit we would have found it just as well try to so here's the statistic for what it's worth we there are 24 planets known around other stars that orbit farther than a tenth of an earth sun distance one of them 24 out of 24 orbits in an elongated eccentric orbit the planets that we've found and other teams the swiss team has contributed several those planets that orbit closer than one tenth of an earth sun distance all of them are in circular orbits and some people speculate that that may be these circular orbits may have resulted from tides raised on the star and indeed tides raised on the planet due to the star which tend to circularize the orbit even if it were not circular to begin with so the interesting statistic is the 24 out of 24 planets beyond a tenth of an earth sun distance that are in these elongated orbits there are several of those though that I think are the planet orbits far enough away from its star that the star probably hasn't had enough time to circularize its orbit so even though they're on circular orbits those circular orbits are probably primordial but as you go farther away as Jeff pointed out they're essentially all eccentric so there's some real puzzles there theoretically to explain okay we'll try to take them into order Randy and then Paul how do you explain the detection gap between the brown dwarfs and the planets that have been found and why hasn't anything been found in the gap brown dwarf companions to stars are rare at least within the inner three times the earth sun distance which is what we're probing now anything within about three earth sun distances of its star we don't find any brown dwarfs we have 1,000 stars under survey we're doing a full sky survey we're going to extend that to 2,000 stars out of 500 stars that we've carefully examined now over three or four years there aren't any we don't see a single object so they're rare brown dwarfs while brown dwarfs are very common there's probably two brown dwarfs in our galaxy for whatever reason the brown dwarfs do not form around stars at least within the inner part of the solar system and that's up to somebody like Alan to explain that's just empirical I can comment on that also we have to remember that discovery space diagram we showed is actually for companions to solar type stars stars like our sun if you did a similar plot for companions to brown dwarf stars you would find they all have a brown dwarf companion at least half of them do the point is that if you look at binary stars tend to have roughly equal mass companions or at least a companion that's perhaps no more than a third or a fifth of its mass you start getting down to a 101 mass ratio which is where you would be for a brown dwarf companion to a sun like star those things just are very rare we know from looking at other binary stars so when you form two stars they tend to be roughly equal mass so that's why the solar type stars essentially have zero brown dwarf companions but that gives us a great handle for making the argument that we've really found something completely different here that we're finding planets not brown dwarf stars if it had turned out that brown dwarfs could coexist with planets we'd be in a real mess right now we wouldn't be able to tell the two from each other but luckily nature's been kind and it's given us a fairly clear signature as to what we're seeing so we're really at the birth of a field here as we understand the the relation between different planets in a planetary system and this is just the very very beginning of a field that's brand new to us we simply haven't had the ability the power to see these very small and dim objects in the universe okay, front row, we'll try to press on and get all the questions in that we can but we'll have a heart out here in about in eight minutes Marsha Freeman with 21st Century Science and Technology Magazine you had some other announcement last November which was very important relates to something that Dr. Boss said which was a transit of a planet across the face of a star which was a second confirming technique of a planet an extra solar planet and I'm wondering if you could mention what you learn additionally about a planet from a second confirming technique let me just make one comment the first and most important thing that you learned from that was this is what people thought they were it was not it was not a variation in a star it was not some completely other thing that we don't understand yeah, I guess I would start by adding as Alan Boss mentioned it's glorious in science and it's the hallmark of science as a human activity to have an entirely independent technique by competitors perhaps even establish the reality of some physical phenomenon which had already been suspected by one technique so that's the ground base that we like to operate from various groups attacking a problem from different directions getting the same answer the one thing we've learned that I think is the most exciting about the transiting planet is its size its diameter we had known the mass roughly from our wobble measurements and we had known its orbit its physical dimension we now know that these planets are as large as our own Jupiter approximately and that tells us that they are made of gas rather than solid of course an object that's solid would be more compressed and smaller and the transiting planet by virtue of the fact that the star dims as the planet crosses in front allows us to measure the size of the planet and hence determine that it's large and hence gaseous Leonard? Leonard David with Space.com maybe jump into the future a bit several decades listening to Dan Golden he's challenged you kind of folks to image continents or oceans maybe a couple weeks ago he might have said bald spots on aliens I don't know what but it's certainly a challenge what do you think the prospect is in your own mind the march of technology and again space based capabilities of doing that and already gave a really nice summary of what NASA's plans are and how they plan to proceed and I think that most of us recognize that that's a great plan we also recognize that it's a tough thing to do we all understand that the interferometry method is the way we have to go this is where you take multiple telescopes and you can separate them by great distances and yet you can combine the signal to give you a telescope that has the effective area of the difference between the two that's really what we're going to need to really be able to see the continents on another planet I think another thing that most of us believe is feasible in the next why don't we say in our lifetimes because that's probably the right timescale here is to use these very large telescopes or arrays of telescopes to measure the chemical signatures of what is on these planets to use spectroscopy to really detect whether or not there's ozone or methane or some other kind of signature that is there I think most of us think that's probably more likely than taking pictures of these planets and it's going to give us more quantitative information about planets around other stars but after we get a spectrum of a planet and find out that it has ozone it's going to be I think inevitable that we're going to want to eventually take the big leap and build a terrestrial planet imager this may be 30 40 years in the future but it's something that's going to be an imperative for our civilization it'll be a lot easier to do that than it will be to build a spacecraft to go there which we'll talk about even farther in the future it's one of these other giant steps that we'll be taking and if we know that there are earths out there nearby we're going to want to take a picture of them and we'll figure out a way to do it and just to give you a little perspective looking backwards 15 years when these guys were starting on this 15 years ago people laughed at them for doing this Jeff and I have a friend in astronomy who's got a really great test for how to tell whether a project is worth doing and that is you go out to dinner with them and you describe the science and if they all burst out laughing then it's a great thing to do so you know sometimes it takes very new ideas and these guys are working on something that 15 years ago was a joke maybe not quite a joke but but most astronomers were not on the wagon that this is a very cool stuff and you know when we sit now and look into the future 15 or 20 years you've really got to push your ideas we have time for about maybe one or two more I just wanted to clarify I think Jeff said about one fourth of the stars that you look at are not good candidates for planets and out of what kind of a population is that that you've looked at all together let's see I'm not quite sure what you're referring to with regard to one quarter of the stars or planets not being that was Paul's statement you know where the one quarter came from just about one quarter of our stars are flat I see how many stars have you looked at is what he wants to know 200, 500 we have right now about 1,100 stars on our sample about 500 stars we've looked at over a long enough period of time that we can say at least over three or four years we're not seeing variations our goal is to ramp up the survey to about 2,000 stars which will allow us to do every star within about 150 light years and we hope to ramp up within the next two or three years and get every nearby star under survey unfortunately we have run out of time we're losing the satellite and we're going to have to close down you'll be available for questions here in the NASA headquarters thank you very much for joining us today we have a slate coming up here showing you the website where you can get the images and the other information from today also information about the missions that were discussed you can find at the NASA website space science nasa.gov we'll see you again next time thank you very much