Looking for extra-solar planets

[bdewoody]

I was just reading about the Keplar program where astronomers will be looking for extra-solar planets by trying to detect transits across the star being observed. I'm wondering since the universe has no up or down what percentage of stars that have orbiting planets will have their orbital plane edge on from our point of view so that a transit can be observed? It seems like they won't get a very high percentage. <div class="Discussion_UserSignature"> <em><font size="2">Bob DeWoody</font></em> </div>

 

[MeteorWayne]

That is correct, it's not a very high percentage, but returns a lot of unequivocal data.<br /><br />Using the doppler shift method you only get a maximum mass estimate, whereas with a transit you get info on mass, size, etc.<br /><br />I saw a number somewhere (around 3%?), but it would have to be only a rough estimate. The closer the orbit, the more likely it would transit, so such a search would still be biased toward "Hot Jupiters". <div class="Discussion_UserSignature"> <p><font color="#000080"><em><font color="#000000">But the Krell forgot one thing John. Monsters. Monsters from the Id.</font></em> </font></p><p><font color="#000080">I really, really, really, really miss the "first unread post" function</font><font color="#000080"> </font></p> </div>

 

[heyscottie]

Yes, they will be missing the vast majority of systems by using the transit method. But the aim of the mission is not just to find systems, but more to find an approximate count of the systems. Since we can predict what percentage we would be missing, we can simply multiply up the number we find to get an approximate count in the space we search, which should then give us an approximate count within similar regions of the galaxy as a whole.

 

[bdewoody]

Another observation. As I understand it the orbital plane of most of the planets in our solar system is approximately in the same plane as the arms of our galaxy so might not the spin of the galaxy influence the orbital plane of stars with planets that are located this far away from the center of the galaxy or are they still totally random? <div class="Discussion_UserSignature"> <em><font size="2">Bob DeWoody</font></em> </div>

 

[MeteorWayne]

Evidence is pretty this on this point, but as far as I recall, the distribution is rather random.<br /><br />Even for our solar system, the alignment between the plane of the solar system and that of the galaxy is not that close. <div class="Discussion_UserSignature"> <p><font color="#000080"><em><font color="#000000">But the Krell forgot one thing John. Monsters. Monsters from the Id.</font></em> </font></p><p><font color="#000080">I really, really, really, really miss the "first unread post" function</font><font color="#000080"> </font></p> </div>

 

[robnissen]

It does return a lot of data. Also, we should get a lot of new planets out of this. If only 10% of planets have stars, and only 3% of planetary systems are alligned with earth, that would still be 300 new planets discovered (100,000*.1*.03) and that is assuming that we don't discover mutiple planets per star.<br /><br />I do have one question, however, that I have never seen answered. How many light years away could Kepler detect earth orbiting the sun before the sun's light got to dim to detect the further dimming caused by earth (assuming that earth/sun was perfectly aligned with the remote Kepler). In other words, I would like to know the limit of how far Kepler could detect an earth-like world.

 

[MeteorWayne]

An excellent question I don't know the answer to.<br />But would like to <img src="/images/icons/smile.gif" /><br />Perhaps it's buried somewhere on the Keplar mission pages...I'll look around. <div class="Discussion_UserSignature"> <p><font color="#000080"><em><font color="#000000">But the Krell forgot one thing John. Monsters. Monsters from the Id.</font></em> </font></p><p><font color="#000080">I really, really, really, really miss the "first unread post" function</font><font color="#000080"> </font></p> </div>

 

[heyscottie]

Just a note on the "aligned with earth" numbers: If orbital plane distribution truly is random, then detecting an earth sized planet at an earth-like orbit around a sun-like star will succeed only about .48% of the time, not 3%. This means we would only find such systems about 1/210 of the time. Still, looking at 100000 stars, we should be able to find many of them if they are out there...

 

[MeteorWayne]

The 3% may have inlcluded the preponderance ofHot Jupiters discovered so far.<br />I'm not sure where the 3% came from, I read so many things. <div class="Discussion_UserSignature"> <p><font color="#000080"><em><font color="#000000">But the Krell forgot one thing John. Monsters. Monsters from the Id.</font></em> </font></p><p><font color="#000080">I really, really, really, really miss the "first unread post" function</font><font color="#000080"> </font></p> </div>

 

[bushuser]

Many of you read this part of Leonard David's Kepler article:<br />There’s a new advance in the possibility for direct observation of exoplanets, reported Webster Cash, Director of the Center for Astrophysics and Space Astronomy here at the University of Colorado.<br /><br /><br />Cash has been working on an idea tagged as the New Worlds Observer, receiving early financial support and encouragement from the NASA Institute for Advanced Concepts (NIAC). ...<br /><br /><br />The proposal calls for two spacecraft—a large, self-propelled, flower-shaped starshade and a conventional-quality telescope positioned far apart in space. <br /><br /><br />“The key thing is that the starshade stops light from the star from ever getting into the telescope,” Cash said. “This is really what you want. This is the ideal goal for direct study of exoplanets.”<br /><br /><br />The huge starshade—at least 100 feet (30 meters) from tip-to-tip—is a space-based occulter that blocks light from a target star. “If luck is with you, you’ll see a little cluster of tiny faint planets…an actual direct signal from the planets with no interference from the big bright star,” Cash noted.<br /><br /><br />Viewed as a low-cost replacement for the Terrestrial Planet Finder, Cash said the New Worlds Observer starshade would have a price tag of some $500 million, optimized for use with about a 13-foot (4 meter) diameter telescope costing upwards of $1 billion.<br /><br /><br />I wonder if this mission could be successful using the refurbished Hubble telescope [2.4 meter mirror] in concert with the occulting spacecraft which Cash describes?<br />

 

[weeman]

You're right, the chances of finding other planets aren't very high. Being an astronomer who is part of this program, would certainly require very, very good patience <img src="/images/icons/smile.gif" /> <div class="Discussion_UserSignature"> <p> </p><p><strong><font color="#ff0000">Techies: We do it in the dark. </font></strong></p><p><font color="#0000ff"><strong>"Put your hand on a stove for a minute and it seems like an hour. Sit with that special girl for an hour and it seems like a minute. That's relativity.</strong><strong>" -Albert Einstein </strong></font></p> </div>

 

[depthoffocus]

The thing you will be looking for is not a planet or a planet's shadow. You will be looking for a bulge in the electromagnetic plume our electromagnetic equipment detects as any star.<br />Consider, we can see individual items in galaxies so far away that mathematics cannot do justice to the distance, but we can't get a visual image of the planets in the next solar system. There is a certain discrepancy in this.<br />Actually, some wonderful folks recently detected these bulges in the electromagnetic discharge of a nearby star. Them be planets!<br />Please for a moment consider that the images we see as stars are not transmitted, coherent light, but instead are the harmonics of other things. That is a rather bizarre perspective, but I am so fond of them as a group. The best part is that I can demonstrate it in a paragraph.<br /><br />Imagine a star in a distant galaxy. You are asked to pretend that the light emitted by this star is headed toward Earth and your detection system - eye, telescope... BUT, as the light travels, it must pass beside a nearer solar system. The electromagnetic signal gets bent by the electromagnetic/gravitational well of the solar system and we don't get to see it. If you consider that the nearer solar system is in the distant galaxy as well, then by being bent only by an insideously small angle, we don't see it. The problem is that, as telescopes get better, we can look at any portion of the sky and see stars bunched-up, right next to each other. You are asked to believe that either light cannot be bent by prismatic gravity and electromagnetic fields, or that light does not travel between stars. Look up into the night sky...dark with tiny dots. <br /><br />I am standing here, declaring that millennia of stargazing and over a century of quanta have been barking up the wrong tree, and that what we see as starlight, starbright is actually the harmonics of gravity between the two suns. I can appreciate what some of you are winding up to toss... but I

 

[doubletruncation]

<font color="yellow">I do have one question, however, that I have never seen answered. How many light years away could Kepler detect earth orbiting the sun before the sun's light got to dim to detect the further dimming caused by earth (assuming that earth/sun was perfectly aligned with the remote Kepler). In other words, I would like to know the limit of how far Kepler could detect an earth-like world.</font><br /><br />The answer to this depends on the size of the star and the orbital separation between the planet and the star (closer planets have more transits so you can build up a bigger signal). But for an earth-sized planet orbiting 1 AU from a sun-like star, they claim detectability for stars brighter than about V=13 or so. That would give a distance of about 400 pc, or 1400 light years. (you can see plots here: http://kepler.nasa.gov/sci/basis/sizes.html ). <br /><br />Of course this will only be true if they can actually achieve their theoretical photometric precision limit. Any systematic sources of variation will make it very challenging to actually achieve this level of precision, and incredibly impressive if they actually are able to do it (stellar variability, which we really don't know much about at this level, is one such possibility, but more mundane things like scattered light, temperature drifts in their camera/optics, difficult to model charge transfer inefficiencies and flaky pixels are all quite likely to show up, especially as the camera is exposed to harmful radiation over time).<br /><br />Regarding what types of stars they're planning to look at. They had originally designed the mission for F, G and K main sequence stars (the sun is a G star), but I believe they're now also planning to include M dwarfs (they're actually not going to download all the data from their CCD due to bandwidth issues but will only send down about 1% iirc, which means they have to pick the <div class="Discussion_UserSignature"> </div>

 

[doubletruncation]

One thing to note when talking about transit surveys is that you don't actually get the mass of the planet from the transit - you still need to measure radial velocity wobbles to get the mass. You get the radius from the transit. While Kepler will be able to say that there is a good chance that an Earth radius object is orbiting some star, it will be much more convincing if they are able to also detect radial velocity variations and prove that it really is a planet. (Ground based transit surveys have found something like more than 10 false positives for every real planet... of course there are ways you can eliminate these without actually detecting the mass of the planet and ulitimately one might be able to rule out all conceivable false positives... but it still would be a bit of question as to whether or not the thing is actually a planet). The problem though is that it will be exceedingly difficult to measure the RV wobble from an Earth mass planet at 1 AU (something like 10 cm/s) which would take like year on Keck to observe... There are some plans to build a new spectrograph on a big telescope that might be able to detect it in less time... we'll see. <div class="Discussion_UserSignature"> </div>

 

[doubletruncation]

<font color="yellow">As I understand it the orbital plane of most of the planets in our solar system is approximately in the same plane as the arms of our galaxy so might not the spin of the galaxy influence the orbital plane of stars with planets that are located this far away from the center of the galaxy or are they still totally random?</font><br /><br />The ecliptic plane and galactic plane are at more than a 60 degree angle from each other, so we're very much tipped with respect to the galaxy (as you can see from a dark site by noticing that the milky way passes way up north through cassiopeia). The assumption of random inclinations seems to be good - both in terms of percentage of stars detected in eclipsing binaries compared to those detected as just spectroscopic binaries, and for star clusters where you can measure the rotation period and velocities of stars to infer their rotation inclinations. <div class="Discussion_UserSignature"> </div>

 

[robnissen]

Great information. Posts like yours are the reason I visit this board. Thx. <img src="/images/icons/smile.gif" />