My guess is that the 10389 day period was made under the assumption that the transiting planet was in a circular orbit, which would result in an orbital motion slow enough to explain the lengthy transit time.

But I think it is considerably more likely that we are instead looking at a highly eccentric body like HD 80606 b (e = 0.921, p = 111.436 days) seen transiting its primary near apoapsis, and hence moving so slowly that its transit signal resembles that of a much more distant low eccentricity planet.

That transit is highly symmetric. Fold it over itself and you will see this.

Neptune's orbit is 60,190 days. Uranus, 30,800. That Kepler would find an object >10,000 days is neither surprising nor unusual.

It isn't surprising that such a planet exists, I'll agree with that.

But it it actually a nice surprise for Kepler to find one as fast as it did. Recall that Kepler detects planets by observing them transit across the stellar disk. For this to happen, their orbits have to be orientated in a certain way for such transits to occur.

For your typical hot Jupiter, there's a ~10% chance of it transiting, assuming a randomly oriented orbit. That number drops steeply with increasing semi-major axis.
If we assume a circular orbit, then the probability of a planet transiting is given by (R_p + R_star) / a, where a is the semi-major axis of the planet.

For a 1 solar-mass star, ~10,000 days corresponds to a transit probability of 0.000541351523
Out of 150,000 stars, if we assume every one has a planet like this, 81 should transit.

But Kepler has been observing for 43 days, that's 0.43% of the orbital period. So out of these 81 planets, Kepler should have observed 0.3483

So the odds of such a planet having been found in the first 43 days of data were around 34% (again, under the assumption that all of them have a planet at 10,000 d).

Sure, the candidate is a G-type giant, but considering the overwhelming prevalance of non-giant stars, I don't belive this significantly affects the probability of detecting such a planet.

Another way to get a 10,000+ day object so quickly in the search is that there are many more such objects than we suppose.

There are, after all, far more 'hot Jupiters' than might have been expected 20 yeas ago, maybe 'cold Neptunes' are a dime a dozen.



{wink}

This paper suggests that cold Neptunes may outweigh cold Jupiters by at least 3 to 1.

http://arxiv.org/abs/0912.1171v1

Hmph.

There goes Mother Nature messing up another perfectly good joke with actual reality.






Crikey, just how many planets are we going to have to keep track of??



Oh, wait a minute, that's a good thing . . . .

Interesting paper accepted for publication in Astrobiology (Segura et al.) with implications for Kepler discoveries... It has been widely assumed that earth-like planets orbiting M dwarfs would have habitable zones lying rather too close for comfort to the star. But this study suggests that life on such a world would probably be able to weather even strong stellar flares, if one makes reasonable assumptions about the atmospheric chemistry.

There's a new manager's update (here). I like the last sentence: "Meanwhile, the Kepler science team is busy preparing scientific papers for publication about Kepler discoveries."

Just want to remind Bay Area people about this lecture. It is scheduled for tonight, about 5 hours from now. I can't make it unfortunately, but hopefully somebody can and will take notes for the rest of us!


RANDALL MUSEUM ANNOUNCES
SAN FRANCISCO AMATEUR ASTRONOMERS’ LECTURE SERIES
Extrasolar Planets & NASA’s Kepler Mission
A Presentation by Dr. Jack J. Lissauer
NASA’s Ames Research Center


What: San Francisco Amateur Astronomers’ Lecture
Extrasolar Planets & NASA’s Kepler Mission
A Presentation by Dr. Jack J. Lissauer, NASA’s Ames Research Center

When: 7:30pm, Wednesday, July 21, 2010

Where: Randall Museum
199 Museum Way, SF, CA

http://www.dailymail.co.uk/sciencetech/art...l#ixzz0ua68fFAU

I was going to post it too. There are a few things that I don't understand. How many Earth sized planets have been discovere? 140 or 60. are they reporting real discoveries or just projections of Kepler's capabilities?

140


60 - is the expected number of analogs of the Earth (the planet the size of Earth in the habitable zone)

60 out of 140 are in the HZ? Time to change Drake's equation again

While Kepler discovery short-periodical planet, far from habited areas.
http://www.youtube.com/watch?v=Ch5MilCuxK8

These are all CANDIDATES. The Kepler team has said they expect that as many as 50% of the CANDIDATES are false alarms (eclipsing binaries, etc.)

"Despite overzealous news headlines this week, NASA's Kepler spacecraft has not indentified more than 100 Earth-like planets in the galaxy...[Dimitar] Sasselov was referencing only possible planets among the Kepler data, scientists said...'What Dimitar presented was "candidates,"' said David Koch, the mission's deputy principal investigator at NASA's Ames Research Center in Moffett Field, Calif. 'These have the apparent signature we are looking for, but then we must perform extensive follow-up observations to eliminate false positives, such as background eclipsing binaries. This requires substantial amounts of ground-based observing which is done primarily in the summer observing season'...However, Sasselov did say that what Kepler has learned so far about extrasolar planets offers tantalizing hints that our planet may not be unusual. Among the hundreds of candidate planets, a large percentage of them appear to be Earth-like – that is, small and rocky, rather than large and gassy, like Jupiter. 'Even before we have confirmed the planets among these hundreds of candidates, we can see statistically that the smaller-sized planets will be more common than the large-sized (Jupiter- and Saturn-like ones) in the sample,' Sasselov explained...

For the complete article on Space.com, go here.

Woof. They need to get rid of the term 'Earth-like' immediately & replace it with 'small & rocky', or at the very least 'Earth-sized'.

This could spin right into an unprecedented PR fiasco all too quickly.

Isn't an earth-sized blink of a transiting planet much smaller than an eclipsing binary? If the dip in light is that small, are the other reasons for it? Oh, "background eclipsing binaries"... how does that create a dip that resembles an earth-sized planet? Another mystery to plague my tiny, earth-sized, amateur brain!

That's what happens when people start quoting headlines from The Daily Mail. Never a good idea on any type of science forum.

I think that the real danger here is the potential negative impact on public perception of Kepler & by extension extrasolar planet research as a whole. Interested amateurs like ourselves know the score, of course, but we've very much in the minority. The general public's been bombarded by SF visions for decades, and dashing incorrectly elevated hopes just doesn't ever end well.

White dwarfs is the size of the Earth.



http://arxiv.org/abs/1006.5030
The difference can reach up to 10 magnitudes. Ie if a candidate in the Earth-like planet brightness of 12 magnitude, then the background eclipsing 22 magnitude. But in the near infrared band difference fortunately falls to 6 magnitudes.



So Spitzer Space Telescope will observe the transit in the near infrared.

http://nexsci.caltech.edu/workshop/2009/Fo...warmspitzer.pdf

I'm confused by that last slide+quote. An atmosphere will give you a transit depth that is dependent on colour. Why should the transit depth be colour-invariant??

edit- I guess the differences aren't as big as in the above example....

Imagine a star with a background eclipsing binary. You can't resolve any components, but you have a light curve that is comprised of the total luminosity of the three stars. The eclipsing binary will make regular transit-like signals in your data. Alone these transits would be to deep to resemble a planet (unless it was a grazing transit), but with the third, unrelated foreground star, the transit signals get washed out. The result is that the transits have a depth that resembles a planet transit.

Agreed. I will note that in the lectures I've attended, the team has been very careful to not use "Earth-like."

With much help from others on this forum, I wrote this post last year:

http://www.unmannedspaceflight.com/index.p...st&p=149307

in hopes we could avoid the ambiguity.

Thanks, Byran and H4i.

Regarding an eclipsing white dwarf, it would have significantly more mass than a small rocky Earth-sized planet, so it would create a substantially greater wobble. The perturbations caused by background binaries seem more likely to be the culprits in a false-positive planet signature. It's gonna be fun finding out!

Indeed. And, for the case of an transiting white dwarf, the deep secondary tarnsit (where the white dwarf passes behind the star) should be a dead giveaway about a non-planetary nature.

Kepler examples include KOI-74 and KOI-81.

Spaceref has a good article on the Kepler data being over reported by news orgs....
http://www.spaceref.com/news/viewnews.html?id=1415

With regard to the frequency of Earth-diameter objects vs. Neptune or Jupiter-diameter ones, I think we need to keep in mind that this mostly applies to inner orbits. I'm happy to learn (if I'm interpreting this correctly) that "hot Jupiters" are relatively rare. I'd like to believe that most solar systems are more or less like ours, with warm terrestial planets near the star and cold jovian planets further out.

Kepler won't see many of the planets further out. Its methodology is unavoidably doubly biased towards inner planets: from geometry, inner planets are more likely to have transiting orbits in the first place, and from time limits, inner planets are more likely to transit during the lifetime of the mission. (Although big planets only have to transit once, while smaller ones need to do it three times, the example of our own solar system suggests that this doesn't help a whole lot.) So the evidence doesn't really show that small rocky planets are more common than gas/ice giants. It only suggests that this is true in the regions close to stars.

Still, this is exciting new info. Not as exciting as what the Mail claims, but still pretty cool. It makes me think that, with luck, by the end of the summer, Kepler will announce the first "Hot Earths."

--Greg

Oh please, PLEASE don't encourage them to use the term "Hot Earth"; such a planet would be nothing remotely like Earth in anything other than size. People reading or hearing the term "Hot Earth" will just think of a cuddly, Earth-sized jungle planet, or a rainforest planet inhabited by Ewoks. Seriously, I shudder at the very thought of that.

Jupiter is the prototypical gas giant. A Jupiter-like planet is one where the majority of the mass is gaseous.
Neptune is the prototypical ice giant. A Neptune-like planet is one where the majority of mass is in ices.
Earth is the prototypical terrestrial planet. An Earth-like planet is one where the majority of mass is in solids.

There's a significant difference between habitable and Earth-like. Venus is an Earth-like planet, but it is not habitable. The atmosphere of a terrestrial planet comprises such a tiny, insignificant fraction of the planet that I don't think it reasonable to distinguish them based on it.


You're correct. And, this isn't a new development. Radial velocity surveys find that 0.8% of sun-like stars host hot Jupiters. Thus, the idea that hot Jupiters were rare pre-dates the Kepler mission.

I too worry that the Kepler team's task of Outreaching and press-communicating the whole "Earth-like" tag is going to be walking across a minefield covered in eggshells. Most members here read "Earth-like" and think in terms of mass, orbit, etc. Your average person "out there" will reada press story announcing the discovery of an "Earth like" world and will think of a planet with crystal blue skies, roaring oceans and fluffy kittens stretching in their sleep in butterfly-filled meadows...

Haha that is true. Maybe some education should go along with the outreach.

"Kepler has found [x] Earth-like planets, that is, planets with radii similar to Earth, ..."

I'm sure when the average person thinks "Jupiter-like," they understand the planet doesn't necessarily have a GRS. Education + outreach could perhaps extend this understanding to Earth-like planets.

I'd drop "Earth-like" and go straight for the does-what-it-says-on-the-tin "Earth-sized". Saying anything is "like" something else instantly suggests to people the two are EXACTLY alike. I hope the Kepler team avoid using "Earth-like" full-stop. That term should be kept until we find a world that truly is "Earth-like", i.e. in terms of size, orbit, and habitability.

I hereby trademark "Lava Earth"! Any and all usage will cost $.01 payable to UMSF as a donation.

I've posted elsewhere - these are 'Earth-like' planets, in the same way a cow is 'Car-like'

Well, this leak story has even reached Science:

http://news.sciencemag.org/scienceinsider/...h-in-earth.html

Although the planets that Sasselov presented ARE candidates and he said that in his presentation, many people are reading into the very confident, affirmative way he said that "there are Earth-like planets" in the data. His chart also says ~1160 planet candidates, which is a big boost from the 706 stars with candidates announced in June. But, the other chart says 700. So, whose to know?

So, although most of these are probably baking-hot corona-grazers, Sasselov seems to be hinting that some are in the HZ.

The ~1160 planets is the 700 Kepler candidates plus the ~460 already known from various non-Kepler surveys.

Ah, got it. That makes much more sense, thanks.

I'm not too worried about public perception. We don't need the science team backtracking on announcements, since that looks bad and calls the data into question. But if the popular press just misinterprets the data, that's primarily an educational opportunity. True, there will be some lingering confusion. But at the end of the Kepler extended mission, we'll have everything wrapped up in a nice digestible package that can get boilerplated into the textbooks and amateur websites.

Well said.

I recall reading somewhere a mention by a member of the Kepler team that they intend to submit a follow on Discovery proposal for a spectroscopic telescope to characterize planets. Has anyone heard about that or know more?

I suppose we might find out once the Discovery proposals are due in a few months.

It's too early for such a mission. Kepler's planets are spectroscopically inaccessible to even the next generation of telescopes due to their extreme distance, star faintness, etc. Kepler's goal is to quantify η_Earth to aid in the design and planning of the next step, a search for the Earth-like planets among the nearest stars with something like SIM-lite. Those planets will be accessible to direct imaging and spectroscopic characterisation of their atmospheres, in what will be the third step, a TPF/Darwin type mission.

Nearby transiting earths in the habitable zone of their stars like those which may be discovered from the MEarth project may be accessible to telescopes like JWST but only with considerable time and effort (watching many transits to build up a sufficient S/N ratio to characterise their atmospheres).

Considering that a metal-rich star is one with lots of carbon and oxygen, and an ice giant is a planet containing lots of steam, I'm going to suggest they call these "mushroom planets."

--Greg :-)

http://www.unmannedspaceflight.com/index.p...st&p=159109

Basically what you've already said. It was Bill Borucki who said they'd propose a spectroscopy mission. I haven't heard/read anything about it since that time.

MSNBC is reporting that NASA will issue some sort of clarification tomorrow.

http://cosmiclog.msnbc.msn.com/_news/2010/...k-causes-a-stir

It doesn't say exactly what form that will take, though.

--Greg

I think that is this one:
http://thesis.iap.fr/
http://ces.jpl.nasa.gov/documents/thesisWh...er_v7-swain.pdf

Thanks, Remcook! That's exactly the type of thing I was looking for. Perhaps the European mission will be competed for the next selection of M or L class missions? It sounds like that type of mission would be tough on a Discovery budget. Maybe NF-class, but astrophysics missions aren't eligible for NF.

Sasselov clarifying things:
http://blogs.nasa.gov/cm/newui/blog/viewpo...blogname=kepler

First Eleanor Cameron reference here that I am aware of!

Thanx for the memories.


Maybe someone on the Kepler team would get some cover art for a little memento . . . .