Kepler Mission Options

[NGC3314]

Why does the mission timeframe matter for ferreting out that sort of thing?

At the moment, nobody has figured out a way to get the photometric precision needed to detect Earth-sized planetary transits against Sun-sized stars from the ground (atmospheric scintillation is a major limiting factor). Kepler or a successor would be the only game in town. On the other hand, if you already had a list of target stars and potential transit times, a less complex and much lower data-rate followup might be able to pick up the search - but compared to that, keeping Kepler going seems like a really powerful thing, especially since it has (AFAIK) no consumables to limit the mission duration.

[stevesliva]

Huh. Wouldn't have guessed the ground was quite that hopeless! I'm of course not augering for an early demise for Kepler, just curious. I guess my misunderstanding all along was not that Kepler was necessarily much more precise than, say, Keck, but that Kepler could just stare away and find these things, rather than looking at rain on Titan.

What sort of spacecraft cameras could do the trick? WISE, Deep Impact, Spitzer...? Thinking along the same lines that once these planets are discovered, their transits could be well observed by something else not dedicated to the task. I'll be somewhat shocked if Kepler and Corot(?) are fairly unique.

[Hungry4info]

WISE is out of comission, Deep Impact lacks the accuracy needed (their constraints to additional planets in known hot Jupiter systems ruled out planets down to ~Neptune radii, and these systems are much brighter than Kepler targets). Spitzer could probably detect it for Earth-size planets orbiting smaller stars, HST maybe.

Other than that, yeah, just Kepler and maybe CoRoT.

[siravan]

...since it has (AFAIK) no consumables to limit the mission duration.

Actually, Kepler - as any 3-axis stabilized spacecraft - has hydrazine thrusters to desaturate the reaction wheels. I think the post launch estimate was that there is enough hydrazine onboard for a 6+ year mission.

[djellison]

And all spacecraft have a consumable that they all need that is very limited. Cash.

[Greg Hullender]

I'm trying to think of a successful mission that wasn't allowed to have an XM, though. For cost-effectiveness, running new experiments on already-orbited hardware is very hard to beat.

In Kepler's case, though, I could see it being a hard sell. More time makes it easier to see more remote planets, but the geometry of the problem makes those much less likely to see in the first place. Of course, as others have mentioned, more time gives you more chances to learn about other planets through their effect on the orbits of the ones already measured. I guess we'll see, but, even so, I'll be surprised if they can't get an XM approved, as long as the vehicle is still working reasonably well.

--Greg

[siravan]

In a sense, the problem is that observing the same field follows the law of diminishing return. One possible option for XM is that Kepler targets a new star field all together. It is a trade-off between possibility of detecting few long-period planets (which, of course, are of great interest) versus detecting lots of short-period planets/hot Jupiters (which improves planet frequency statistics).

[Syrinx]

- Bill Borucki himself has said they will almost certainly win an extended mission. How extended, he did not precisely say. In the context of the discussion it seemed it would be 2-3 years in duration.

- Kepler won't point at a different star field. Kepler takes advantage of star knowledge obtained from ground-based observatories. The current star field was studied for literally years before Kepler was launched.

- Every lecture that I attend it seems like the estimated time to fuel exhaustion (not mission duration) pushes later by a year or two. At first it was three years, then six, then seven, and the last lecture at Stanford Natalie Batalha said up to 10 years.

[Gsnorgathon]

I'm guessing that staring at the same star field longer also increases the s/n, which makes spotting smaller planets (and maybe even moons) more likely. And lets not forget that Kepler can teach us a lot about stars, too!

[NGC3314]

Kepler - as any 3-axis stabilized spacecraft - has hydrazine thrusters to desaturate the reaction wheels.

Ahh, right. Good thing I qualified that. (A further qualification is "any 3-axis stabilized craft outside the geomagnetic field", anyway, since HST does fine managing the reaction wheels with that, and some even manage to do their pointing that way).

A mission extension would make much more sense at the original field, since that gets us new kinds of information (larger orbits) rather than better statistics in a regime already probed.

[Hungry4info]

Not to mention a longer baseline for transit timing variation measurements to discover more non-transiting planets and measure the mass of some who do, but are difficult to detect with radial velocity.

[scalbers]

Catching up on things I noticed an interesting chart of Kepler candidates in the May Sky and Telescope. It shows quite a few planets less than Earth radius (up to about 40 days orbital period). Some are actually down to around 0.1 radius. That's pretty small - perhaps the parent star is also a smaller radius to give a good S/N ratio? Here is a similar chart online (also shown in post #730):

http://arstechnica.com/science/news/2011/0...-candidates.ars

It would follow that given a longer period of monitoring there would very likely be some Earth Sized planets with 1 year orbital period. We'll see if it's around a sunlike star too.

[Rob Pinnegar]

Some are actually down to around 0.1 radius. That's pretty small - perhaps the parent star is also a smaller radius to give a good S/N ratio?

That'd be my guess too. Some of the least massive known red dwarfs aren't that much bigger than Jupiter in terms of radius.

It's not difficult to picture something with 80-90 Jupiter masses having planets at least as large as the Galileans. If a star were to have one-ten-thousandth of the Sun's luminosity, planets circling at distances corresponding to the orbits of the Galileans around Jupiter would have surface temperatures roughly comparable to those of the planets of the inner Solar System. I'm no expert on planetary formation, but one would think that rocky bodies ought to be able to form at those kinds of temperatures. As a bonus, they would be close enough to their parent star to make eclipse-viewing geometry pretty favourable.

What would be *really* neat would be a hot (warm?) Jupiter in a relatively close orbit around this kind of red dwarf. Something like that might actually be larger than its primary in spite of being much less massive. It might be able to completely eclipse the primary. I wonder if a system like this would be stable, in the sense that the planet wouldn't start losing mass to the red dwarf (thereby shrinking)?

EDIT: fixed a mistake (see below... thanks Greg).

[Hungry4info]

One could just look through the paper that listed all 1,235 planet candidates, their radii, and the radii of their host stars...
http://arxiv.org/abs/1102.0541

It's pretty clear the graph is just drawn poorly. Here's a better graph by Mongo with planetary radius being the x-axis and equilibrium temperature on the y-axis.
http://solar-flux.forumandco.com/t282p465-...nd-results#5802

The smallest observed planet candidates have radii at 0.7 R_E.

[Greg Hullender]

It's not difficult to picture something with 80-90 solar masses having planets at least as large as the Galileans. If a star were to have one-ten-thousandth of the Sun's luminosity . . .

I think a star with 80 times the mass of the sun but only one ten-thousandth the luminosity just isn't trying very hard.



--Greg :-)