I have no doubt that that the majority of stars with photometric variability will be intrinsically variable.
But I would sure like to be wrong!

The dataset all comes down together, so the timing of the non-exoplanet science results depends entirely upon the priority the science team gives it, and how many people they have working on it. It's probably not the highest priority, but if something highly unexpected falls out of the data, I suspect they will need to characterize fairly quickly it so they can rule out the possibility that it's caused by some type of exoplanet.

A non-exoplanet discovery would have to be really something to trump the discovery of habitable-zone Earth-twin planets. Unless it was something to do with solving current cosmological head-scratchers, like dark matter or dark energy, then I don't really know what would qualify as a show-stopper.

Discovering a Dyson's Sphere, perhaps....

Either that or a full sized Culture habitat, or a huge Romulan mining ship firing "Red Matter" bombs..?

How do you discover a Dyson's sphere using photometry method? I thought you need infrared imaging for that

Periodic flashes of starlight through the windows...

...or through holes punched in the outer shell by meteoroid impacts..?

They'd have to be some fairly large holes to be detected by Kepler.
But if you build a Dyson sphere, and you still have asteroids in your solar system, you're just begging for it to get hit

A couple of questions. Given how nicely the first results from Kepler have turned out, should it be possible for Kepler to:

a] detect almost-but-not-quite transiting hot Jupiters from the rising and falling phase-induced light curves (i.e. like HAT-P-7b only without the transits).

b] detect the presence of other, non-transiting planets from variations in the timing of the transits of planets they can see?

Yes. The inclination will still be unknown, but constrained as a function of the light curve variation, and models of the planet's reflectivity. I would expect such things to not really be noticed though. I don't know how the Kepler software works. If it's anything like some others, transit-like dips in brightness get flagged as planet candidates, and then examined more closely. You might imagine that a planet detection like one you describe would easily be assumed to be intrinsic stellar variability. (unless it holds up for a very long time, which would indeed be the case if it's a planet. In that event, it all depends on if it's noticed or not. Then the next obstacle is whether or not to devote resources to follow-up on it).

Yes. What really helps is that the planets that Kepler detects will have many, many transits measured, and in high-detail for the bright stars. A long baseline is crucial in detecting transit timing variations. Kepler is uniquely situated to do this, more so than CoRoT due to a longer observation time in a fixed point in the sky.

Yes, Kepler will detect non-transiting, close-in Giant planets.

"The Kepler Mission readily records the modulation of the light reflected by about 870 close-in giant planets as their phases change between superior and inferior conjunction."

http://kepler.nasa.gov/sci/basis/giants.html

In this case, "close-in" means a period of a week or less.

--Greg

But only if they're close in. For planets in Earth-sized orbits, they'd only have 3 or 4 transits during the course of the primary mission. (Or would 3 - 4 transits be enough to detect variations in timing due to other planets?)

Four transits is probably enough to determine that there ARE other planets, but I'd be very surprised if you could usefully extrapolate much of anything about them. Too many variables and not enough data points.

Thanks for the link back to moons... saved me the trouble of searching. There's something that's I've been thinking about regarding moon detection with Kepler data.

Most investigators seem to be focusing on the transit-timing method for detecting moons: the slight advance or delay of transit ingress and egress due to motion of the planet around the planet-moon barycenter. This timing variance will be slight, but within the edge of detectability. I also saw one reference to the possibility of slight additional dimming before or after the main transit due to the physical body of a large moon. For this to be possible, the moon will have to be very large indeed, approaching Earth diameter: much larger than Ganymede, but within the realm of conceivability.

For a moon this size, a possible third method of detection occurs to me. The Hill spheres of hot Jupiters are very small, so that even a moon in a high-inclination orbit is likely to transit and be occulted by its planet as seen from Earth (or Kepler). Also, it is likely to have a very short period even relative to the abbreviated "year" of the hot Jupiter, probably on the scale of hours. Won't we therefore see a signal as the moon passes in front of and behind the planet during transits? This signal will be much fainter than the main transit signal, of course, but shouldn't it be possible to tease it out? Over the course of time, as the number of observed planet-star transits increases, the observed moon-planet transits will increase at some multiple, enabling characterization of both the moon and the planet. Especially coupled with the barycenter/timing method mentioned above, this could tightly constrain masses and hence densities of exoplanets. Assuming, of course, that such moons exist.

I'm sure someone must be working on this; has anyone more familiar with the literature or the community seen or heard a mention?

Won't the transiting moon always be in eclipse? Do you mean you might see signal from a moon when it is not transiting its planet? (And therefore is neither in front or behind the planet and is transiting the star?)

Yes, Steve, that's what I mean. I suppose it depends on what you mean by "signal," but the indication of the presence of a moon would be the magnitude difference between the star transited by both planet and moon and the the star transited only by the planet, because the moon is transiting the planet or being occulted by it. In my quick-and-dirty illustration (not to scale), on the left the star is transited by both star and moon, but on the right, the moon has moved in front of the planet as it tranits, resulting in slightly less occultation of the star. Is that clearer?