Results from the Late Lamented Kepler Mission

I was watching a youtube from early in the Kepler mission (Jon Jenkins ; http://www.youtube.com/watch?v=UNGviQ0LPDQ if you wanna look at it), and it set out the "critical questions" the mission (now unfortunately over) was intended to answer. I will essay to say what they found out, based on listening to a number of NASA and other science website lectures on the subject.

• • Are terrestrial planets common or rare?
  
  
  • What are their sizes and orbital distances?
    
  • How often are they in the habitable zone?

First, it's worth pointing out what Kepler was for and what it did in the broadest terms. It's a space telescope, in solar orbit, that looked at only one small patch of sky in the Constellation Cygnus, along the disk of the Milky Way. It critically examined the "sunlike" stars in that field of view out to an effective distance of about 3000 light years. That's 150,000 stars, more or less similar to the Sun. The point was to try to detect planets, and create some meaningful statistics about what it found.

Turns out what I'll call terrestroid (i.e., small, rocky) planets (e.g., Mercury, Venus, Earth, Mars) are very common. Kepler's methodology, which is detection of transiting planets (crossing the disk of their star as seen from Earth) involves a selection effect: only quite flat ecliptic-plane systems are detected, and only something like 2 or 3% at most of stars viewed in the Kepler field of view will be oriented correctly for detection. Single stars are also heavily favored, because binary star planetary disks, even though the stars (it turns out) may often have planets, are unlikely to be flat enough to have transiting planets.

Still, it's thought that the systems Kepler has and will have detected (once all the data is processed), are relatively well representative of sunlike star systems in general. They found that small rocky planets are actually much more common than gas giants (in the Solar System we have four of each; a typical system might have four or five dense rocky worlds and only 1 or 2 gas planets, and there is not a preference for the rocky worlds to be close in, as formerly thought). So, from this point of view, very roughly earthlike worlds should be common. The question is implicit, what percentage of sunlike stars have planets at all? ... and the answer seems to be most of them, although the exact percentage is not really clear from Kepler data. Unfortunately, the one type of planet that would be most interesting, which is small planets, in the habitable zone of sunlike stars, and with an orbital period approximately similar to Earth's, is a type that the mission did not detect well, because it didn't have a long enough time to form a good baseline. Based on what was seen, there's good reason to believe that such planets (between let's say .9 and 1.5 Earth diameters, and in the zone where water can be liquid on the surface) are not particularly rare. A rough (but informed) guess would be maybe 2 to 5% of sunlike stars have such a planet.

Of note, in the Solar System Earth is the largest of the terrestroid planets, but in most of these systems, rocky planets larger than Earth, up to 2½ Earth diameters (still small enough NOT to have retained their hydrogen envelopes, like Neptune or Uranus), are common. Such planets have been widely labeled "super-Earths" in the popular scientific literature. Various models have been proposed for what the implications of the greater mass and size of these planets would be, but there's a good deal of uncertainty still.

• What is their dependence on stellar properties?

This question is less clearly answered. Kepler did not look at stars dimmer than about K5, nor brighter than F0, so it is intrinsically limited to a category of stars that are actually quite a bit brighter than most stars, which are red dwarfs, on average. Turns out (from other detection methods) that red dwarfs almost always have planets, and the sunlike stars mostly do, too. The "habitable zones" of dimmer stars are very narrow, so it remains likely that sunlike stars are a good deal more likely to have liquid water surface rocky planets than are much dimmer stars. But since dimmer stars (not only the M dwarfs but the dimmer-range sunlike dwarfs, say from G5 to K5) are more common, it's a safe bet that the median, or typical, earthlike world with seas, clouds, mountains, plate tectonics, etc., (let's call them "First Order Earthlike;" the presence of life being another matter), will orbit a star dimmer than the Sun. Just exactly how those numbers will work out is still somewhat uncertain.


Binary stars do, it seems, often have planets (most of this information comes from other planet detection regimes), so it would appear that at least some earthlike worlds could be "Tattooines" (planets orbiting two closely separated stars), or could orbit one of two or more very widely separated stars in a single system (e.g., Alpha Centauri, where planets have been detected). But these, it seems safe to say, will be the exceptions: habitable planets are more likely to be found orbiting single stars, probably by a significant margin.  

• Bottom line question: Are we alone?

Unfortunately, unless there's a big surprise in the final reports, it doesn't appear that Kepler is going to answer this question. There are a number of suggested methods of teasing from the data information about whether life could exist on a planet that's been detected. The atmosphere of a planet transiting a star can affect the spectrum of the star, and from that comparison such things as the presence of ozone in the atmosphere could theoretically be inferred. A planet, for example, that had both methane and ozone detectable in its atmosphere would be so seriously in chemical disequilibrium that, albeit controversially, such a detection would be widely accepted as definitive proof of the existence of life there.* But it hasn't been demonstrated, so far. So while Kepler does not provide any reason to infer that life is not likely, or not present on one or more of the detected worlds, it doesn't confirm that, either. As for the detection of intelligent life; well, not. Again, nothing to rule it out, but nothing to confirm it either.
♦♦

*The point is that if chemicals that will react with each other strongly both exist in the atmosphere of a planet, one or both of them must be continuously replenished, and in the case of oxygen (which yields ozone in the stratosphere), that can pretty much only be life, at least over time... if there were only one such detection, it could be explained away by transitory geological phenomena, but if the signature were detected in several different instances, the conclusion would be that it could pretty much only be a biosignature.