# What is the probability that a star of a given spectral type will have planets?

There is a lot of new data from the various extrasolar planet projects including NASA's Kepler mission on extra-solar planets. Based on our current data what is the probability that a star of each of the main spectral types (O, B, A, etc) will have a planetary system?

-
Interesting news today: sciencedaily.com/releases/2012/01/120111133530.htm –  Brightblades Jan 11 '12 at 22:17

Almost all exoplanets observed are near F, G, and K stars. In part, this is because astronomers are looking for earth-like planets, so they look at stars similar to our Sun, but there are also some physical reasons. Sahu et al (2006) have provided some evidence that red dwarfs (class M) are more likely to have planets than other spectral types, though it is hardly conclusive; in any case, planets have been observed around red dwarfs.

No exoplanets have been observed around very massive UV stars (O and B spectral types), and only a few around still-large A-stars. This is probably because very massive stars blast away the protoplanetary disk before accretion allows formation of planets. This was covered in a recent paper by Gorti and Hollenbach (2009).

Incidentally, the most important predictor of whether a given star will have planets is its metallicity. This has been known for quite a while, but Geoff Marcy and company found this most dramatically in a 2005 survey - they estimate that 25% of high-metallicity stars have planets, while only 3% of low-metallicity stars have planets. It's not totally understood why planetary formation depends so strongly on metallicity, but many reasons have been proposed: metallic stars have lower stellar winds, less total UV flux, and their protoplanetary disks are probably more enriched with silicon and iron, which speeds up planet formation.

-
Additionally, all current methods for finding planets are dependent on the strength of the light of the parent star such that brighter stars have greater planetary detectability. This has a substantial impact on radial velocity surveys since it affects which stars are chosen to be studied. Though it also affects even studies like the Kepler mission since it means that for a given volume of space the higher mass stars will be more conclusively examined for planets. –  Wedge Jun 5 '11 at 18:24
@Wedge right - that's a reason that low-mass stars aren't as frequently observed, but you are perhaps going a little too far in saying that all methods depend on strength of light of parent stars. Gravitational microlensing studies depend on the brightness of a background illuminating star, and this method has been used to discover a few exoplanets. –  spencer nelson Jun 5 '11 at 18:37
"...since these are often dead F, G, and K stars." Do you have a reference for that? I have never heard that before. F and G stars end up being white dwarfs, do they not? And late K stars are red dwarfs. –  voithos Jun 6 '11 at 6:16
good point, though microlensing is a very niche technique in comparison to radial velocity, transits, astrometry, and direct imaging. The important point being that all current planet detection systems have an observational bias of some sort and we're still far away from being able to look at a volume of space and get a survey of most of the planets orbiting stars within it. Which means we have to be careful how we interpret statistics from the data we do have. –  Wedge Jun 6 '11 at 22:49
@Wedge Totally agreed. –  spencer nelson Jun 6 '11 at 22:50

Given what we know about planetary formation (Link 1, Link 2, Link 3 and Link 4), and the theories around it, it would probably be a safe bet to say that ALL stars end up having some left over material that might become planets. I think the bigger question is how many of those planetary orbits stay stable enough throughout the life of the star?

All these links aside, I think that it would just be supposition to declare with any certainty that the chance is 100% or 90%, or whatever number you want to choose. We are still trying to gather the data. And our instrumentation is inadequate to the task at hand. We have a bias towards detecting larger planets (up until the Kepler mission). And the Kepler mission is only looking at a small portion of the sky, and will only detect transiting planets, thereby missing all the systems that are tilted relative to us.

That is why there are so many papers on the subject (such as This one, or this one, and even this one), as well as a few competing theories (in the scientific sense of the word).

The intellectually honest answer is, "We don't know." However, it's a great excuse to explore more and find out.

-
Science is never 100% certain, of course, but I don't think anyone really claims that O-stars allow planet formation, so I'm not sure you can say "ALL stars end up having some left over material that forms planets". There are even some claims that the presence of an O-star in a stellar neighborhood can prevent star-formation in that entire region! –  spencer nelson Jun 5 '11 at 17:39
@Spencer Nelson as I said in my answer, I think the bigger question is how many of those planetary orbits stay stable enough throughout the life of the star? The O types may start with the material, but it blows out. Your answer covers the mechanism. :) As well as going into detail about the metalicity and such. Well done. –  Larian LeQuella Jun 5 '11 at 20:13
Perhaps it would be more correct to say that "all stars end up having some left over material that might form planets." –  voithos Jun 6 '11 at 6:17
But we do know that the lower limit to planet incidence is a large fraction. Your answer is not very informative. –  Rob Jeffries Dec 20 '14 at 22:13

This question was asked a couple of years ago and things have changed since then.

We now know that small planets are found around stars across a broad range of metallicities and that it is only the existence of giant planets that are affected by low metallicity. Nature article here.

It was previously thought that small planets were more common around small stars but the latest Kepler results show that small planets are equally common around stars of all spectral types. See this AAS press conference.

"After accounting for false positives and the effective detection efficiency of Kepler as described above, we find no significant dependence of the rates of occurrence as a function of the spectral type (or mass, or temperature) of the host star. This contrasts with the findings by Howard et al. (2012), who found that for the small Neptunes (2–4R) M stars have higher planet frequencies than F stars." (Preprint here)

-
+1 for the up-to-date data. And welcome to physics.SE! –  Chris White Jan 8 '13 at 17:47