Hot and "warm" Jupiters expelling terrestrial planets?

Question

I've seen in documentaries that if a Jupiter size planet migrates close to its star then it would remove terrestrial planets along the way. This makes sense and I'm sure most models would predict this, but do we have evidence?

Indeed, the Kepler survey (taking Kepler 1-447) found 34 Planets>6 R(Earth) less than 0.1 AU from the star and the only one where another planet was found was Kepler 424 which also has a Jupiter size planet at 0.73 AU. Three "warm" Jupiters (0.1- 1.7AU) out of 47 have earth-size planets discovered in same system and all are "mini-solar systems" with the smaller planets far interior. (K68,90 and 407). If you include "super-earths" (R = 1.25-2.0 R(Earth)), then there are 7 additional stars found with "warm Jupiters" in multiple systems. The super-earths are still all inside the Jupiter's (although K18,89 and 118 have Jupiters less than 0.2 AU).

My question is:

Is this finding proof that migrating Jupiters have thrown out terrestrial planets or do their existence make the signals of "exterior" planets hard to discern? And what about smaller planets migrating in "behind" the Jupiter's?

Answer 1

It's not proof that they've ejected other inner planets, because there are plenty of other explanations for why we haven't observed companions. Steffen et al. (2012) analyzed Kepler data - likely some of the same examples you've looked at - and came up with several explanations besides the no-companions-because-of-planet-planet scattering hypothesis:

• Inner planets never formed. It's possible that mechanisms in systems with hot Jupiters prevent inner planets from forming; the authors are unclear as to what those mechanisms might be.
• The planets are too small to see during transits. This is the first explanation based on detection bias. Obviously, Kepler has its limits, and it's entirely possible that the systems have small bodies it simply can't detect.
• The planets are too low-mass. It is possible that there may be planets with masses so low that transit-timing variations (TTVs) are too small for Kepler to see them.
• The planets have high mutual inclinations. In other words, there might be planets with orbits highly inclined with respect to the hot Jupiters' orbits; we only detect the ones at inclinations favorable to detection from our angle.

The options the authors suggest rest mainly on experimental biases as opposed to theoretical possibilities.

Something I was confused about when I read your question was the assumption that hot Jupiters migrate mainly due to planet-planet scattering. Levison et al. argue that warm Jupiters (i.e. giant planets with semi-major axes of about 1 AU) are more likely to have migrated via planet-planet scattering, because it provides a stopping mechanism via damping, assuming the planetesimal number density scales appropriately. This happens when the energy needed to move planetesimals from their orbits is greater than the energy lost from the change in the planet's orbit. Given that planetesimals cannot survive long in orbits less than about twice the stellar radius, warm Jupiters reach semi-major axes no lower than 0.03 to 0.1 AU (see Murray et al. (1998)).

Hot Jupiters, on the other hand, may be driven largely by Type II gas disk migration, where giant planets create a gap in a protoplanetary disk that subsequently brings in material; this then brings the planet closer to the star, eventually leading to a hot Jupiter. Here's a visualization, from Planet Hunters: