TESTS OF IN SITU FORMATION SCENARIOS FOR COMPACT MULTIPLANET SYSTEMS

Abstract

Kepler has identified over 600 multiplanet systems, many of which have several planets with orbital distances smaller than that of Mercury -- quite different from the Solar System. Because these systems may be difficult to explain in the paradigm of core accretion and disk migration, it has been suggested that they formed in situ within protoplanetary disks with high solid surface densities. The strong connection between giant planet occurrence and stellar metallicity is thought to be linked to enhanced solid surface densities in disks around metal-rich stars, so the presence of a giant planet can be a detectable sign of planet formation in a high solid surface density disk. I formulate quantitative predictions for the frequency of long-period giant planets in these in situ models of planet formation by translating the proposed increase in disk mass into an equivalent metallicity enhancement. I rederive the scaling of giant planet occurrence with metallicity as P_gp = 0.05_{-0.02}^{+0.02} x 10^{(2.1 +/- 0.4) [M/H]} = 0.08_{-0.03}^{+0.02} x 10^{(2.3 +/- 0.4) [Fe/H]} and show that there is significant tension between the frequency of giant planets suggested by the minimum mass extrasolar nebula scenario and the observational upper limits. This fact suggests that high-mass disks alone cannot explain the observed properties of the close-in Kepler multiplanet systems and that migration is still a necessary contributor to their formation. More speculatively, I combine the metallicity scaling of giant planet occurrence with recently published small planet occurrence rates to estimate the number of Solar System analogs in the Galaxy. I find that in the Milky Way there are perhaps 4 x 10^6 true Solar System analogs with an FGK star hosting both a terrestrial planet in the habitable zone and a long-period giant planet companion.