We simulate a Kepler-like observation of a theoretical exoplanet population and we show that the observed orbital period distribution of the Kepler giant planet candidates is best matched by an average stellar specific dissipation function Q_* in the
interval 10^6 ~< Q_* ~< 10^7. In that situation, the few super-Earths that are driven to orbital periods P < 1 day by dynamical interactions in multiple-planet systems will survive tidal disruption for a significant fraction of the main-sequence lifetimes of their stellar hosts. Consequently, though these very-hot super-Earths are not characteristic of the overall super-Earth population, their substantial transit probability implies that they should be significant contributors to the full super-Earth population uncovered by Kepler. As a result, the CoRoT-7 system may be the first representative of a population of very-hot super-Earths that we suggest should be found in multiple-planet systems preferentially orbiting the least-dissipative stellar hosts in the Kepler sample.
Existing exoplanet radial velocity surveys are complete in the planetary mass-semimajor axis (Mp-a) plane over the range 0.1 AU < a < 2.0 AU where Mp >~ 100 M_Earth. We marginalize over mass in this complete domain of parameter space and demonstrate
that the observed semimajor axis distribution is inconsistent with models of planet formation that use the full Type I migration rate derived from a linear theory and that do not include the effect of the ice line on the disk surface density profile. However, the efficiency of Type I migration can be suppressed by both nonlinear feedback and the barriers introduced by local maxima in the disk pressure distribution, and we confirm that the synthesized Mp-a distribution is compatible with the observed data if we account for both retention of protoplanetary embryos near the ice line and an order-of-magnitude reduction in the efficiency of Type I migration. The validity of these assumption can be checked because they also predict a population of short-period rocky planets with a range of masses comparable to that of the Earth as well as a desert in the Mp-a distribution centered around Mp ~ 30-50 M_Earth and a < 1 AU. We show that the expected desert in the Mp-a plane will be discernible by a radial velocity survey with 1 m/s precision and n ~ 700 radial velocity observations of program stars.
