Presentation #102.384 in the session Poster Session.
Planet population synthesis can be used to constrain global models of exoplanet formation by comparison with observations at the statistical level. Here, we perform such a comparison between the Bern model of planetary formation and the planets of the combined HARPS/Coralie sample. For this purpose, we develop a framework to apply observational biases and perform statistical tests.
We find that the nominal synthetic population of our previous work reproduces the close-in super Earth and distant giant planet trend of the HARPS/Coralie sample, but there are relatively too many giant planets and the “planetary desert” for masses between that of Neptune and Saturn is too deep. We compute additional populations to find the model parameters that best reproduce planet masses. Decreasing the model efficiency by increasing the planetesimals size also rebalances the number of super Earths and giants. Then, changing the efficiency of gas-driven migration affects the distribution between super Earths and giant planets; however it cannot be used to control the location of the final planets. Even with improved parameters, our model is unable to reproduce the mass and distance distribution together. This indicates that planets likely originate from further away than our model predicts. Also, this suggests that the planetary desert is due to a combination of runaway gas accretion and planetary migration.