A critical reanalysis of planetary accretion models

Abstract Planet formation models and simulations appear to be converging toward a consensus on how growth may have occured, at least up to the emergence of a few dominant planetary embryos from the vast swarm of planetesimals, which is similar to the scenario described by Greenberg et al. (1978, Icarus 35, 1–26). The various approaches to simulation differ in many details, so comparison is now needed to understand the relation of assumed physical processes and computational approximations to the modeled evolutionary behavior of the system. The algorithms of Greenberg et al. for (a) mutual velocity stirring between planetesimals and (b) mass shifting between adjacent numerical “size-bins” are described in detail. Experiments with the original code, as well as with subsequent revisions, show that the computational methods were accurate to the level of precision appropriate at the time, and the model incorporated the relevant physics including dynamical friction, so its then-novel conclusions remain robust and generally in agreement with various, more recent models. We demonstrate that use of continuum equations is inappropriate for modeling growth with initially all-equal masses; numerical models with discrete building blocks allow incipient runaway growth even with such an initial condition.