Runaway growth during planet formation: Explaining the size distribution of large Kuiper Belt objects

Runaway growth is an important stage in planet formation during which large protoplanets form, while most of the initial mass remains in small planetesimals. The amount of mass converted into large protoplanets and their resulting size distribution are not well understood. Here, we use analytic work, that we confirm by coagulation simulations, to describe runaway growth and the corresponding evolution of the velocity dispersion. We find that runaway growth proceeds as follows. Initially, all the mass resides in small planetesimals, with mass surface density σ, and large protoplanets start to form by accreting small planetesimals. This growth continues until growth by merging large protoplanets becomes comparable to growth by planetesimal accretion. This condition sets in when Σ/σ ∼ α^3/4 ∼ 10 ^-3, where Σ is the mass surface density in protoplanets in a given logarithmic mass interval and α is the ratio of the size of a body to its Hill radius. From then on, protoplanetary growth and the evolution of the velocity dispersion become self-similar and Σ remains roughly constant, since an increase in Σ by accretion of small planetesimals is balanced by a decrease due to merging with large protoplanets. We show that this growth leads to a protoplanet size distribution given by N(> R) α R ^-3, where N(>R) is the number of objects with radii greater than R (i.e., a differential power-law index of 4). Since only the largest bodies grow significantly during runaway growth, Σ and thereby the size distribution are preserved. We apply our results to the Kuiper Belt, which is a relic of runaway growth where planet formation never proceeded to completion. Our results successfully match the observed Kuiper Belt size distribution, they illuminate the physical processes that shaped it and explain the total mass that is present in large Kuiper Belt objects (KBOs) today. This work suggests that the current mass in large KBOs is primordial and that it has not been significantly depleted. We also predict a maximum mass ratio for Kuiper Belt binaries that formed by dynamical processes of α^-1/4 ∼ 10, which explains the observed clustering in binary companion sizes that is seen in the cold classical belt. Finally, our results also apply to growth in debris disks, as long as frequent planetesimal-planetesimal collisions are not important during the growth.