Application of the mixing length theory to assess the generation of melt in internally heated systems

Abstract: The effect of melting in planetary mantles plays a key role in their thermo-chemical evolution. Because of the laterally heterogeneous nature of melting, 3D numerical simulations are in principle necessary prohibiting us from exploring wide ranges of conditions. To overcome this issue, we propose a new analytical framework allowing to estimate the amount and depths of melting in a 1D analytical model for a simplified convective system. To do so, we develop an approach, partly based on an extended version of the mixing length theory, able to estimate the distribution of the hottest temperatures in natural systems. The approach involves several free parameters that are calibrated by fitting 3D numerical simulations. We demonstrate that our algorithm provides the melting profile at steady-state as well as the long-term evolution in fairly good agreement with the ones obtained in 3D numerical simulations. We then apply our framework for a wide variety of planetary sizes and heating rates. We find that an increase in planetary size increases the depth of melting for small planets but decreases for large planets. This change in the trend is caused by the pressure dependence of the solidus.