Exploring the evolution of lava planets using planet-resolving models

It is thought that most terrestrial planets pass through a magma ocean stage at some point in their lifetimes. This could originate from accretional energy or a giant impact, and potentially be maintained by stellar/tidal heating or a greenhouse effect. Understanding the major processes that cotemporally shape lava planet evolution is key to relating their historical conditions to currently observable states. For example, by explaining how a terrestrial planet entirely loses its atmosphere despite supplicant outgassing of volatiles from the interior.

To this end, we have developed a 1D numerical code which resolves a model lava planet from its core to space. This allows feedbacks between the atmosphere and interior to be captured, as well as the processes that occur within each component of the planet (e.g. convection, radiative transfer). Subject to parameters such as the initial volatile endowment, the model planet can then be evolved over time from an entirely molten state with primordial atmosphere to a fully-solidified state with a secondary atmosphere; in some cases it may not solidify at all. Using a 1D approach allows disequilibrium processes (e.g. photochemistry) to shape planetary evolution and atmospheric composition, which will be reflected in observations made by the James Webb Space Telescope. Another advantage of a generalised planet-resolving model is that it can be applied to a wide range of cases with rocky components: Earth, Venus, the terrestrial TRAPPIST-1 planets (which may or may not currently have atmospheres), as well as sub-Neptunes and super-Earths.

Our preliminary results indicate that steady-state atmospheric composition is heavily dependent on interior conditions (e.g. oxygen fugacity, nitrogen concentration), in contrast to the canonical assumption of steam-dominated atmospheres. Our chemical kinetics module indicates that these atmospheres are typically inhomogeneous, with melt-vapour equilibrium at the surface not being representative of the upper-atmosphere composition. Using the radiative-convective module of our model, we also verify recent literature which has indicated that such thick and hot atmospheres may feature a radiative layer at the surface.