On the relative importance of thermal and chemical buoyancy in impact-induced melting on Mars

Kurzfassung

Convection in planetary mantles is driven by buoyancy that results from density variations, which may have thermal or compositional causes. We study the relative importance of the thermal and compositional buoyancy of melt-induced density heterogeneities in Mars by coupling dynamical 2D convection models with a model of the mineralogy and material properties of martian rock. For the purely thermally driven models, the compositional contribution was suppressed by forcing the density to remain at the value of undepleted mantle. The main focus lies on the anomalies created by giant impacts, which lead to particularly intense melting that may reach deeper than the regular asthenospheric melting zone. The impacts are represented as instantaneous thermal anomalies. As we model them after existing martian craters, we deduce impact parameters such as the impactor size from observed final crater diameters D using empirical scaling laws. Impacts of three different sizes (D: 470-3380 km), all occurring at 4 Ga, are considered. Most models assumed a bulk water content of 36 ppm by mass, but with respect to the ongoing discussion concerning the water content of the martian mantle, we also ran some models with the fourfold initial concentration; the principal effect of this parameter concerns mantle viscosity. In models with both thermal and compositional buoyancy, the strongly depleted compositional anomaly from the impact spreads beneath the lithosphere and remains there as a stable layer, which is progressively incorporated into the growing thermal lithosphere. By contrast, the compositional anomaly in the purely thermal models is mixed back into the mantle and leaves no coherent trace that survives to the present. The thermal anomaly decays by diffusion within a few tens of millions of years in both cases. The crustal thickness at the impact site results from the combination of additional melting, crater excavation, and ejecta deposition. The results suggest that it can be locally overestimated by up to 4-8 km if impact-induced density anomalies in the mantle are neglected. The different behavior displayed by the two model variants is due to the additional density deficit caused by compositional changes of the melting rock. It suggests that the signature of an impact-generated compositional anomaly may be detectable by gravimetry, but a detection by seismics would not be expected with instrumentation whose deployment on Mars can be expected within the next decades.