Thermochemical Evolution of Mercury’s Lower Mantle Linked to Early Volcanism

Abstract:

The surface of Mercury is dominated by ancient volcanic plains with a range of compositions similar to Mg-rich basalts and basaltic komatiites. This observed geochemical diversity cannot solely be the result of fractional crystallization of magmas derived from a single source composition, and instead requires the volcanic plains to be products of partial melting of at least two distinct mantle lithologies, consistent with a compositionally heterogeneous interior. Whereas studies of mantle convection on Mercury to date have constrained general parameters for mantle thickness, viscosity, and radiogenic heating, such models have yet to utilize the link between magma generation and past mantle states to impose thermochemical constraints on the interior. To that end, we use a three-dimensional thermochemical mantle evolution model (CitcomS) to investigate the effects of compositional heterogeneity on the form and vigor of mantle convection and the extent and duration of partial melting. We investigate a range of compositional scenarios for Mercury’s mantle that are broadly consistent with planetary formation and differentiation from enstatite chondrite and carbonaceous (CB) chondrite bulk compositions. Our results indicate that the magma generation implied by widespread volcanic plains emplacement until ~3.8–3.6 Ga must have occurred predominantly in the lower half of the mantle. Although low-degree partial melting is possible in the shallow mantle for the first ~500 My after planetary differentiation, we find that temperatures in the shallow mantle are not sufficiently high to account for the prolonged and extensive magmatism required to produce both intercrater and smooth plains. Additionally, our results suggest that during the period of widespread plains emplacement, layered mantle convection would permit the lower mantle to convect vigorously beneath a sluggishly convective or conductive upper mantle. A compositionally stratified mantle on Mercury not only could yield the substantive magmatism necessary for volcanic plains emplacement, it would also allow for multiple mantle source regions as indicated by the observed geochemical diversity of surface volcanic materials.