Cumulate Overturn on Mercury May Have Produced Varied Mantle Source Regions

A view of Mercury’s horizon taken by the Wide-Angle Camera (WAC) of the Mercury Dual Imaging System (MDIS) on board NASA’s MESSENGER spacecraft. Credit: NASA/Johns Hopkins University Applied Physics Laboratory/Carnegie Institution for Science.

It is generally thought that the terrestrial planets were partially or entirely molten early in their history, i.e., they had magma oceans. A magma ocean allows efficient segregation of dense metal to form a core, leaving silicate melt crystallizing into a mantle and primary crust. These primitive mantles probably consisted of a series of distinct layers of cumulates (rock types formed as concentrations of one or two minerals as they crystallize and settle out of the magma), which could have later experienced density-driven overturn. Mercury’s low iron, high carbon, and high sulfur composition suggest that it is an endmember case for magma ocean evolution and volcanic mantle source development.

Mercury’s two largest volcanic terranes are the Intercrater Plains and Highly Cratered Terrane (IcP-HCT) and Borealis Planitia (BP). The mineralogies of these terranes require a deep lherzolitic source (composed of the minerals olivine, orthopyroxene, and clinopyroxene) and a shallower harzburgitic source (composed of olivine and orthopyroxene), respectively. However, during magma ocean crystallization, the first minerals to crystallize are olivine, followed by orthopyroxene, so that harzburgitic layers would form at greater depths than lherzolitic ones.

A new study by Megan Mouser (Carnegie Institution for Science) and Nick Dygert (University of Tennessee, Knoxville) offers a potential solution to this dilemma by invoking cumulate mantle overturn driven by density differences relating to Fe-free silicate minerals and compositionally diverse sulfides during magma ocean crystallization on Mercury. Unlike the Moon, where cumulate overturn would have been driven by the increasing abundance of iron in late magmas, Mercury’s magma ocean would have mainly been Fe-free. Thus, Mercury’s later crystallizing pyroxenes (especially high-Ca clinopyroxenes) could have been denser than the olivine, while sulfides with variable densities could either rise (CaS) or sink (FeS). Mouser and Dygert modeled mantle overturn with these density-driven instabilities and found that it is possible to produce the appropriate deep lherzolitic source for the IcP-HCT and a shallower harzburgitic source for BP within 100 Myr of magma ocean crystallization. READ MORE