^1,2,3Tabb C.Prissel,^1,3,4JulianeGross
Earth and Planetary Science Letters 551, 116531 Link to Article [https://doi.org/10.1016/j.epsl.2020.116531]
¹Lunar & Planetary Institute, Universities Space Research Association, 3600 Bay Area Blvd., Houston, TX 77058, United States
²Astromaterials Research & Exploration Science Division, NASA Johnson Space Center, 2101 NASA Parkway, Houston, TX 77058, United States
³Department of Earth & Planetary Sciences Rutgers, the State University of New Jersey, 610 Taylor Road, Piscataway, NJ 08854, United States
⁴Department of Earth & Planetary Sciences, American Museum of Natural History, Central Park West at 79th Street, New York, NY 10024, United States
Copyright Elsevier

We investigate lunar troctolite petrogenesis with a series of forward models. We simulate the cumulate mantle overturn hypothesis by modeling the adiabatic ascent and decompression melting of primary mantle cumulates produced during differentiation of a lunar magma ocean (LMO). Combined equilibrium and fractional crystallization of candidate liquids generated by the melting model can reproduce the predominant constituents of the lunar magnesian-suite (Mg-suite: troctolites and norites), contrary to previous hypotheses. Model results are consistent with previous studies challenging the proposed and long-standing genetic relationship between Mg-suite and gabbronorites.
Our Mg-suite petrogenetic model validates a direct temporal and chemical link between Mg-suite melt production and pressure-release melting of primary LMO cumulates. If so, Mg-suite crystallization ages (4345 ± 15 Ma) can be used to constrain the onset and duration of melting associated with mantle overturn. Based on our model results, we propose an alternative mantle overturn hypothesis whereby upwelling olivine-dominated cumulates experience decompression melting to produce the Mg-suite primary melt (∼1.9% melt at ∼2.1 GPa), but that this melt was extracted from depth akin to lunar picritic glass magmas (low-degree partial melts at depths corresponding to ∼1.3–2.5 GPa). Thus, our revised mantle overturn hypothesis reconciles Mg-suite petrogenesis without the expanse of an olivine-dominated upper mantle (as suggested by the current paradigms, but contradicted by orbital data). This hypothesis supports the presence of a low-Ca pyroxene dominated upper mantle, consistent with mantle stratigraphy constrained by experimental and numerical simulations of LMO differentiation and proposed mantle exposures within impact basins.