Distribution of Radioactive Heat Sources and Thermal History of the Moon

International audience; The initial distribution of heat sources in crustal and mantle reservoirs plays a major role in the thermal evolution of the Moon. We use new constraints on the thickness of the crust, the size of a nearside low in crustal magnetization, surface composition data from orbit, Apollo samples, and mass balance considerations to generate a set of plausible post magma ocean initial conditions. We then test those initial conditions using the 3-D thermochemical mantle convection code Gaia and compare with observables. Models that use Lunar Prospector gamma-ray spectrometer values of thorium throughout the highland crust cannot sustain long lasting volcanic activity, as low abundances of heat-producing elements are left in the mantle to keep an Earth-like bulk silicate composition. The low magnetic field intensities of the innermost Procellarum KREEP Terrane are consistent with a higher heat production than in the outermost portion and delayed cooling below the Curie temperature of iron metal until after 3.56 Ga when the dynamo field strength is known to have decreased by an order of magnitude. The distribution of crustal heat sources also influences the depth evolution of isotopic closure isotherms for a range of isotopic systems relevant to radiometric dating, which may be important for sample age estimation. Core crystallization can sustain a continuous dynamo for about 1 billion years, after which dynamo activity is potentially more episodic. Plain Language Summary The lunar surface possesses different features on the nearside and farside hemisphere. In particular, extrusive volcanism is localized on the nearside and is correlated with a surface enrichment in heat-producing elements (Th, U, and K). How representative this enrichment is of the lunar interior and which processes governed the initial localization are important questions to understand the dynamics of planetary differentiation. In this contribution, we investigate how different initial distributions of those elements within major geochemical reservoirs affects long-term evolution through several observables (such as volcanism, paleomagnetic record, and present-day surface heat flow). Among other things, we found that the deep highland crust must contain less heat-producing elements than the surface values to account for farside volcanism. We also discuss the implication of lateral variation in heat-producing element distribution on radiometric dating of lunar samples.