Differentiation of lunar basin impact melt sheets and the thickness of mare basalt fills

Basalt thicknesses in mare basins have been determined using assumptions about the premare topography of partly buried craters and by comparison to the Orientale basin. Differences in those assumptions have led to a factor of four difference in mare thickness estimates. Further, knowledge of thickness is restricted to areas in which buried craters are present. Using mixing models applied to multispectral images acquired by the Clementine spacecraft, we have shown that craters in the mare sometimes excavate highland material from below the mare cover. Using such craters and assumptions about their depth of excavation, we obtain independent estimates of basalt thickness. Our results are in agreement with the lower end of previous thickness estimates. These results allow us to determine a volume of 40,000 km{dollar}sp3{dollar} for basalts in Mare Humorum, considerably less than the approximately 110,000 km{dollar}sp3{dollar} from previous estimates. We also confirm a diameter of 425 km for Humorum based on the Clementine gridded global topography, assuming an original morphology like Orientale.; Differentiation of lunar basin impact melt sheets would have important implications for our understanding of the lunar crust. Modeled differentiation of various lunar basin impact melt starting compositions show that it is possible to generate vertical compositional profiles with strong FeO and TiO{dollar}sb2{dollar} variations. Cratering models show excavation of materials from a differentiated melt sheet can produce radial FeO and TiO{dollar}sb2{dollar} profiles that can be discriminated using Clementine UVVIS data. Analysis of craters and their ejecta in the Orientale melt sheet show no detectable FeO and TiO{dollar}sb2{dollar} variations. This supports the model of Orientale melt sheet as an undifferentiated body.