QUANTITATIVE MINERALOGY OF THE MOON

Recently we developed a methodology to extract quantitative major mineralogy and chemistry from Clementine reflectance spectra of the Moon. The mapping approach uses a radiative transfer model to precompute a large number of mixtures covering the full range of lunar mineral compositions, and includes the effects of mineral chemistry and space weathering. The model mixtures include 858 different lunar compositions at 10% intervals covering the system plagioclase-olivine-clinopyroxene-orthopyroxene, in which lunar highlands rocks are classified, and ilmenite at 0, 10 and 20% by volume. These 858 compositions were also parameterized in the form of the Mg#, (Mg# = atomic Mg/(Mg+Fe)). Mg# is varied from 50 to 95 at intervals of 5. For each composition and chemistry spectra were computed at ten levels of intensity of space weathering. Space weathering effects were included in the form of submicroscopic iron and iron bearing glass. This process results in just over 85,000 model spectra forming a large lookup table. Calculated spectra from this lookup table are compared against Clementine spectra (750, 900, 950, 1000 nm bandpasses) at 1km resolution. The OMAT parameter is used to restrict and only locations with values above 0.3 are used, which translates to only about 2.3% of the surface of the Moon is analyzed.

On the basis of this model we find that the distribution of the large olivine-bearing mare deposits in Oceanus Procellarum contain ~50% olivine by volume. No other regional olivine-rich mare deposits are found. Orthopyroxene-rich mare deposits are found distant from the Apollo landing sites, and may constitute a new mare basalt type underrepresented in the sample collection. The Feldspathic Highlands Terrane is shown to be compositionally diverse; large portions of the plagioclase-rich highlands have a mafic assemblage completely dominated by olivine and lacking pyroxene. The South Pole-Aitken (SPA) basin is shown to be regionally low in olivine relative the rest of the lunar surface, which shows 5-10% olivine by volume. This supports the conclusion that the SPA impact event did not excavate significant amounts of mantle material. In addition, based on the high abundance of high-Ca pyroxene in SPA, mare basalt is likely far more extensive than previously believed; thus, complicating the task of sampling the SPA melt sheet.