NASA NTRS2004
The successful models for the internal evolution of the Moon must consider the volume, distribution, timing, composition and, ultimately, the petrogenesis of mare basaltic volcanism. Indeed, given the paucity of geophysical data, the internal state of the Moon in the past can be gleaned only be unraveling the petrogenesis of the various igneous products on the Moon and, particularly, the mare basalts. most useful in constraining the depth and composition of their source region [Delano, 1980] despite having undergone a certain degree of shallow level olivine crystallization.The bulk of the lunar volcanic glass suite can be modeled as the partial melting products of an olivine + orthopyroxene source region deep within the lunar mantle. Ti02 contents vary from 0.2 wt % -1 7.0wt [Shearer and Papike, 1993]. Values that extreme would seem to require a Ti- bearing phase such as ilmenite in the source of the high-Ti (but not in the VLT source) because a source region of primitive LMO olivine and orthopyroxene, even when melted in small degrees cannot account for the observed range of Ti02 compositions. The picritic glasses are undersaturated with respect to ilmenite at all pressures investigated therefore ilmenite must have been consumed during melting, leaving an ilmenite free residue and an undersaturated melt [Delano, 1980, Longhi, 1992, Elkins et al, 2000 among others]. Multi- saturation pressures for the glasses potentially represent the last depths at which the liquids equilibrated with a harzburgite residue before ascending to the surface. These occur at great depths within the lunar mantle. Because the liquids have suffered some amount of crystal fractionation, this is at best a minimum depth. If the melts are mixtures, then it is only an average depth of melting. Multisaturation, nevertheless, is still a strong constraint on source mineralogy, revealing that the generation of the lunar basalts was dominated by melting of olivine and orthopyroxene.