Combined use of band shape algorithm, linear spectral un-mixing on Clementine & Moon Mineralogy Mapper data for identifying the imprints of magmatic differentiation – A study around Aristarchus Plateau

Abstract

In this study, we have implemented a band shape algorithm (BSA) to identify the surface distribution of different minerals on parts of the Aristarchus Plateau using selected bands of Clementine and hyperspectral Moon Mineralogy Mapper (M3) data. BSA mineral maps derived using spectral bands of Clementine and M3 sensors could determine broadly similar surface distribution of minerals in the study area. Here we show that M3 based BSA mineral map could detect additional mineralogical details in selected places, especially around Vallis Schroteri, Aristarchus etc. areas. Therefore, we implemented linear spectral un-mixing (LSU) method on hyperspectral bands of M3 data using the reference spectra of Reflectance Experiment Laboratory (RELAB) to estimate the relative abundance of different minerals in above-mentioned areas. Based on the relative abundance of minerals in the LSU map, we could infer the presence of different rocks such as dunite/troctolite, norite, harzburgite, olivine-pyroxenite in those sites of the study area. The relative abundance of each mineral in a particular place was used as the basis for delineating specific rock in that place of the study area. These mineralogically diverse rocks with distinct proportions of mafic mineral and plagioclase content are exposed along a lunar structure that has been delineated using Gravity Recovery and Interior Laboratory (GRAIL) data. We hypothesize that some of these rocks might have formed by a magmatic differentiation process triggered by the structure by separating magma from the crystallized rocks and allowing the magma to compositionally evolve to produce different rocks under the influence of syn-magmatic structural activity.