Radiative transfer modeling of near-infrared spectra of lunar mare soils: Theory and measurement

We examine the ability of a radiative transfer model based on the theory of Hapke (1981, 1993, 2001) to reproduce reflectance spectra of known composition, as well as extract compositional information from reflectance spectra. We test this model using spectra of two-component mineral mixtures, spectra of lunar mare soils studied by the Lunar Soil Characterization Consortium (LSCC), and a telescopic spectrum of the Apollo 11 landing site. The model is able to accurately reproduce spectra of two-component mineral mixtures and can be used to accurately predict mineral abundances, mineral chemistry, and particle size. Reflectance spectra of the lunar mare soils are modeled using the mineral abundances and chemistries reported for each soil. We collect our own mineral and chemical information for one of these samples, 12001, in order to examine the effects of several simplifying assumptions employed by the LSCC and conclude that the classification of glass (volcanic/impact versus agglutinitic) can have large consequences on the predicted spectra. Model spectra can generally mimic the lunar mare sample spectra, though with consistent errors in contrast and continuum slope, and absorption bands offset to shorter wavelengths. The Apollo 11 telescopic spectrum is successfully modeled with mineralogy and chemistry from sample 10084, though the fit is improved with slight variations in mineralogy or mineral chemistry. We find that varying pyroxene chemistry can have as large an effect on spectral shape as varying mineral abundance.