Optimal Inversion Of Optical Constants Of Lunar Minerals And Simulation Of Spectra Of Lunar Soil In High Accuracy

It has profound significance for the study of the mineral composition, geological structure and historical evolution of lunar soil in use of the method of remote sensing to extract information of lunar soil. Due to so few of landing site on the moon, the ground truth data of lunar soil is scarce, which limits the further study on the lunar science. Thus to carry out spectra simulation of lunar soil is a necessary supplement to so limited number of the ground truth spectral data and it can quantitatively analyze the effect of viewing geometries, particle size, space weathering, mineral composition and abundance on spectra of lunar soil.This paper briefly introduces the Hapke radiative transfer model which has been widely used in the spectra simulation. Spectral feature of minerals of lunar soil has also been analyzed. Base on the Hapke model, the photometric parameters can be inversed optimally using multi-angle reflectance spectral data. The inversion results of single scattering albedo which is the most important parameters for the spectra simulation are stable, making it have both mathematical and physical significance. The inversion of the Legendre polynomial coefficients of phase function can be used to simulate the variation of the spectra with the angle accurately. In this paper, comparison of inversion results of the photometric parameters by multi-angle and single-angle spectra data shows that parameters inversion based on multi-angle spectra data can throw light on the set up of parameters of phase function when inversing photometric parameters by single-angle spectra data, which can reflect the changes of the spectra with the angle more accurately compared to the empirical parameters given by Lucey(1998). Based on the Hapke model, the optical constants of lunar minerals can be calculated from reflectance data applying the Monte Carlo method.In use of the inversed optical constants of lunar minerals, reflectance spectra of various particle sizes, maturity, viewing geometries and mineral abundances can be simulated. This paper analyzes the multiple solution and the applicable scope of the Hapke model using Newton interpolation and the least square optimization method. It is found that the method is more suitable for the simulation of spectra, and it is difficult for the inversion of mineral abundance. In the spectra simulation of lunar soil, olivine was subdivided into forsterite and fayalite, and the two end-members of olivine were mixed tightly to approximately estimate the effect of chemical composition of olivine on spectra, which has been proved to be feasible. We simulated the spectra, mineral abundance, particle size and SMFe(lunar submicroscopic iron) content of 57 mare and highland samples of Lunar Soil Characterization Consortium(LSCC) in size group of 10 um, 10-20 um and 20-45 um, to verify the reliability of the simulation method. The simulated spectra and the measured spectra fitted well with the correlation coefficients more than 0.99 and the root mean square error at magnitude of 10-3. The position of absorption center wavelength is difficult to simulate accurately. This may be due to the effect of space weathering, which weakened the spectral features, making it difficult to simulate the absorption position of spectra. The simulated abundance shows a good linear fitting with the measured abundance. Simulation accuracy of volcanic glass, ilmenite, plagioclase, pyroxene and olivine is relatively high with R2 higher than 0.92, and simulation accuracy of agglutinate is relatively poor with R2 close to 0.81. The simulated SMFe content has a distinct linear relationship with the relative maturity index Is/FeO. There are 86% samples whose simulated particle size strictly falling within the measured particle size range, and 100% samples falling within the range of 0-45 um.Based on the simulated spectra of lunar soil, the variation law of lunar spectra can be quantitatively analyzed, which can deepen the study of mechanism of lunar spectra and improve the accuracy of identification of the lunar soil by remote sensing method. This paper analyzes the mechanism, reliability and applicability of the "spectral characteristic angle parameter" proposed by Lucey(1995, 1998, 2000). This method is only suitable for the macro analysis of the entire moon and the error is large when it is used for specific areas especially with high abundances of forsterite or ilmenite.