Principal components analysis of Martian NIR image cubes to retrieve surface spectral endmembers

Presented here is a discussion of the complete principal components analysis (PCA) performed on all photometric NASA Infrared Telescope Facility (IRTF) NSFCAM spectral image sets from 1995–2001 and Mars Reconnaissance Orbiter (MRO) Compact Reconnaissance Imaging Spectrometer for Mars (CRISM) spectral image sets from 2006–2008, detailing the similarities and differences and overall interpretation of the PC dimensional spaces. The purpose of the analysis is to use the PCA to recover surface spectral endmembers to be used in a full radiative transfer modeling program to recover ice cloud optical depths (and thus water content) over diurnal, seasonal, and interannual timescales. The PCA results show considerable consistency across all seasons, and can be optimized to increase the consistency through both spectral and geographic restrictions on the data.