Analysis of the potential of kimberlite rock spectra as spectral end member using samples from Narayanpet Kimberlite Field, Andhra Pradesh

Mineral spectra as an end-member have been used for spatial mapping of the mineral deposits and associated potential lithovariants (altered rocks, etc.) while processing the remotely sensed hyperspectral data. But in nature, minerals occur in a mixture in rocks. Therefore, characterization of rock spectra is important. Here we study the role of texture, grain size and relative mineralogical abundances of constituent minerals in shaping the spectral features of the rock spectra in the visible–near infrared and shortwave infrared (VNIR–SWIR) domain. In this regard, analysis of kimberlite rock spectra is carried out to understand how absorption features of its constituent minerals are preserved in the spectral profiles of two different types of kimberlites with distinct mineral assemblages; one serpentine-rich and the other carbonate-rich. It has been observed that the spectral signature of the rock is controlled by the diagnostic absorption features of dominant constituent minerals. However, wavelength and depth of the diagnostic absorption feature of the dominant constituent mineral are modified in the rock spectra due to nonlinear spectral mixing with spectral features of other constituent minerals of the rock. It is observed that the spectral profiles of the rocks are not influenced by the fabric and grain size variation, except for the variation in the albedo or background reflectance of the spectral profiles. Spectral features of the rock also remain stable with the changes in the spectral measurement parameters. Therefore, it is concluded that the rock spectra can be used as an end-member or reference for spatial mapping of the economic rock instead of its constituent minerals.