Targeting Limestone and Bauxite Deposits in Southern India by Spectral Unmixing of Hyperspectral Image Data

This paper presents a study about the potential of spectral unmixing of hyperspectral satellite image data for targeting and quantification of mineral content in limestone and bauxite rich areas in southern India. ASTER image (acquired in the VNIR and SWIR regions) has been used in conjunction with SRTM – DEM and field studies. A new approach of spectral unmixing of ASTER image data delineated areas rich in carbonates and alumina. Various geological and geomorphological parameters that control limestone and bauxite formation were also derived from the ASTER image. All these information, when integrated, showed that there are 16 cappings (including the existing mines) that satisfy most of the conditions favouring bauxitization in the Kolli Hills. The sub-pixel estimates of carbonate content in the limestone area of Ariyalur, south India, match well with the geochemistry of the samples collected from the study area. The study concludes that spectral unmixing of hyperspectral satellite data in the VNIR and SWIR regions may be combined with the terrain parameters to, not only target limestone and bauxite deposits accurately, but also estimate the quality of these deposits.