Classification of geologic features in the open pit mines using high resolution hyperspectir imagery

Abstract

The areas of open pitting and strip mining are common methods of ore extraction in the wide-area disseminated gold and porphyry copper deposits in the Western United States. The processes of open-pitting remove much of the surface cover in the affected areas and expose geologic features of interest at various depths, often at great detail and can yield clues to the formation of the ore deposit as a whole. Using high resolution hyperspectral imagery acquired over several open pit mines (Sacaton Pit, AZ; Borax Pit, CA) it is possible to delineate the principal elements of geologic diversity: veins, contacts, mineralization host rocks and various zones of hydrothermal alteration on the basis of their characteristic spectral absorption features. Another useful element of hyperspectral remote sensing over mining operations is general characterization of the landscape cover types and mineralization as a whole, prior to mine development for design of life cycle of mine reclamation, monitoring the preservation of key environmental attributes during mine development and finally monitoring the effectiveness of environmental remediation during mine closure. Automated classification of geologic / landscape features using radiance data aids in generation of distribution maps for particular features (i.e. alteration type) whereas analysis of reflectance data allows for abundance mapping. The combined process greatly improves the overall quality of the acquired data, reduces the overall processing time for the analysis of hyperspectral imagery and may be a useful tool application of remote sensing in geological/geophysical surveying and exploration worldwide.