Lithotype Classification in Geothemal Area by the Use of Hyperspectral Data

This work aims to characterize the surface of an Italian geothermal field, Parco Naturalistico delle Biancane (PNB), by using hyperspectral data and define the main diagnostic spectral features of lithotypes affected by mineral alteration due to geothermal activity. Hyperspectral data acquired by PRISMA (Hyperspectral Precursor of the Application Mission) and AVIRIS-NG (Airborne Visible / Infrared Imaging Spectrometer – Next Generation), coupled with a spectral library of the main lithotype of the area, represent the dataset used for the analysis. All the spectral data cover the VNIR (Visible and Near InfraRed) and SWIR (Short-Wave InfraRed) spectral range. The Material Identification and Characterization Algorithm (MICA) has been used to perform the comparison between the spectral features from the spectral library and the PRISMA and AVIRIS hyperspectral images in order to obtain an automatic lithotype classification map.