HYPERSPECTRAL ANALYSIS OF THE ADVANCED SPACE-BORNE THERMAL EMISSION AND REFLECTION RADIOMETER (ASTER) DATA: A CASE STUDY FROM THE NEOPROTEROZOIC SAWAWAIN BANDED IRON FORMATION, ARABIAN SHIELD

The Advanced Space-borne Thermal Emission and Reflection Radiometer (ASTER) data have been increasingly used in geoscientific studies because of their improved spatial and spectral resolution. ASTER data has three bands in the visible and near infrared (VNIR), six bands in the short wave infrared (SWIR) and 5 bands in the thermal infrared (TIR) portion of the electromagnetic spectrum. These can be effectively used for geological and structural mapping especially in arid regions. We have used ASTER data to map the Neoproterozoic Sawawin banded iron formation in the northern Arabian Shield in Saudi Arabia using hyperspectral digital image processing techniques. These techniques include the identification of pure end-members using minimum noise fraction, pixel purity index images, and N-dimensional visualization followed by spectral angle mapper, spectral feature fitting and spectral unmixing methods to identify the pure pixel from the image. We then compare their reflectance spectra with the end members and thereby map various lithological units in the region including the banded iron formation. Hyperspectral digital image processing techniques proved to be more effective than the conventional digital image processing techniques including supervised and unsupervised classification, density slicing and principal component analysis.