1999 年 19 巻 5 号 p. 423-439
Recently, many airborne hyperspectral sensors have been developed for studying and managing the Earth's resources. Furthermore, in the next century several spaceborne sensors are expected to be available. The hyperspectral sensors, that is, imaging spectrometers acquire spectral data in many contiguous narrow spectral bands for each picture element (pixel). Such spectral data can be compared with laboratory and field spectra to identify terrestrial materials. Data processing of hyperspectral imaging data applying to geologic survey needs to be established for such future spaceborne sensors.
The data processing consists of two steps-retrieval of surface reflectance and surface mapping based on spectral features. First, the retrieval process estimates column water vapor in pixel by pixel basis, and then surface reflectance is retrieved based on estimated water vapor condition under the assumption of a horizontally uniform Lambertian surface. The estimation of water vapor was examined for 0.94 and 1.14pm water vapor absorption bands. This retrieval of surface reflectance is based on a radiative transfer code, MODTRAN without any external information on surface materials. The retrieval process was applied to a set of hyper spectral imaging data acquired by Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) over Cuprite mining district, Nevada. The estimated water vapor from 1.14, um seems to be less affected by reflectance of surface materials in water absorptions bands than that from 0.94.um. Furthermore, two alteration minerals, alunite and kaolinite are mapped based on the retrieved reflectance data. The retrieved surface reflectance and mineral map were compared with a detailed alteration map by Ashley and Abrams (1980) and laboratory spectra of rock samples from Cuprite. Generated mineral map clearly shows alunite and kaolinite distribution on the alteration zoning by Ashley and Abrams (1980). The results have demonstrated hyperspectral imaging data has a capability to add information on surface geology.