Spectral analysis and classification of metamorphic rocks from laboratory reflectance spectra in the 0.4-2.5 μ m interval: A tool for hyperspectral data interpretation

• Published in: International journal of remote sensing Vol. 22; no. 18; pp. 3763 - 3782
• Main Authors: Longhi, I., Sgavetti, M., Chiari, R., Mazzoli, C.
• Format: Journal Article
• Language: English
• Published: Abingdon Taylor & Francis Group 01.01.2001; Taylor and Francis
• Subjects: Applied geophysics; Earth sciences; Earth, ocean, space; Exact sciences and technology; Internal geophysics; Indian Ocean Islands; Madagascar; basement; carbonate composition; Indian Ocean; infrared rays; textures; absorption; reflectance; simulation; spectral analysis; remote sensing; classification; Precambrian; metamorphic rocks; quality; mineral assemblages; airborne methods; instruments; interpretation; imagery; siliceous composition; kaolinization; petrography; albedo
• ISSN: 0143-1161; 1366-5901
• DOI: 10.1080/01431160010006980

Laboratory reflectance (0.4-2.5 w m ) spectra of 41 samples of metamorphic rocks from the Precambrian basement of Madagascar were analysed on the basis of absorption band position and shape, and classified on the basis of recurrent associations of absorption bands. Petrographic analyses allowed us to interpret the absorption features in compositional terms. Spectral and petrographic classes coincided when the principal mineralogy was also spectrally dominant (e.g. in carbonate rocks). When the principal mineralogy did not produce diagnostic spectral features (e.g. in siliceous rocks in the visible-short wave infrared region), the classification was based on spectrally dominant secondary phases. The reflectance spectra were measured on both freshly cut and exposed surfaces of the samples. Apart from a few cases of spectral features obliteration due to kaolinization, or overall albedo change related to texture variation, the two sets of spectra did not significantly differ. The responses of airborne MIVIS and AVIRIS hyperspectral sensors were simulated from spectra representative of the spectral classes, showing that significant identification and classification of well exposed metamorphic rocks are potentially possible using remote instruments providing high quality spectra. Although at present there are no plans for a spaceborne instrument of this quality, TM simulations and band composite images showed that a preliminary gross discrimination of the rocks belonging to the different classes was however possible.