GSA Annual Meeting, November 5-8, 2001

Paper No. 0
Presentation Time: 2:30 PM

THERMAL INFRARED MULTISPECTRAL ANALYSIS OF MAFIC VOLCANIC ROCKS NEAR GILA BEND, ARIZONA


KNUDSON, Amy Trueba, Geological Sciences, Arizona State Univ, Box 871404, Tempe, AZ 85287-1404 and CHRISTENSEN, Philip R., Geological Sciences, Arizona State Univ, PO Box 871404, Tempe, AZ 85287-1404, amy.knudson@asu.edu

Mapping the variability of rocks on the surface of Mars provides one means of studying the geologic history of the planet. The primary igneous rocks identified on Mars using the Thermal Emission Spectrometer (TES) include basaltic to andesitic compositions. While these rock types have been distinguished in the thermal infrared using the spectral resolutions afforded by spectrometers, they have not been well characterized using multispectral instruments such as the Thermal Emission Imaging System (THEMIS) currently en route to Mars. The Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) and the Thermal Infrared Multispectral Scanner (TIMS) offer an opportunity to study terrestrial mafic rock compositions with multispectral instruments where there is the opportunity to compare results to field based studies and laboratory analysis of collected samples. Both the TIMS and ASTER instruments have coarser spectral resolution than THEMIS. TIMS provides 6 bands and ASTER provides 5, in contrast to the 9 unique bands for THEMIS. However, if these rock types are distinguishable at lower spectral resolutions in field conditions on Earth, THEMIS is likely to exceed their abilities at Mars.

In this study, several basaltic to andesitic flows ranging in age from Tertiary (~20 Ma) to Quaternary (3 Ma) in the Gila Bend Mountains of southwestern Arizona are analyzed using ASTER and TIMS remote sensing data. The multispectral remote sensing data are deconvolved using laboratory mineral spectra resampled to the ASTER and TIMS instrument spectral resolutions. Comparison to laboratory and portable spectrometer data, as well as thin section analysis, provides the means to determine the accuracy of the deconvolution. Since these rocks differ in age, they exhibit variable amounts of weathering and rock varnish. These variable properties can be used to determine the proportion of the spectral signature that is due to coatings and weathering rinds versus host rock composition.