2006 Philadelphia Annual Meeting (22–25 October 2006)

Paper No. 5
Presentation Time: 1:30 PM-5:30 PM

MAPPING THE DISTRIBUTION OF ALTERATION MINERALS IN AN ACTIVE GEOTHERMAL SETTING USING SATELLITE-BASED SPECTRAL REFLECTANCE: AN EXAMPLE FROM THE UZON CALDERA, KAMCHATKA, FAR EAST RUSSIA


CROWE, Douglas1, JORDAN, Thomas2, BAKER, M. Scott1 and SCHROEDER, Paul3, (1)Department of Geology, Univ of Georgia, Athens, GA 30602, (2)Department of Geography, University of Georgia, Athens, GA 30602, (3)Department of Geology, Univ of Georgia, 210 Field Street, Athens, GA 30602, crowe@gly.uga.edu

The Uzon caldera, located 300 km north of Petropavlovsk, Kamchatka, hosts numerous active geothermal fields. The plumbing system beneath the fields is complex, such that individual pools within a given field that are < 5m apart can vary substantially in terms of temperature, pH, Eh, major ions, and stable isotope ratios. This leads to substantial differences in alteration mineralogy around each pool, and makes mapping the spatial distribution of these alteration facies very difficult on a large scale. However, with the advent of commercially available satellite imagery, we are able to produce maps showing the distribution of alteration facies at high resolution (+/- 0.6 m).

Initial use of ASTER image data demonstrated that the 15m (VNIR) and 30m (SWIR) pixel size was insufficient to distinguish changes in mineralogy. For this reason, high resolution Quickbird images were acquired (panchromatic and VNIR bands). The data were corrected for atmospheric interference and then processed using the Spectral Analysis Workstation in Leica Geosystems Imagine 8.7, and displayed using conventional ArcMap software. Data were initially evaluated using an unclassified search approach, where each pixel was color-coded depending on the degree of similarity to all other pixels. Although this provides maps that show differences in mineralogy, it does not identify any given phase, and is limited in use. Therefore, the reflectance spectra for each pixel was then compared to the USGS and JPL mineral databases, and best fit spectra were generated for each pixel. Maps were then made of specific thermal fields showing the distribution of discrete phase assemblages.

Maps for the North, East, Orange and Central thermal fields show distinct mineral zonation. Of particular note is the ability to see overall redox changes related to proximity to individual pools. Typically, reduced phases such as pyrite occur in the immediate vicinity of active pools, phases that are more oxidized, such as elemental sulfur, occur outboard from the reduced zones, and the most oxidized phases, including hematite, scorodite, and alunite group minerals, are furthest from the active pools. Additionally, we have been able to distinguish illite and alunite group minerals from their ammonium-bearing analogs solely on the basis of spectral reflectance signatures.