M S Ramsey, J. H. Fink, R. Wessels, M. Poland, and S. Ruff (Dept. of Geology, Arizona State Univ., Box 871404, Tempe, AZ 85287-1404; ph. 602-965-1790; email: ramsey@elwood.la.asu.edu)

Silicic domes commonly form the late stage products of an eruption sequence at composite volcanoes. Variations in the surface texture may indicate changes in volatile content or lava extrusion rate and can signal a collapse that initiates pyroclastic flows. These surface changes are commonly difficult to observe due to the dangers of continued activity. The use of thermal infrared (TIR) remote sensing provides a measure of safety for field investigators and is uniquely sensitive to compositional, textural, and thermal changes. However, TIR data have rarely been used to investigate active domes mainly due to the lack of repetitive high resolution, multispectral data. The first NASA Earth Observing platform, scheduled for launch in mid-1999, will carry the Advanced Spaceborne Thermal Emission and Reflectance Radiometer (ASTER) instrument. ASTER will provide the first global multispectral imagery from visible to TIR wavelengths, with spatial resolutions that increase from 90 m (TIR) to 15 m (visible), and along-track DEM generation. Active volcanoes are a primary target set, however complications are expected from atmospheric absorption, DEM location accuracy and surface variations. To better understand these factors, calibration field studies were initiated on the Holocene domes of Medicine Lake volcano, CA.

TIR emissivity spectra of a lava surface can be deconvolved to produce an estimate of the vesicularity using a new technique developed by the lead author. To evaluate this approach as well as quantify atmospheric effects, we collected spectra using NASAs portable field spectrometer. This is the first time TIR spectra of undisturbed silicic dome surfaces have been acquired and used to calibrate airborne TIR data. We also employed two real-time differential GPS units for the sub-meter location accuracy of several calibration pixels, and a notebook computer to correct 30-80 m georeferencing image errors. Results of this field study are important because we have been able to quantify each sub-pixel element and refine our TIR vesicularity estimates. We plan to tie these calibrations into ASTER data over the same areas next year, and use those results to expand the study to recently-active volcanoes. Although detailed field studies will rarely be possible, this study will lay the groundwork for calibration of ASTER imagery.

------------------
Presented at: American Geophysical Union Fall Meeting
Date: 1998