RETRIEVAL OF MULTISPECTRAL INFRARED EMISSIVITY FROM THERMALLY-MIXED VOLCANIC SURFACES FOR MORE ACCURATE COMPOSITIONAL MAPPING

The Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) sensor was launched in December 1999 as one of five instruments on the Terra satellite, part of NASA’s Earth Observing System (EOS), and has proven effective for the detection and monitoring of volcanic eruptions and their associated products. However, continuous advancement in analytical techniques remains essential to better understand the data acquired from active volcanoes. These images commonly contain features that are below the spatial resolution of the data; are more indicative of the state of volcanic unrest; and tend to saturate thermal infrared (TIR) sensors due to their high emitted radiance. In addition, compositional, textural, and thermal heterogeneities can vary greatly over the area of just one 90m ASTER TIR pixel, and without more advanced techniques the accurate retrieval of the temperature and composition of that surface becomes impossible. Previous studies of isothermal, compositionally heterogeneous pixels have shown that the emitted radiance mixes linearly, such that the spectrum is a combination of the energy from each component in proportion to its areal percentage. However, where thermal mixing occurs, the linear approach is no longer valid. We present a new approach for these thermally mixed pixels in an ASTER TIR scene using predetermined thermal components in order to model the associated errors in the emissivity spectra. These results have been used to determine temperature thresholds and corrections for basalt spectra from the active flow fields of Kilauea volcano, Hawaii. ASTER data of Kilauea acquired during an active effusive phase in October of 2006 have been deconvolved with the new approach, which identifies the thermally mixed pixels and then separates them into their hot and cool thermal components using the higher spatial resolution short wave infrared (SWIR) bands. Results provide more accurate non saturated temperature estimates as well as corrections to the emissivity for better compositional and textural mapping of the surface. This approach also serves as rapid means for identifying surface breakouts and minimizes the processing time, therefore allowing critical information to be disseminated expeditiously. It could prove to be an invaluable tool for understanding other high temperature processes and hazards, which are commonly obscured in low to medium spatial resolution orbital data sets.