CHARACTERIZATION OF VOLCANIC THERMAL ANOMALIES BY MEANS OF SUB-PIXEL TEMPERATURE DISTRIBUTION ANALYSIS

Abstract

The simultaneous solution of the Planck equation (involving the widely used “dual-band” technique) using two shortwave infrared (SWIR) bands allows for an estimate of the fractional area of the hottest part of an active lava flow (f h) and the background temperature of the cooler crust (T c). The use of a high spectral and spatial resolution imaging spectrometer with a wide dynamic range of 15 bits (DAIS 7915) in the wavelength range from 0.501 to 12.67 μm resulted in the identification of crustal temperature and fractional areas for an intra-crater hot spot at Mount Etna, Italy. This study indicates the existence of a relationship between these T c and f h extracted from DAIS and Landsat TM data. When the dual band equation system is performed on a lava flow, a logarithmic distribution is obtained from a plot of the fractional area of the hottest temperature vs. the temperature of the cooler crust. An entirely different distribution is obtained over active degassing vents, where increases in T c occur without any increase in f h. This result indicates that we can use scatter plots of T c vs. fh to discriminate between different types of volcanic activity, in this case between degassing vents and lava flows, using satellite thermal data.