DETRENDED FLUCTUATION ANALYSIS FOR CHARACTERIZING VOLCANIC TREMOR PRIOR TO ERUPTIONS
The analysis of the seismic signals recorded near volcanoes is one of the most effective tools for predicting volcanic eruptions. During the onset of the eruptive process, an increment in the seismic activity is usually observed. In this study, we use the detrended fluctuation analysis (DFA) method to characterize the seismic energy recorded in seismograms preceding volcanic eruptions. We used seismograms from three different active stratovolcanoes: the Mount Pavlof and Mount Shishaldin in Alaska, USA, and the Bezymianny volcano in Kamchatka, Russia. We present our preliminary result on how the variations in seismic energy, characterized using the DFA scaling exponent (), can serve as the basis for developing new volcanic eruption alert algorithms based on the fluctuation analysis methods.