The time series of radon (222Rn) concentration in soil gas at a fault, together with the environmental parameters, have been analysed applying two machine learning techniques: (i) decision trees and (ii) neural networks, with the aim at identifying radon anomalies caused by seismic events and not simply ascribed to the effect of the environmental parameters. By applying neural networks, 10 radon anomalies were observed for 12 earthquakes, while with decision trees, the anomaly was found for every earthquake, but, undesirably, some anomalies appeared also during periods without earthquakes.
At Cazzaso (Friuli) in northeast Italy, radon (222Rn) activity concentration in soil gas in a borehole at a depth of 80 cm has been monitored continuously (at a frequency of once an hour) since May 2004, using a Barasol probe (Algade, France). In addition, environmental parameters (air and soil temperature, barometric pressure) have been recorded. The results have been evaluated and the relationship between radon levels and seismic activity is discussed. Correlation between radon concentration and barometric pressure has been observed. Preliminary results have shown a distinct radon anomaly prior to some earthquakes.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.