Earthquake precursory study using decomposition technique: time series soil radon monitoring data from the San-Jie Station in Northern Taiwan.

Soil-gas radon concentration data recorded for one year at the San-Jie monitoring station in northern Taiwan has been analysed in this study. To detect genuine pre-seismic anomalies, the hourly radon data was decomposed into seasonal, trend, and residual components. Anomalous fluctuations in hourly and residual radon data have been correlated to earthquakes and rainfall. Preliminary results indicate that this procedure can be useful for the identification of pre-seismic anomalies in radon time series that may be masked by rainfall or other meteorological factors. Furthermore, Pearson correlation was employed to determine radon relationships with temperature, pressure, precipitation, and humidity. [ABSTRACT FROM AUTHOR]