Non-linear analysis of soil 222Rn time series recorded at Jiaosi in north-east Taiwan for possible earthquake precursor study

Abstract

A nonlinear technique was used to analyse soil radon concentrations measured over a year and four months in Jiaosi, northeast Taiwan, to find precursory signals for regional earthquakes. The recorded data demonstrated that numerous meteorological and geophysical parameters influence radon gas emission from the soil, resulting in a complex soil radon time series. Empirical Mode Decomposition (EMD) was used to remove periodic factor components before applying the Hilbert-Huang Transform (HHT) to find pre-seismic anomalies. The discovered anomalous variations were associated with regional earthquakes of magnitude 5 or more that occurred within 100 km of the monitoring site. The effect of precipitation on soil radon emissions has also been discussed. The results show that the adopted technique can be used to analyse radon concentrations in soil gas for the prediction studies of regional earthquakes.