Identifications and removal of diurnal and semidiurnal variations in radon time series data of Hsinhua monitoring station in SW Taiwan using singular spectrum analysis

Abstract

Over the past three decades, anomalous temporal changes in radon concentrations have been reported in relation to earthquake occurrences. However, radon anomalies in all cases are not only controlled by seismic activity, but also by meteorological parameters which make isolation of earthquake precursory signals complicated. In the present study, characteristics of temporal variability of soil gas radon concentrations at the Hsinhua monitoring station, southern Taiwan, have been examined using singular spectrum analysis (SSA). In order to make continuity and regularity of the data before applying the SSA, the radon data were carefully edited for gaps and discontinuous jumps following intervals of malfunctioning of equipments. Digital filter has been applied in eliminating the long-term trend in the data and retains variations of <30 days. The radon changes exhibit dominant daily variations, which are controlled by atmospheric temperature inducted evaporation in surface water saturated soil (capping effect). The causal relationship is marked by a clear phase lag of 2–3 h in the sense that peak in the daily variation of radon succeeds the peak in temperature. Aperiodic variations in soil radon intensity in the range of 2–10 days are negatively correlated with temperature, while positively correlated with pressure. However, the negative correlation of the soil radon with temperature is found to be pseudo-effect arising due to interrelation between temperature and pressure.