Robust identification of periodic behavior in the time dynamics of short seismic series: the case of seismicity induced by Pertusillo Lake, southern Italy

In this paper, we investigate the time dynamics of the seismicity of the area of the artificial Pertusillo Lake (southern Italy) by using two advanced statistical techniques: the robust periodogram and the singular spectrum analysis. Such methodologies are well suited to unveil the main characteristics (periodicities and long-term trend) of short time series, for which the application of standard statistical techniques fail. Our findings are that the monthly number of earthquakes occurred in the investigated area from 2002 to 2012 reveals the presence of the annual periodicity, which is reasonably linked with the yearly loading/unloading operational cycle of the lake. This indicates that the seismicity of the Pertusillo Lake area is effectively induced by the time-varying water level of the artificial lake. This result from one side sheds light on the nature of the earthquakes occurring in the investigated area as reservoir-induced, but, more generally, from another side suggests a robust manner to investigate the time dynamics of short seismic series.