Predictability of 2012 Ahar-Varzaghan Earthquake, Using the RTL Method

Investigation of seismicity pattern before large earthquakes usually reveals the precursory activities. The seismic precursors have some advantage with respect to the non-seismic ones, including the wide range of their spatial and temporal domains. Two phases in seismicity change before a forthcoming, major earthquake are widely observed, which are the seismic quiescence and activation. The RTL (Region- Time-Length) analysis is a statistical method developed by SOBOLEV and TYUPKIN (199, 1997) to detect seismic anomalies preceding isolated large earthquakes. The RTL algorithm has been successfully used to detect precursory seismic activation and quiescence in different seismotectonic region around the world. We, therefore, consider the RTL algorithm a helpful technique to reveal the signi cantly precursory activity occurred before large earthquakes of Iranian plateau. The precursory seismicity in and around the epicentral zone of the 2012 Ahar-Varzaghan (Iran) dual earthquakes, by applying the RTL algorithm to earthquake catalogues derived from the Iranian Seismological Center for the period 2006 2012. First, we have determined the completeness magnitude for the selected region around the desired events. In order to avoid any bias in the results due to different magnitude conversion formulas, the original magnitudes in the catalogue (i.e. Mn) have retained. The RTL algorithm codes have been written using the Matlab software. By applying the RTL prognostic parameter, a clear quiescence stage followed by a period of foreshock activation is observed before two events of Ahar-Varzaghan, 2012. RTL performances are sensitive to the choice of spatial and temporal parameters. Following the method for automatic parameters selection developed by Chen and Wu, we calculated many sets of the RTL functions with various combinations of r0 and t0, then computed the correlation coefficients over pairs of the RTL functions. It is expected that high correlation between two RTL functions could be found when their values of r0 and t0 approach the optimal values mostly stabilizing the pattern of the RTL function. The best performance was obtained using r0 = 30 km and t0 = 1 year. However, the results are robust as we change r0 and t0 values. This implies that the RTL changes are not artifacts and could stem from some real and physical preparedness stage.