Temporal and spatial variations of seismicity parameters before the occurrence of MN≥5 earthquakes in northeast Iran

In the present study, we investigate some statistical features of earthquake precursors, namely the variations of the b-value and Z-value parameters in time and space, before the occurrence of MN≥5.0 earthquakes in northeast Iran. Based on the literature, the b-value of the Gutenberg–Richter distribution is connected to the field of seismic stress, so that it decreases linearly with increasing differential stress and vice versa. Still, a high differential stress at later stages of an earthquake cycle causes failure of large patches. The Z-value parameter is also associated with seismic quiescence periods which imply the regional preparedness for the occurrence of strong earthquakes.    It is worth mentioning that our study covers ten MN≥5.0 target earthquakes that occurred between 2010 and 2022. However, due to the large number of target earthquakes, the text body is focused on one of the events accompanied with all investigated anomalies, namely the 2015 Kashmar MN 5.2 earthquake.    In order to study the anomalies of seismic parameters, an earthquake catalog, reporting post-2006 MN≥0.3 events, was extracted from the Iranian Seismological Center. As the raw catalog includes small size events, it was preprocessed via removing possible non-tectonic events. These events, being mostly quarry blasts, have been statistically recognized by an unrealistic increase in the number of events during working hours (i.e. 6.00 to 16:00 O’clock). Accordingly, the Rq method, implemented in the ZMAP software package, has been used for removing possible quarry blasts. The estimation of seismic parameters has been done by assuming the Poisson distribution of the occurrence of earthquakes. Therefore, dependent events (i.e. aftershocks and foreshocks) were also eliminated from the used catalog, using a declustering procedure. However, in the case of the 2015 Kashmar earthquake, the de-quarrying and declustering procedures remove 6859 events out of 22143. Having estimated the magnitude of completeness (Mc) for the region of the Kashmar earthquake, we removed earthquakes with magnitudes less than Mc=1.8. Furthermore, as earthquake catalogs commonly involve in a sort of temporal fluctuation of Mc which is mostly due to increase in the number of seismographs over time, the stability of Mc=1.8 for the region has also been checked. In total, the residual catalog includes 7606 MN≥1.8 earthquakes over the Kashmar region.    Our results indicate that at least one of the investigated anomalies can retrospectively be observed prior to all target earthquakes. Still, in some cases both anomalies can simultaneously be detected in the space and time domains. Moreover, the results show that regions characterized by both b-value≤1.0 (after experiencing a temporally decreasing trend of b-value) and Z-value≥3.0 are most prone areas for future MN≥5.0 earthquakes. Furthermore, it has been found that there exists a significant correlation between the explored anomalies, so that the correlation coefficient is higher than 0.6 for some cases. In conclusion, our results confirm that temporal and spatial variations of seismicity parameters can somehow be interpreted as effective indicators for the areas prone to earthquakes with significant earthquakes.