Study of seismotectonic and seismicity of Isfahan region

This paper presents the seismicity and seismotectonic characteristics of Esfahan and adjoining regions, an area bounded between 49.5-54o E and 31-34.1o N. According to Mirzaei et al. (1998) this region is placed between two major seismotectonic provinces: Zagros and Central-East Iran. The boundary between these two provinces is known to be the Main Zagros Reverse Fault. Geological maps with scales of 1:100000 and 1:250000 were used to provide the faults map of this region. We determined several probable faults in the region that can help us to introduce the potential seismic source more precisely. Among a total of eighteen focal mechanism solutions of earthquakes with Mw ≥ 4.3 in the region of interest, one event is normal, six events show dominant strike-slip components, and eleven events have dominant reverse components, which confirms the dominance of reverse/thrust faulting. Using global database of earthquakes provided by USGS/NEIC and ISC, as well as catalog of historical (pre-1900) and early- instrumental (1900-1963) earthquakes provided by Ambraseys and Melville (1982), a uniform catalog of earthquakes for the interest region has been provided that involves 170 instrumental and 10 historical earthquakes. To unify the scale of earthquake magnitude for each seismotectonic province, we established empirical relations to convert mb to Ms. Because historical earthquake in Iran have been defined in Ms and lack of events with magnitudes above the saturation level of Ms in the interest region, surface wave magnitude is the most appropriate magnitude scale for the region, and on a broader scale, is appropriate for Iran. In order to establish empirical relations between mb and Ms we applied the orthogonal regression (OR) method that takes into accounts the errors of measurements in both variables. Based on this uniform catalog, seismicity of the interest region was studied, and seismicity parameters were calculated utilizing the method proposed by Kijko and Sellevoll (1992), in which one can consider magnitude uncertainty and completeness of data in calculations. In order to achieve more reliable results, the completeness of catalog and uncertainty of magnitudes have been estimated and considered in our calculations. In this method, dataset of the catalog should be divided into extreme and complete part, and each complete part can be subdivided again into several complete parts that have their own completeness threshold. In this work, the whole data was separated into six complete parts with threshold magnitudes between Mc=3.2 (for events after 1996) to Mc=5 (for events after 1939). To estimate the magnitude threshold (Mc) combination of two methods were used. The first, is maximum curvature (MAXC) and the second, is goodness of fit test (GFT). In cases where lack of data does not allow using GFT, only MAXC method is used. Magnitude uncertainty of each event is considered to be 0.2 and 0.3 for modern-instrumental and early-instrumental earthquakes, respectively, when Ms has been assigned directly based on seismogram readings. For the events that their surface-wave magnitudes have been obtained by empirical conversion relations, magnitude uncertainty is considered to be 0.4. In the case of historical events, uncertainties are considered to be in the order of 0.4 to 0.8 (Mirzaei et al., 1997a), based on their quality (a, b, c, d) that was assigned by Ambraseys and Melville (1982). For Central-East Iran part of the interest region, the results show that b-value is equal to 0.81 ± 0.12, λ is equal to 0.48 ± 0.129 and Mmax=7.8 ± 1.74. For Zagros part, b-value is equal to 0.95 ± 0.05, λ is equal to 4.976 ± 0.413 and Mmax=7.4 ± 0.67. The impact of classification of data on seismicity parameters has been investigated, and the results show that it has a significant impact on the b-value (or β) and recurrence rate of the seismic events (λ); while its effect on Mmax is negligible. Furthermore, classification of complete part of the catalog significantly decreases the uncertainty in the evaluated parameters of Mmax, λ and b (or β). The minimum impact on uncertainty of parameters appears in b, and the maximum appears in Mmax.