Discrimination of quarry blasts and microearthquakes using adaptive neuro-fuzzy inference systems in the Tehran region

Abstract

The purpose of this research is to demonstrate the use of Adaptive Neuro-Fuzzy Inference System (ANFIS) for discrimination between quarry blasts and microearthquakes in the Tehran region using data from the Broadband Iranian National Network Center (BIN). In the south and southeast of Tehran, a large number of quarry blasts contaminate the earthquake catalog. In order to identify the real seismicity (tectonic earthquakes) in the region, we need to discriminate quarry blasts from natural earthquakes in the catalog. ANFIS classifiers were used for identifying and categorizing the two kinds of seismic events. Neuro-fuzzy coding was applied using six features as ANFIS inputs. We investigated waveforms of 506 seismic events (0.9<ML<4.2) from the BIN network and determined differences between earthquakes and quarry blasts based on input features. These features include the original time of event, distance (source to station), latitude of epicenter, longitude of epicenter, magnitude, and two features for spectral analysis of their seismograms. The results of this study indicate that, out of a total of 506 seismic records, the neuro-fuzzy ANFIS approach identified 24.5% (124 events) and 75.5% (386 events) were quarry blasts and natural earthquakes, respectively. Our revised earthquake catalog provides improved data for realistic earthquake hazard assessment. Moreover, active faults will be detected correctly.