P-Wave Detection Alert System for Multi-Seismic Stations

This paper proposes an automatic P-wave picking algorithm based on discrete wavelet transform (DWT). This algorithm is applied to multi-station seismic data from the Egyptian National Seismic Network (ENSN) to enhance the accuracy and reliability of P-wave detection. The real-time detection of P-waves suggests a focus on practical applications for implementing earthquake early warning (EEW) systems. The integration of this algorithm into the system generates an event report and sends it via email, containing all the stations that recorded this event within the sub-network. The proposed algorithm has the major advantage of providing accurate P-wave arrival times with minimal processing time. In addition, this algorithm can detect even the smallest events, such as micro-earthquakes, due to different frequency sub-band analyses of discrete wavelet transform (DWT) with varying thresholds for every station. These features might be suitable for developing an Earthquake Early Warning (EEW) system. In a challenging test environment, the system achieved a remarkable 95.97% detection rate with minimal false alarms in the Cairo subnetwork, 80.88% in the Red Sea subnetwork, and 94.2% in the North Coast subnetwork.