Automatic Event Detection and Picking of P, S Seismic Phases for Earthquake Early Warning: A Case Study of the 2008 Wenchuan Earthquake

We develop an automatic seismic phase arrival detection and picking algorithm for the impending earthquakes occurred with diverse focal mechanisms and depths. The polarization analysis of the three-component seismograms is utilized to distinguish between P and S waves through a sliding time window. When applying the short term average/long term average (STA/LTA) method to the polarized data, we also construct a new characteristics function that can sensitively reflect the changes of signals' amplitude and frequency, providing a better detection for the phase arrival. Then an improved combination method of the higher order statistics and the Akaike information criteria (AIC) picker is applied to the refined signal to lock on the arrival time with a higher degree of accuracy. We test our techniques to the aftershocks of the Ms8.0 Wenchuan earthquake, where hundreds of three-component acceleration records with magnitudes of 4.0 to 6.4 are treated. In comparison to the analyst picks, the results of the proposed detection algorithms are shown to perform well and can be applied from a single instrument within a network of stations for the large seismic events in the Earthquake Early Warning System (EEWS).