Wavelet Based Automatic Phase Picking Algorithm for 3-Component Broadband Seismological Data

A phase picking algorithm has been developed for P-wave arrival detection and picking on a 3-component broadband seismogram. The algorithm is based on Wavelet Transform, Akaike Information Criteria (AIC) picker and voting mechanism. The mathematical functions have been used to cut up data into different frequency components and each component with a resolution matched to its scale is studied. Continuous Wavelet Transform (CWT) has been applied to pre-filtered 3-component broadband seismograms to compute scales of interest. Wavelet coefficients have been calculated in a sliding time window and AIC picker is applied to these coefficients. Consistency of 3 components picks is the primary criteria of phase detection. Further, for phase picking, AIC picker is applied on the filtered seismogram in the same time window where phase is detected. The comparison of the present algorithm has been checked with the time domain automatic phase pickers and manual picks by seismologists. The algorithm has been tested on regional earthquake data acquired through deployment of broadband seismological station in Garhwal Kumaon Himalayan region and shows a good agreement between analyst picks and auto picks.