Abstract:
The aim of this work is to detect and accurately pick the onset time of P wave (primary wave) arrival in broadband seismograms. P-waves are the fastest kind of seismic waves, which reach the recording station first and therefore appear first on the seismogram. So its detection and onset time estimation is of utmost importance for accurately and quickly finding the epicenter of earthquake. In this work, phase picking algorithms for P-wave arrival detection and picking in Broadband Seismogram has been developed. The detection and picking is achieved using two methods which are quite deviant from each other. The first method is Phase picking using single component (1-C): (At the recording station there may be three component recording i.e. East-West, North-South and Vertical component or Single component recording i.e. Vertical component of ground motion): The algorithm used for 1-C phase picking is based on wavelet transform, Akaike Information Criteria (AIC) picker and waveform correlation (among the AIC values at different scales) based detection. In this method only Vertical component of seismogram is used because of its highest sensitivity to P-wave arrival. The second method is Phase picking using three components (3-C): The algorithm also employs wavelet transform and AIC picker. In addition a phase detection criterion is used for consistency check using voting mechanise developed for the purpose.
The algorithms developed in the present study have been tested on the seismic data obtained from seismically active Garhwal Kumaon Himalayan Region. The results obtained provide accurate and robust automatic picks on a large experimental data. Out of 450 P-phase onsets with SNR above 1 from Garhwal Kumaon Himalayan Region, the first algorithm gives 62 % detection with mean difference of 0.12 sec with respect to manual picks and the second technique gives 64 % detection with mean difference of 0.13 sec with respect to manual picks. Picks that extend beyond 0.5 sec with respect to manual picks are considered to be false alarms.
