Rapid determination of source parameters for the 2017 Mw 8.2 Mexico earthquake based on high-rate GPS data

摘要

Aiming to rapidly determine source parameters (i.e., static offset, seismic wave arrival, moment magnitude, and hypocentral location) for the Mexico 8.2 earthquake, the records of 7 high-rate GPS stations are utilized. Smoothness Priors Method (SPM) is introduced to rapidly extract the static offset from GPS displacement waveforms, and the result suggests this approach is feasible. The seismic wave arrival detection is carried out with the aid of S-transform, and the propagation velocity of seismic wave detected by most GPS stations are within 4-5 km/s, implying the seismic wave captured by GPS station may be S wave or surface wave. After that, an empirical regression model is adopted to characterize the moment magnitude for this earthquake, and this empirical formula can obtain reliable magnitude in comparison with the reference magnitude. The convergence time of average moment magnitude is 298 s, suggesting that a reliable and robust magnitude (Mw 8.37) can be estimated by the 7 GPS stations with about 298 s after earthquake occurrence. Considering the influence of the spatial distribution of GPS stations, 4 nearest GPS stations evenly distributed on both sides of the fault are selected to determine the warning-magnitude for the Earthquake Early Warning (EEW). A reliable and robust moment magnitude (Mw 8.2) can be estimated by the 4 stations with about 251 s, which is 57 s ahead of 7 stations. Finally, the coordinates of 6 GPS stations and corresponding seismic wave arrival time are utilized to determine the hypocentral location, and the latitude and longitude of estimated location is 14.925 degrees N and 93.765 degrees W, which is 17.9 km from the reference location. The results indicate that the source parameters required for EEW can be rapidly determined based on high-rate GPS displacement waveforms, and integrating real-time GPS into a joint EEW system will be crucial going forward. (C) 2019 COSPAR. Published by Elsevier Ltd. All rights reserved.