Probabilistic Assessment Of Station Site Hazard Based On Initial Seismic P-wave Information

Earthquake is a sudden natural disaster and one of the major factors causing damage to buildings and people's lives and property.The occurrence of an earthquake is generally unpredictable.Only in the short time before the destructive seismic wave reaches the station the information such as the size and location of the earthquake can be forecasted and corresponding measures can be taken to reduce the loss.Based on the seismic data of Sichuan region,I studied the estimation methods of key parameters such as magnitude and epicenter distance in earthquake warning and evaluated the potential hazards of station sites.Firstly,the commonly estimation methods for earthquake magnitude and epicenter distance were introduced.In combination with factors such as geographical environment Bayesian methods were used to improve these common methods.The Bayesian Gibbs method and the empirical Bayesian Gibbs method were proposed.Secondly,based the 3 seconds data of the seismic P-waves the fitting effects of various estimation methods for magnitude and epicenter distance were compared and it was verified whether the various estimation methods for magnitude and epicenter distances had validity and accuracy in dealing with actual earthquake warning problems.The comparison showed that the empirical Bayesian Gibbs method is more accurate when it comes to solving actual problems.Then due to the deficiencies of the general ground motion intensity estimation model in processing actual data the leave-one-out cross-validation method was used to select and evaluate the general ground motion intensity estimation model and the model that error obey the skew normal distribution.Ground motion intensity estimation model based on skew normal distribution of error Terms was established.In the end,based on the ground motion estimation model and probabilistic analysis method,a probability assessment method for station site hazard is proposed,which could be a reference for earthquake warning and earthquake risk assessment and earthquake prevention.