Study On Regional Differentiation Of Ground Motion Attenuation Relationship

Despite the rapid development of ground motion observation technology and the increasing of strong ground motions data around the world, in recent decades, developing ground-motion predication equations for most regions with scarce records is still an outstanding problem because of the maldistribution of earthquakes and ground-motions data. Concerning this issue, scientific and reasonable adjustment based on ground-motion predication equations established by large amounts of records for those regions with rich ground-motions data for target regions is a feasible way. Nevertheless, the characteristics of ground-motions are distinct between different regions, and there may be regional differentiation between ground-motion predication equations, as the function to describe the relations between ground-motions and magnitude, distance, and site condition. Therefore, it’s critical to demonstrate the existence of regional differentiation of ground-motion models, to discuss the origin and major influence factors of regional difference, and how those factors affect the ground-motion models. Based on these cognitions, a quantitative description of regional differentiation can be modeled, which is of great significance in developing ground-motion models for the regions where strong ground-motion data are absent.In this study, regional characteristic parameters of earthquakes were inverted using strong ground-motion data. Based on the fundamental form ground-motion model, effects of regional characteristic parameters on coefficients and related terms of ground-motion model were analyzed and quantitative functions between these coefficients or items and regional characteristic parameters were constructed. Thereby, a regional differentiation model was developed, based on which adjustment equations to consider regional differentiation between various regions were determined according to regional characteristic parameters.The main content of the thesis are as follows:1. Combining the variability of regional geologic tectonic, crustal structure, weathering and erosion environment, the possible differences in source, attenuation, and local site effect were discussed on the base of general recognitions of ground-motion model. A definition of regional differences of ground-motion model was given according to the discussion.2. Four important earthquake regions around the world including western China, western U.S., Japan, and New Zealand were selected. Source effects, distance attenuation, and site effects for those four regions were extracted step-by-step from 5% damping acceleration response spectra. According to systematic comparison and study, the existence of regional differentiation was demonstrated.3. Generalized inversion technique(GIT) was used to obtain regional characteristic parameters and the effects of each characteristic parameter on ground-motions and ground-motion model were analyzed qualitatively. The Comparison of regional characteristic parameters from the four regions shows that, for different regions, average stress drop of earthquakes, crustal depth, Q value of S-wave in crust, and high-frequency decay factor are markedly different.4. Point-source stochastic simulation method was used to simulate ground motions with different regional characteristic parameters. A ground-motion predication equation was established. The effects of stress drop ??, crustal depth, Q value of S-wave in crust, and high-frequency decay factor0? on the magnitude scaling, distance scaling, and constant term of ground-motion model were discussed, and then, the quantitative describing functions of the effects were established. To determine the relationship between site effect of soft rock with site regional characteristic parameters 0? and Z2.5, square-root-impedance method was used to analyze the site amplification of soft rock with different 0? and Z2.5.5. The regional differentiation model to consider the differences of source, crustal media, and local site characteristics of various regions was developed, which is the basis of adjustment to ground-motion model. The influences of each regional characteristic on regional differentiation were discussed, and the intra-region variability and inter-region differentiation of ground-motions and ground-motion models was compared.6. Adjustment equations for ground-motion models between different regions were determined utilizing the regional differentiation model. Observed records and ground-motion models from target regions were used to check the correctness and validity of the adjustment equations. The comparisons reveal that, the predicted values of ground-motion models adjusted by our regional differentiation model conform to the observed records very well. What’s more, the ground-motion models adjusted from different reference regions were compared, and the results showed high consistency.7. Using regional differentiation model, we developed a ground-motion model for the Eastern China by regional differentiation adjustment. Comparing with Yu(2006) model for Eastern China, predicted ground-motions of our model are with higher amplitudes totally, especially for small-moderate magnitude(Mw≤6). The predicted ground-motions of our model show much slow attenuation with distance at T≥0.5s. The predicted ground motions for 2012 Gaoyou-Baoying earthquake by our ground-motion model were in agreement with obversed records well.