Physically Based Modeling of Spatially Variable Seismic Ground Motions

地震地面运动的空间变化对桥梁、管道等生命线的响应有重要影响；如果结构的支撑受到相同的地震运动，它可能会在结构中引起显著的附加荷载。地震地震动的空间变化是由于地震地震动波形在地面上的表观传播及其在支承结构不同位置上的形态差异造成的。在之前的一项研究中，PI开发了一种替代方法，用于研究密集仪器阵列记录的空间可变地震地面运动。它将地震运动建模为正弦函数的叠加，用它们的振幅、频率、波数和相位来描述。将该方法应用于台湾洛东SMART-1阵列记录的数据，发现共相干分量与实际运动的近似程度非常令人满意，并且空间变异是由与波通过站点地形向上传播相关的阵列站的到达时间延迟引起的，以及由共同分量特征周围运动的振幅和相位的变化引起的。所开发的方法同时考虑了地震运动的振幅和相位变化。将其应用于记录数据，识别振幅和相位变化之间的相关模式，围绕运动的相干近似；这一观察结果可能导致空间变异性描述的重大发展。然而，该方法只应用于一个地点，为了推广其结果，有必要将其应用于遭受各种地震的不同地点。因此，本研究使用在岩石和土壤地点记录的大量额外数据集。根据地震震级进行区分：对同一地点的小地震和大地震（如主地震和余震）进行调查。场地（“湿”和“干”场地）地下水位的影响也被考虑在内。

The spatial variation of seismic ground motions has an important effect on the response of lifelines, such as bridges and pipelines; it may induce significant additional loads in the structure than the ones obtained if the structure's supports are subjected to the same seismic motions. The spatial variation of seismic ground motions results from the apparent propagation of the waveforms on the ground surface and the differences in their shape at the various locations of the supporting structure.In a previous study, the PI developed an alternative methodology for the investigation of spatially variable seismic ground motions recorded at dense instrument arrays. It models the seismic motions as superpositions of sinusoidal functions, described by their amplitude, frequency, wavenumber and phase. The application of the methodology to data recorded at the SMART-1 array in Lotung, Taiwan, revealed that the common, coherent component approximates to a very satisfactory degree the actual motions,and that the spatial variability is caused by arrival time delays at the array stations associated with the upward traveling of the waves through the site topography, and by variabilities in both the amplitudes and phases of the motions around the common component characteristics.The developed methodology considers simultaneously the amplitude and the phase variation in seismic motions. Its application to recorded data recognized correlation patterns between amplitude and phase variabilities around a coherent approximation of the motions; this observation may lead to significant developments in the description of spatial variability. However, the methodology has been applied to one site only, and, for the generalization of its results, it is necessary that it be applied at different sites subjected to various earthquakes.Therefore, the present study uses extensive additional sets of data recorded at rock and soil sites. Differentiation is made in terms of the earthquake magnitude: small and large events (e.g., main event andaftershocks) at the same site are investigated. The effect of the water table at the sites ('wet' and 'dry' sites) is also taken into consideration.