Unified approach for evaluation of horizontal site amplification factors with special reference to history of studies on the effects of surface geology on seismic motion

Abstract

Following the 1923 Kanto earthquake in Japan, Japanese researchers noticed the strong effects of surface geology on seismic motion (ESG) and began to investigate these effects to quantify the site amplification factors (SAFs) associated with soft surface sediments. On the other side of the Pacific Ocean, ESG received limited attention until the 1985 Michoacan, Mexico earthquake revealed significant long-period amplification inside Mexico City that manifested as the source of devastating damage to high-rise buildings. Since then, seismologists and earthquake engineers have performed a lot of studies on various ESG issues worldwide. We have not yet reached common conclusions on how to quantitatively predict SAFs over a broad frequency band of engineering interest, 0.1 to 20 Hz, for moderate to strong input from different types of earthquakes in different tectonic settings. However, we found here several basic guidelines useful for successfully modeling ground motions as a common approach to ESG studies. First, in this letter, we briefly review our history of understanding ESG, which is closely related to the key settings required for reliable quantifications of SAFs, and then introduce various emerging techniques for broadband quantitative evaluations of SAFs based on the vast amount of observed ground motions primarily from dense Japanese strong-motion networks. Based on the findings of our investigation and the physical relationships behind the parameters, the authors would like to recommend that researchers on ESG and related topics would refer to the five basic guidelines proposed in the conclusions for the successful implementation of techniques to delineate SAFs in a specific region of interest, such as the use of Fourier spectra instead of response spectra. We have started applying the proposed techniques to regions outside Japan. The implementation of the statistical validation exercises will follow. Graphical Abstract: [Figure not available: see fulltext.].