VALIDATION OF MICROTREMORS ARRAY METHOD

Lijing SHI,Xiaxin TAO,Jisheng ZHAO

DOI: https://doi.org/10.3321/j.issn:1000-6915.2006.08.025

2006-01-01

Abstract:By the microtremors array observation and velocity structure inversion in the field of an engineering project, the accuracy of microtremors array methods (MAM) for exploring the detailed shallow velocity structure is studied with observation system, dispersion curve extraction and inversion methods. The shallow S-wave velocity structure is inversed by the hybrid method of genetic algorithm (GA) and simplex algorithm (SA) from the surface wave dispersion curve, which is inferred by the spatial auto-correlation method (SAC) from microtremors array records. The studies show that the relative average error of each layer is about 20% compared with that detected by borehole method. The average velocity obtained by SAC method and frequency wave-number method (F-K) is almost the same. However, F-K method only gives an average velocity for the upper 20-meter layer. The accuracy is compared with that of spectral analysis of surface wave (SASW) method. In the fields of geotechnical engineering and earthquake engineering, the direct goal of velocity structure exploration is site dynamic characterization. From the viewpoint of site seismic responses, the accuracy of MAM is further validated by ID equivalent linearized analysis. The analytical results show that the difference between the response spectra from inversed velocity structure and that measured in borehole is really acceptable, whereas that for the simplified single layer model with the average velocity is quite large. The study suggests that it is not a good idea in seismic design to characterize site condition only by the average velocity; and the MAM has the potential to be an effective method for the shallow velocity structure exploration.