LAUSR

Abstract Surface wave tests such as the multi-channel analysis of surface waves (MASW) and microtremor array measurements (MAM) offer a fast and convenient way of measuring the shear wave velocity (Vs) of the ground for geotechnical site investigation due to their non-intrusive nature. However, it has been widely reported that the accuracy and resolution of the Vs profile depend on the construction of the dispersion image and its interpretation. While many advanced computational techniques have been proposed to process surface waves into a final Vs profile, we opted for the classical ‘picking and inverting’ approach in the current study. This is in view of the lower required computation power as well as the relatively uniform geology of the site. Surface wave tests (MASW, MAM and horizontal-to-vertical spectral ratio) were carried out at a remote reclaimed land in Singapore. A systematic methodology for the modal analysis of Rayleigh waves based on combined MASW and MAM tests is presented. The MAM dispersion image was constructed using the phase shift method, the extended spatial autocorrelation (ESAC), and the refraction microtremor (ReMi) method but a meaningful dispersion curve could only be obtained using the phase shift. Due to the modes of energy in the low frequency regime being unclear in the dispersion image, an idealized synthetic model of the site was set up to guide the mode identification process. The combined dispersion curve and HVSR profile were jointly inverted to construct the final Vs profile, which agreed quite well with the seismic cross-hole tomography experiments and bore-log information. It is also demonstrated that the inversion of the fundamental mode alone led to slight errors and low resolution in the final velocity profile, even for such a simple soil stratigraphy.