Love wave dispersion and associated analytic partial derivatives are theoretically derived using the reflection/transmission (R/T) method. The numerical implementation is then applied to linearised inversion of synthetic and field data. For simple cases dominated by the fundamental mode, Love wave sensitivity and inversion stability is higher than the Rayleigh wave dispersion. However, in general, Rayleigh wave inversion converges more rapidly than Love waves, although similar low misfits can be achieved. When assumed interfaces are used, the inversion of fundamental-mode Love wave dispersion of the normally dispersive profile provides a more accurate result, because the Love wave dispersion is independent of the Poisson's ratio. In more realistic, irregularly dispersive profiles, deep structure is only interpreted using higher-mode Love and Rayleigh wave dispersion.A field test over a shallow geologic fault with coincident Love and Rayleigh wavefields shows the fundamental-mode Love wave dispersion above 20 Hz to have at least 10% higher phase velocities than Rayleigh waves. However, at lower frequencies, Love wave dispersion is at least 10% slower. The resulting inverted models show up to 25% difference in shear-wave velocity. This is attributed to transverse isotropy of V SH and V SV in shallow fluvial sediments and allows improved geological horizon interpretation and soil discrimination over conventional, single-component surface wave inversion. A second test at a seismograph station site shows the Love wavefield less scattered and mode identification is simple, unlike Rayleigh waves over the same line, and the inversion results correlate well to downhole shear-wave velocity logs.

No abstract

No abstract

Joint inversion codes of the effective Rayleigh wave and multimode Love wave dispersion curves based on modified GA have been developed. We focus on the attempt to improve the estimated shear‐wave velocity profile for a reversal subsurface structure. Two different profiles representing the high‐velocity layer (HVL) and low‐velocity layer (LVL) cases are used. Effective Rayleigh wave and multimode Love wave dispersion curves are synthesized from these profiles using full P‐SV waveform and full SH‐waveform reflectivity, respectively. The dispersion curves are then inverted using a modified genetic algorithm with two different inversion approaches, namely single inversion of Love/Rayleigh dispersion curves and joint inversion of Love and Rayleigh dispersion curves. To asses the accuracy of each inversion approach, differences between the true and inverted shear‐wave velocity profile are quantified in terms of shear‐wave velocity error, ES. Our numerical modelling showed that the shear‐wave velocity errors of the joint inversion approach are relatively smaller than the single inversion approach. These errors indicate that the accuracy of shear‐wave velocity reversal can be improved by jointly inverting Rayleigh wave and Love wave dispersion curves. All of these approaches have been applied on the real data acquired at a site in Niigata prefecture, Japan. In this field example, our inversion results show good agreement between the calculated and experimental dispersion curves and can well detect LVL targets.

No abstract