Scale of fluctuations (SOFs) of spatially variable soil properties have been regarded as one of the important parameters for performing reliability-based design in geotechnical engineering. However, the information required to estimate the SOFs in practice is limited, especially in the horizontal direction. In this study, the potential use of Multichannel Analysis of Surface Waves (MASW) to estimate the SOFs of shear wave velocity (VS) in horizontal direction is investigated through a series of numerical investigations. 2D random field models with a vertically increasing trend of VS were first simulated with different levels of variability controlled by the coefficient of variation (COV) and SOF. Afterwards, a large number of numerical MASW tests were performed using a 2D finite difference method, where the survey lines were progressively shifted. Results show that the COV of VS can be determined from the scatter of the dispersion curves, whereas the horizontal SOF of VS can be appropriately estimated from the phase velocity profile presented in the wavelength form. Additionally, in-situ MASW tests were conducted to estimate the horizontal SOF, and the obtained results align with those estimated by different methods. It is highlighted that the accuracy of the estimation depends on survey length. The interpretation using a long array reflects an averaged site condition of the survey area, thus losing the variability information. Favorable predictions are produced when survey length is shorter than 1.0 SOF. However, it should be noted that use of short survey length may limit the depth of investigation.

中文翻译：

---

通过表面波的多通道分析估计横波速度水平变化的新方法


空间变化土壤特性的波动尺度 （SOF） 一直被认为是岩土工程中执行基于可靠性的设计的重要参数之一。然而，在实践中估计 SOF 所需的信息是有限的，尤其是在水平方向上。在本研究中，通过一系列数值研究，研究了表面波多通道分析 （MASW） 在水平方向上估计横波速度 （VS） 的 SOF 的潜在用途。首先用变异系数 （COV） 和 SOF 控制的不同水平的变异性模拟具有 VS 垂直增加趋势的 2D 随机场模型。之后，使用 2D 有限差分法进行了大量数值 MASW 测试，其中测量线逐渐移动。结果表明，VS 的 COV 可以从色散曲线的散射确定，而 VS 的水平 SOF 可以从波长形式呈现的相速度分布中适当估计。此外，还进行了原位 MASW 测试以估计水平 SOF，获得的结果与不同方法估计的结果一致。需要强调的是，估计的准确性取决于调查长度。使用长数组的解释反映了调查区域的平均现场状况，因此丢失了可变性信息。当调查长度短于 1.0 SOF 时，将产生有利的预测。但是，应该注意的是，使用较短的调查长度可能会限制调查的深度。