Abstract Soil properties (or parameters herein) vary randomly in space because of different depositional conditions. The knowledge of the spatial variability characteristics of unsaturated soils can be critical to the reliability assessment of geological and geotechnical engineering problems such as the stability analysis of unsaturated soil slopes and the seepage analysis of earth dams. To investigate the spatial variability characteristics of unsaturated soils, the soil parameters, including physical, deformation, and hydraulic properties, are measured through laboratory tests in this study. It is found that the variability of the physical parameters is smaller than that of the deformation parameters, and smaller than that of the hydraulic parameters. Five existing methods are adopted to evaluate the horizontal and vertical auto-correlation distances that measure the spatial variability, and it is found that the space average method and the correlation function method are more suitable for evaluating the auto-correlation distances when the number of test data is small. To further explore the spatial variability characteristics, the Wilcoxon rank analysis is conducted, and the results show that the auto-correlation distance is an inherent property of soils and there is little difference among the results obtained based on the physical, deformation, and hydraulic parameters of the unsaturated soils.

中文翻译：

---

合肥非饱和粘土空间变异特征评价

摘要 由于不同的沉积条件，土壤性质（或这里的参数）在空间中随机变化。非饱和土的空间变异特征的知识对于地质和岩土工程问题的可靠性评估至关重要，例如非饱和土边坡的稳定性分析和土坝的渗流分析。为了研究非饱和土的空间变异特征，本研究通过室内试验测量了土壤参数，包括物理、变形和水力特性。发现物理参数的变异性小于变形参数的变异性，小于水力学参数的变异性。采用现有的5种方法对衡量空间变异性的水平和垂直自相关距离进行评价，发现空间平均法和相关函数法更适合在测试次数较多时评价自相关距离。数据很小。为进一步探究空间变异特征，进行了Wilcoxon秩分析，结果表明自相关距离是土壤的固有属性，基于物理、变形和水力参数得到的结果之间几乎没有差异的非饱和土壤。发现空间平均法和相关函数法在测试数据较少时更适合评估自相关距离。为进一步探究空间变异特征，进行了Wilcoxon秩分析，结果表明自相关距离是土壤的固有属性，基于物理、变形和水力参数得到的结果之间几乎没有差异的非饱和土壤。发现空间平均法和相关函数法在测试数据较少时更适合评估自相关距离。为进一步探究空间变异特征，进行了Wilcoxon秩分析，结果表明自相关距离是土壤的固有属性，基于物理、变形和水力参数得到的结果之间几乎没有差异的非饱和土壤。