Coupled Solid-Deformation/Fluid-Flow Simulation of Failure Initiation in Variably Saturated Slopes

当土质沿着剪切带破裂后迅速下坡时，就会发生滑坡。泥石流与滑坡的区别在于被动员物质的普遍的、流体状的变形。山体滑坡和泥石流威胁着世界各地的生命和财产。尽管在过去的二十年里，在理解水文驱动的边坡破坏方面取得了良好的进展，但在变饱和土壤中，基于三维物理的流体流动和水文驱动的边坡失稳方面还没有进行重要的研究。该奖项资助跨学科研究，重点是在集水区尺度上对陡峭山坡的水文响应/斜坡稳定过程进行基于物理的表征。该模型将变饱和土壤的固体变形与流体流动过程耦合起来，并量化地表和地下连续体之间的水交换。这使我们能够更好地理解地表径流、蒸散发和渗滤对变饱和斜坡内饱和度、有效应力和变形模式的时空变化的影响。该耦合模型将以库斯湾实验集水区（CB1）的全面和详尽的数据，以及最近使用综合水文模型（InHM）在同一集水区进行的模拟的数值结果进行测试。本研究还将利用最近开发的稳定低阶有限元近似方案，采用等阶插值法对固体位移和孔隙压力场进行插值。高度监测的CB1边坡在1996年11月发生大规模泥石流，因此提供了大量数据，可与模型预测进行比较。研究团队将结合斯坦福大学岩土工程、计算地质力学和定量水文地貌学方面的专业知识，开发和测试基于物理的边坡破坏启动模型。据pi所知，目前文献中还没有一个边坡破坏起裂模型可以定量地解决变饱和度的影响。我们认为，有限元方法已经达到了这样一个高级阶段，它现在不仅可以处理复杂的几何形状，而且可以处理可变饱和度的影响。这项研究的另一个智力上的优点在于用可用的数据集测试和验证所提出的数学方法的巨大机会。CB1数据集允许在复杂地形和可变饱和度的大尺度斜坡上进行模型测试和验证。CB1的高质量水文和岩土数据的可用性将有助于限制问题的参数，从而为更好地理解控制边坡不稳定的重要过程提供巨大的机会。这项研究是一个及时的贡献，有助于提高对控制边坡不稳定过程的理解，在一个由近地表水文和土壤本构特性的严格表征驱动的系统中。模拟工作将有效地证明基于物理的边坡稳定性模型的实用性和/或局限性，这些模型在类似于CB1的现场条件下不如这里开发的模型全面。建议的研究还将利用计算流体动力学的进展来应用于岩土工程和地球科学问题。两个pi都认真致力于确保本科生和代表性不足的学生充分参与这个项目。这可以从他们在斯坦福大学指导、建议、监督和毕业本科生和代表性不足的学生的良好记录中收集到。

Landslides occur when earth material moves rapidly downhill after failing along a shear zone. Debris flows are differentiated from landslides by the pervasive, fluid-like deformation of the mobilized material. Landslides and debris flows threaten lives and property worldwide. Despite the fact that good progress has been made within the last two decades relative to understanding hydrologically-driven slope failure, important research has yet to be conducted in 3D physics-based fluid flow and hydrologically-driven slope instability in variably saturated soils. This award funds interdisciplinary research focused on a physics-based characterization of coupled hydrologic response/slope stability processes for steep hillslopes at the catchment scale. The model will couple solid deformation with fluid flow processes in variably saturated soils, as well as quantify the exchange of water between the subsurface and surface continua. This allows us to better understand the effects of surface runoff, evapotranspiration, and percolation on the spatial and temporal variations of degree of saturation, effective stress, and deformation pattern within the variably saturated slope. The coupled model will be tested with comprehensive and exhaustive data from the Coos Bay experimental catchment (CB1), as well as with the numerical results of recently conducted simulations on the same catchment using an Integrated Hydrology Model (InHM). This research will also utilize a recently developed stabilized low-order finite element approximation scheme employing equal orders of interpolation for the solid displacement and pore pressure fields. The highly instrumented CB1 slope failed as a large debris flow in November 1996, thus providing large volumes of data with which to compare the model predictions.The research team will combine expertise in geotechnical engineering, computational geomechanics, and quantitative hydrogeomorphology available at Stanford University to develop and test a physics-based model of slope failure initiation. To the knowledge of the PIs, no slope failure initiation model currently exists in the literature that addresses the effect of variable saturation in a quantitative way. We believe that the FE method has reached such an advanced stage that it can now handle not only complex geometry but also the effect of variable saturation. A further intellectual merit of this research lies in the tremendous opportunity for testing and validation of the proposed mathematical approaches with the available data set. The CB1 data set allows model testing and validation on a large-scale slope with complex topography and variable saturation. The availability of high-quality hydrological and geotechnical data for CB1 will help constrain the parameters of the problem, thus providing tremendous opportunity to gain a better understanding of the important processes controlling slope instability.The study is a timely contribution towards an improved understanding of the processes that control slope instability in a system driven by a rigorous characterization of the near-surface hydrology and soil constitutive properties. The simulation effort will effectively demonstrate the utility and/or limits of physics-based slope stability models less comprehensive than the one to be developed here for field conditions similar to CB1. The proposed research will also utilize the advances in computational fluid dynamics for application to geotechnical and geosciences problems. Both PIs are seriously committed to ensuring full involvement of undergraduate and underrepresented students in this project. This can be gleaned from their proven track record of mentoring, advising, supervising, and graduating undergraduate and underrepresented students at Stanford.