Influence of soil characteristics on structural soil crust development

Sustainability plays a pivotal role in the successful long-term performance of geotechnical applications. Recently, green infrastructure (GI) programs have been used in the management of surface water runoff in many metropolitan areas. In addition, GI has been shown to provide a local cooling effect, as well as being aesthetically pleasing to neighborhoods. However, challenges remain in sustaining soil properties after post-installation of GI because it requires long-term hydraulic sustainability. Soil properties are subject to change as a result of the mechanical energy imparted by raindrops that impact soil surface. This, in turn, leads to soil crusting over time. The soil crust is defined as a relatively thin and dense layer formed at or near the soil surface. This crustal development influences hydraulic sustainability, and thus the long-term performance of GI. In addition to the problems encountered in this application of GI, the problem has arisen in the fields of agriculture (i.e., seedling emergence) and sediments (i.e., soil erosion). The goal of this work is to provide a mechanical model of soil crust development caused by raindrop impact. Since soil crust formation is attributed to a series of complex interactions between rain and soil, a numerical simulation is thus utilized to take into account the mechanism by which the crust develops as a result of raindrop impact. Understanding the mechanism explains how soil characteristics influence crustal development. This research model simulates the crustal development using a two-dimensional (2D) discrete element model (DEM). The calibration of the numerical model was essential for further study and showed good agreement with the experimental results obtained when micro computerized tomography techniques were used. The simulation results revealed that the amount of kinetic energy transferred from the raindrop to the soil particles was less dominant in the development of soil crust, than was the transfer from the soil to the soil particles. The primary reason for crustal formation was the translocation of fines in the upper soil as a result of raindrop. Using the calibrated model, the effect of soil properties on the crustal development, such as initial porosity, grain size distribution (GSD), and matric suction, was evaluated. The results firstly revealed that the initial porosity had a great influence on the crustal development, regardless of GSDs. Secondly, the results were combined by considering only the coefficient of uniformity (Cu ) and the matric suction level to provide a hypothesis about when or if a material would develop a crust. At zero suction values (dry soils), all Cu 's underwent crust development. However, as the Cu increased, the thickness of the crust decreased. When suction was added, low Cu soils no longer developed crusts, and only the higher Cu 's retained susceptibility. With the transition to higher suctions, only the highest Cu soils continued to develop crusts, but at a thinner level. Finally, with high suctions, no soils underwent crust development, as the high tensions suppressed development. Finally, an attempt was made to predict the change in the seepage property of a numerically simulated profile of pre- and post-raindrop. To suggest a solution to the matter, the Lattice and Boltzmann method (LBM) was utilized. The results revealed that the LBM was a good technique to model the fluid flow in a particulate porous media, but the application of the 2D LBM in this study did not appear to be appropriate for estimating the change in the seepage property. As a preliminary study, 3D LBM simulations were conducted using the micro CT images and showed the effect of soil crust formation with respect to the permeability. Although 3D simulations showed a discrepancy between the numerical and experimental result, it appeared only one order of magnitude difference whereas two orders of magnitude by the 2D simulations.