In the Netherlands dikes protect land from flooding by the numerous rivers flowing through. The stability of the Dutch dikes and the failure mechanisms that might occur are well understood, thanks to the improved knowledge of soil mechanics in recent decades. However, one process for which we would like to know more is the so-called backward erosion piping (BEP) (Beek, 2015). BEP is an erosion process driven by a hydraulic gradient which occurs at high water levels on the riverside of a dike. When a crack or well is present in the hinterland this hydraulic gradient might cause transport of sand particles and cause pipe formation as shown in Figure 1a.
This is a very complicated process for which the analytical Sellmeijer piping model (Silvis, 1991) was developed at Deltares in the 1990s to capture the secondary erosion of a pipe and its growth in depth. The Sellmeijer model has ever since been used widely in both academic research and practical applications. However, one drawback of this analytical model is that it is not based on the microphysical processes at the particle scale, namely, the behaviour of individual particles and its interaction with the groundwater flow as shown in Figure 1b. A revisit to the Sellmeijer model is urgently needed to verify and possibly extend its capabilities to handle raising water levels.
Assignment
Recent advances in grain-scale modeling and particle-fluid coupling makes such a revisit possible. The idea is to quantify the physics accurately at the grain and pore scale and then obtain upscaled quantities to examine the Sellmeijer model. In this assignment we want you to investigate if the BEP process can be modelled at particle level using a combination of the Discrete Element Method (DEM) and the lattice Boltzmann method (LBM). This will include getting a thorough understanding of the physics of the problem such as the groundwater flow conditions and the mobility and interaction between particles. In addition, a comparison can be made between the Sellmeijer model and the DEM-LBM model for verification in simple test cases.
Learning goals:
- Understanding the physics of the erosion process
- Translating physics to models
- Assessing the quality of the model
Figure 1 On the left a schematic is shown of the BEP process at dike level where the pipe progresses towards the river due to the hydraulic gradient caused by the river. On the right we have zoomed in on the erosion process right at the interface between clay and sand.
References
- Beek, V. v. (2015). Backward erosion piping: initiation and progression. Thesis, Deltares, Delft University of Technology.
- Silvis. (1991). Verificatie Piping Model: Proeven in de Deltagoot - Evaluatierapport. Delft: GeoDelft.
Supervision
Are you interested in this assignment? Contact the Master thesis coordinator:
