Embankment breaching by overtopping : laboratory and field experiments monitored by photogrammetry
(2026)
Files
MasoumehEbrahimiFullThesis_FinalVersion_Dial.pdf
Open Access - Adobe PDF
- 13.89 MB
Details
- Authors
- Supervisors
- Abstract
- (en) Earthen embankments—encompassing dams, dikes, and levees—have played a vital role in human history. The ultimate goal of these structures is to guarantee an integrated fluvial system within watersheds, providing essential services such as flood protection, water storage, and navigation. Climate change impacts the stability and performance of levees, dikes, and earthen dams by simultaneously increasing the magnitude of hydraulic loads acting upon them and degrading their geotechnical resistance. With regards to hydraulic load magnitude, climate change introduces "non-stationarity," patterns where historical weather patterns are no longer reliable predictors of future loads. This results in intensified hydraulic stresses both in terms of frequency and intensity of extreme weather events on earthen structures. Within this context, laboratory and field experimental studies play a crucial role in capturing and documenting the fundamental physical mechanisms of failure under different circumstances and providing high-quality datasets for validation of predictive tools. This thesis developed a methodological experimental framework for monitoring these failure process, quantifying the erosion, and tracking the breach progression into earthen embankments under wave overtopping and overflowing loads. A major innovation of the study is the application of close-range Structure-from-Motion (SfM) photogrammetry to accurately to in-situ experiments conducted on a prototype levee at the Living Lab Hedwige Prosperpolder in frame of the Interreg Polder2C’s and regional Lime in Clay projects. Furthermore, this research enhanced this photogrammetric technique for highly dynamic laboratory environments, specifically during small- and medium-scale sand embankment breaching. To bridge the critical gap between small-scale laboratory tests and full-scale reality, the research features novel medium-scale (1-m-high, conducted at the Hydraulics Research Laboratory of SPW MI at Châtelet, Belgium) and small-scale (0.2-m-high, conducted at the UCLouvain LEMSC Laboratory) experimental modeling of sand embankments. Recognizing the fundamental challenge of satisfying simultaneous hydraulic and geotechnical similitude, this study adopts a 'scale-series' approach while implementing the Froude similarity to embankment dimensions and overtopping load. The medium-scale models eliminate the scale effects induced in small models, especially inherent in models with sediment modeling, and enable the replication of complex processes at a closer scale to the reality. A series of new medium-scale dike breaching tests have been conducted to examine the influence of (i) sediment granulometry and (ii) overtopping conditions (constant reservoir level versus constant reservoir inflow) on the temporal evolution of breach through sand embankments. The results of the present study were combined with well-documented experimental data from the literature for sand embankments ranging from 0.3 m to 1.6 m in height. This extended comparison highlights the profound influence of reservoir dimensions on breaching dynamics, across varying overtopping conditions. To investigate the scale effects, two distinct analytical methodologies were applied: (i) hydraulic and morphologic comparative analysis, where direct comparison of upscaled variables (water levels, hydrographs, breach width) using Froude scaling were performed; and (ii) dimensionless analysis where a verification of the sediment transport regimes using the grain Reynolds number was used to quantify the divergence between model and prototype physics. The relevancy of Settling Velocity Adjustment application to sand embankment breaching was further investigated.
- Affiliations
Citations
Ebrahimi, M. (2026). Embankment breaching by overtopping : laboratory and field experiments monitored by photogrammetry. https://hdl.handle.net/2078.5/276839