Breaching of small embankment dams: tool for cost-effective determination of peak breach outflow

---

Robert M. Boes, Lukas Vonwiller, David Vetsch

Thursday 2 july 2015

16:00 - 16:15h at Oceania (level 0)

Themes: (T) Water engineering, (ST) Computational methods

Parallel session: 13E. Engineering - Computational

---

Increasing public awareness of flood risk requires more rigorous analyses of the impact of dam failure nowadays. While until recently the Swiss dam safety legislation prescribed to take into account a sudden failure assuming a standardized breach geometry for all dam types including earthen embankments, there have been initiatives to consider more physically-based progressive breach mechanisms for embankment dam failures to determine the peak outflow. In a study of some 90 small earthen embankment dams (height up to 10 m, reservoir volumes up to 150’000 m3) in the canton of Zurich, Switzerland, both a 2D numerical model and a parametric model have been used to determine outflow peak discharges due to dam failure by overtopping. The numerical model was calibrated by data from laboratory modelling, while an existing parametric model was then adapted and implemented based on a systematic parametric variation using the numerical model. The parametric model was validated based on field test data from the European IMPACT project. A comparison of breach outflow peaks using the standard breach and the parametric model revealed that in most cases the peak outflow is overestimated by the former due to the assumption of an instantaneous breach. A progressive breach underlying the parametric model appears to be more realistic. However, and more importantly from a risk assessment point of view, the outflow peaks are underestimated by the standard breach approach for the evaluated reservoirs with large volumes, independent of dam height, as well as for relatively large reservoirs impounded by low dams. The parametric model presented herein allows for a quick and efficient calculation of peak breach outflows for small homogeneous embankment dams. Currently, the model is limited to certain parameter ranges as investigated in the current research project, but there are plans to extend its range of applicability to allow for a wide-spread use.