SIMPLIFIED BREACH ANALYSIS MODEL FOR HOMOGENEOUS EMBANKMENTS: PART II, PARAMETER INPUTS AND VARIABLE SCALE MODEL COMPARISONS

Authors: Hanson, Gregory; Temple, Darrel; MORRIS, MARK - HR-WALLINGFORD, UK; HASSAN, MOHAMED - HR-WALLINGFORD, UK; COOK, KEVIN - USDA-NRCS

Submitted to: United States Society on Dams Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 3/1/2005
Publication Date: 6/6/2005
Citation: Hanson, G.J., Temple, D.M., Morris, M., Hassan, M., Cook, K. 2005. Simplified breach analysis model for homogeneous embankments: Part II, Parameter inputs and variable scale model comparisons. Proceedings of 2005 U.S. Society on Dams Annual Meeting and Conference, Salt Lake City, Utah. p. 163-174.

Interpretive Summary: During extreme flooding events, the details of how dam embankments fail and the resulting reservoir outflow are rarely recorded, and yet this type of information is essential to understanding and predicting dam failures and their potential impacts. During the last 5 years, embankment failure tests have been conducted in different countries (US, UK, and Norway), on embankments of various sizes (2 m, 0.6 m, and 6 m), and with different materials and construction practices. The key objectives of these tests have been to observe embankment failure processes, and to collect data on timing failure and the rate of water released as a result of failure. Another objective of these tests has been to provide realistic information to improve and verify dam failure computer models. A SIMplified Breach Analysis computer model (SIMBA) has been developed that is capable of reproducing the key features that have been observed in these failure tests. The model is presently research oriented, but work is underway to incorporate it into a dam overtopping analysis routine for application to field conditions. The objectives of this paper are: (1) to provide practical approaches to determining key material input parameters used in SIMBA; and (2) to compare results of SIMBA breach simulation to the results from these failure tests.

Technical Abstract: During extreme flooding events, embankment failures including breach initiation, formation, and resulting reservoir outflow are rarely recorded and yet this type of information is essential to understanding, and predicting failures and potential impacts. During the last 5 years, overtopping failure tests of homogeneous cohesive embankments have been conducted in different countries (US, UK, and Norway), at varying scales (2 m, 0.6 m, and 6 m), and with different materials and construction practices. The key objectives of these tests have been to observe embankment failure processes, and to collect data on timing of breach initiation, breach formation, process rates (i.e. erosion, headcut migration, and widening), and breach discharge. One of the other underlying objectives of these tests has been to provide realistic information to improve and verify dam breach simulation models. A SIMplified Breach Analysis computer model (SIMBA) has been developed that is capable of reproducing the key features that have been observed in these failure tests. The model is presently research oriented, but work is underway to incorporate it into a dam overtopping analysis routine for application to field conditions. The objectives of this paper are: (1) to provide practical approaches to determining key material input parameters for algorithms used in SIMBA; and (2) to compare results of SIMBA breach simulation to the results from these failure tests.