Dam safety |
Rethinking dam breach analysis
As climate risks rise and infrastructure ages, it’s time to evolve beyond deterministic dam breach models. Eric M. Toth, Priyanka K. Jain, Chris Goodell, and Ben Cary explore why probabilistic analysis offers a smarter, safer path forward for dam safety
Below: Figure 1. Example probabilistic breach discharge hydrographs
700,000 600,000 500,000 400,000 300,000 200,000 100,000 0 0 1 2 3 4 Time (hours) 5 6 7 8
0.2% EP 1% EP 5% EP 10% EP 50% EP 90% EP 95% EP 99% EP
IN THE WORLD OF dam safety, lives can literally depend on the accuracy of a model. For decades, dam breach analyses have leaned on deterministic methods –choosing one “conservative yet realistic” set of inputs to predict breach outflow rates and the associated failure consequences. But as infrastructure ages, climate extremes grow more volatile, and our understanding of risk evolves, a more nuanced approach is needed. It’s time we talk about probabilistic dam breach modelling – not just as a theoretical ideal, but as a practical, necessary evolution of dam safety practice.
The limits of determinism Traditional dam breach analyses rely on fixed input
parameters to describe the breach size, shape, and formation time. While these parameters are chosen using guidelines from agencies like FERC and USACE, they are ultimately estimates that carry a lot of uncertainty. To account for this uncertainty, regulatory conservatism has often skewed these estimates toward worst-case scenarios, but even then, they can be misleading. The deterministic approach gives us a single output – a snapshot that fails to capture the uncertainty inherent in any real-world dam failure. This can lead to over- or underestimating flood impacts, which complicates everything from emergency preparedness to infrastructure investment. And in an era when the stakes include climate- amplified flood events and aging dams built under outdated assumptions, relying on a single version of an uncertain reality is simply not enough.
Embracing probabilistic modeling Probabilistic dam breach modeling reframes the
analysis as a spectrum of possibilities rather than a singular prediction. Using statistical distributions of breach parameters and Monte Carlo simulations, engineers can generate thousands of dam breach outcomes, each with a probability of occurrence. The result? A robust, more informative picture of flood risk that assigns probabilities to a range of outcomes – from moderate breaches to catastrophic failures (Figure 1). At East Bay Municipal Utility District (EBMUD), this method was recently put into action for Dike 2 at Camanche Reservoir in Northern California. Working alongside Kleinschmidt Associates, the team
About the authors
Eric M. Toth, PE and Priyanka K. Jain, PE are civil engineers at East Bay Municipal Utility District in the Water Resources Planning Division. Chris Goodell, PE, BC.WRE and Ben Cary, PE are hydraulic engineers with Kleinschmidt Associates and frequent educators on HEC-RAS modeling. Together, they bring decades of experience in water resources, risk analysis, and dam safety innovation.
Eric M. Toth
Priyanka K. Jain
Chris Goodell
Ben Cary
20 | August 2025 |
www.waterpowermagazine.com
Discharge (cfs)
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45
