Non-Seismic Liquefaction: What it is and how dam owners can reduce their risk

By Carlin Grundemann – Senior Project Engineer

Non-Seismic liquefaction became a hot topic in 2021. That’s when it was cited as the cause of the 2020 failure of Edenville Dam’s earth embankment near Midland, MI. The Edenville Dam released waters of the Tittabawassee River from its own impoundment and caused the failure of the Sanford Dam downstream. The failure displaced residents from their homes, destroyed businesses, and cost the community billions. After the disaster, the Federal Regulatory Energy Commission (FERC) has completed a forensic investigation which resulted in a higher level of scrutiny on hydroelectric projects that included embankments like Edenville’s with a request to look closer at the potential to trigger this type of failure. Let’s take a brief look at how Non-Seismic liquefaction occurs, how you can determine if your dams may be at risk, and what we know about the changing landscape for owners of dams with these kinds of embankments.

First, what exactly is liquefaction? Liquefaction in soils is a phenomenon where saturated, loose, fine-grained soils lose their strength and behave like a liquid under undrained conditions. This can occur when the soil is subjected to an increase in stress, such as during an earthquake (seismic), or by the weight of additional non-seismic loads (i.e., crane loads, steel sheet pile installations, maintenance equipment loads). This phenomenon causes the water pressure within soil voids to increase.

When the water pressure becomes high enough, it can weaken the soil as the soil particles become suspended in the water, and the soil mass starts to flow like a liquid. This sudden loss of strength can lead to catastrophic slope failures, landslides, or the settlement of structures built on such soils.

So, what have we learned about this process since the Edenville failure?

One thing we’ve learned is that liquefaction can only occur when three key components are present:

(1) loose, granular soils

(2) the soils are saturated, and

(3) a load large enough to exceed the peak shear strength is applied to the embankment.

We can easily screen for the presence of the first two components, but the third is a bit trickier to assess. Plenty of loads may lead to this liquefaction response both during construction and under normal operating conditions. But we know they must be sufficiently large enough to push these liquefaction-susceptible soils past their peak strength to be a problem.

How much is too much? The challenge lies in developing a method to evaluate the magnitude of the load as compared to the capacity of the embankment to handle the load. That’s why it’s important to lean on the information on hand. When our teams are involved in these methods, we review past research and data as well as the work we’ve performed in tailings dams and other studies on loose earth embankments. We’ve used this data and experience to compare the applied load to the available capacity of the soils to predict whether a load will trigger liquefaction.

We continue to refine this data and discuss how we can improve. We have successfully applied this evaluation process on multiple dams since 2020, resulting in implementation of specific monitoring and definable load limitations to reduce risk. This experience in determining liquefaction potential has helped better define whether dams are at risk of this type of failure.

This process continues to evolve as the industry works to understand the factors for these types of failures. Ongoing conversations with owners and regulators, along with our work will help to define risk tolerance and parameters for dam safety. Until then, we can put this evaluation process to work. For dam owners with similar structures, you’ll want to work with a consultant who has experience understanding the data and causes of non-seismic liquefaction. The situation is fluid (pun intended), and more discussions will need to take place before any final processes and procedures are in place.

In the meantime, we’re continuing to learn how to leverage the data with each project. I’m at the Clean Currents Conference this week in Cincinnati if you would like to discuss this further. Otherwise, contact me to learn more about how we can apply this process to your project to identify potential triggers and help reduce your risk.