On a beautiful June morning in 1976, workers near the newly constructed Teton Dam in southern Idaho noticed a small leak in the 405-foot-high dam as the reservoir behind it was being filled. By noon that day, the dam had failed completely, emptying 251,000 acre-feet of water onto downstream communities, killing 14 people, and resulting in large economic losses.

How did the dam fail, in spite of its state-of-the-art construction technology? Was it a flawed design, bad material, or faulty construction? Panels appointed by the Bureau of Reclamation and the state of Idaho explored all those possibilities, but a conclusive explanation eluded them.

Now a new explanation has resulted from collaboration between Balasingam Muhunthan, associate professor of civil and environmental engineering at Washington State University, and V.S. Pillai, a consulting geotechnical engineer who has been heavily involved for 33 years in the design and construction of large earth-fill dams throughout North America. Pillai offered to share his expertise in soil liquefaction and the theory of “critical-state soil mechanics,” which could lead to an explanation of the dam’s failure. Soon after, Muhunthan left for a sabbatical at Cambridge University, where A.N. Schofield and others had developed the theory many years ago.

Schofield himself had investigated the failure of Teton Dam and disagreed with the official conclusions of the investigative panels. Muhunthan conferred with Schofield on failures of three dams, including Teton, and they compiled a paper on “Liquefaction-Failures of Dams.”

Reinforced by the Muhunthan-Schofield paper, Pillai re-examined the Teton failure, checking the mechanical/liquidity properties of the low-plasticity silty soils that formed the impervious core, the varying valley geometry of the dam site, and the influence of stresses on the materials of the impervious core that could cause internal cracking. Pillai came up with possible vertical internal cracks in the upper portion of the dam, with the deepest one, “a 32-ft deep crack,” extending from the top of the dam in the right abutment, near where the breach was triggered. Pillai proposed a full-fledged research project to Muhunthan. In early 2002, the National Science Foundation awarded a grant to Washington State University to support the research.

Nearly a half-ton of the core material from the dam was transported to the Soils Laboratory at WSU, and Muhunthan’s research group began an extensive analytical program, based on their new theory, “state-based soil mechanics,” which, says Muhunthan, asserts that “highly compacted soils of low plasticity tend to crack in an environment of low liquidity index, low confining stresses and high shear stresses.”

The research confirmed Pillai’s initial findings, that the dam had internal open cracks at two locations. One set of cracks extended to a depth of 32 feet from the top of the dam, where the actual breach was initiated. The other one was much shallower in the left abutment. When the water level of the reservoir rose to the bottom of the deepest crack in the early hours of June 5, water simply flowed through the open vertical crack(s), which eroded the crack into a large tunnel, leading to the breach of the dam hours later.