New Studies Published on ASDSO's Dam Failure Website

As dam failures and incidents occur both nationally and internationally, there is a pressing need to understand the underlying causes of failure to help minimize such occurrences in the future. Current information on historical dam incidents is sometimes scattered, incomplete, missing, or misleading – making it difficult for owners and practitioners to easily access meaningful information that could assist them with critical design and operational decisions. If lessons learned and best practices are not effectively communicated, there is a possibility that poor practices will be repeated, and preventable incidents will not be averted.

Presented within DamFailures.org are links to individual case studies and 'lessons learned' pages that summarize historical dam incidents and failures and the valuable information gleaned from them. Each page contains background and description, photographs, videos, best practices, and other resources related to the case study or lessons learned being addressed. The content of this website encompasses a range of failure modes, dam types, and dam safety topics, including best practices regarding engineering and design practices, human factors, emergency planning and response, operation and maintenance, and regulatory issues.

The site now has over 125 studies, including more than 80 case studies and 45 lessons learned!

Thank you to all the contributors, the ASDSO Dam Failures and Incidents Committee, and the FEMA National Dam Safety Program, which provides research funding. 

2026 Case Studies

St. Francis Dam (California, 1928) - Updated Content

Researcher: Michael Bennett, P.E.
Reviewers: Bill Johnstone, Ph.D., P.E. and Seth Thompson, P.E.

The St. Francis Dam breached due to underseepage in March 1928, just two years after its completion. Designer William Mulholland had enjoyed a brilliant half-century career in water supply engineering in Los Angeles, and his handiwork had helped build the city into a metropolis. However, he never welcomed outside review. Thus, he made design decisions for the new dam himself, and the structure therefore fell far short of the standard of care for dam foundation engineering laid out in mid-1920s technical literature. Mulholland and his team did not perform detailed geological studies or a meticulous subsurface investigation at the site. Moreover, they utilized neither cutoff trenches nor grouting beneath the dam and included only a handful of uplift wells beneath it, despite known concerns about the site’s bedrock. Construction errors such as improperly excavating rock, raising the dam without widening it, and using underweight, unreinforced concrete compounded Mulholland’s design errors, and he improperly interpreted warning signs of the impending failure. The resulting disaster ended his accomplished career and led to stricter licensure laws for civil engineers. Read More
 

Dale Dyke (UK, 1864)

Researcher: Michael Bennett, P.E.
Reviewers: Bill Johnstone, Ph.D., P.E. and Tom MacDougall, P.E.

The Dale Dyke Dam breached in March 1864 due to internal erosion, killing roughly 250 people. The dam was built to impound water for the rapidly growing city of Sheffield and featured a clay core, gravel shoulders, an emergency spillway, and two discharge pipes. Errors in the dam’s design and construction led to leaking, settlement, and – after the dam’s first filling – failure. The breach represented the deadliest civil engineering disaster in Victorian Britain and prompted immediate public outcry into what had caused the failure. Both the government and the dam’s owner hired teams of civil engineers to complete forensic investigators of the disaster. Each side drew several valid conclusions but left critical questions unanswered, and the mystery of the primary causes of the Dale Dyke breach persisted for over a century. Only in the 1970s, after 50 years of development in soil mechanics and the discovery of a key construction drawing, did dam engineers finally trace the failure back to a combination of differential settlement, inadequate zoning of shoulder material, and hydraulic fracturing of the core. Read More
 

Lake Okeechobee (Florida, 1926 & 1928)

Researcher: Seth Thompson, P.E.
Reviewers: Mark Baker, P.E. and John W. France, P.E., BC.GE

The Great Miami Hurricane of 1926 and the Okeechobee Hurricane of 1928 overtopped and breached low agricultural levees around Lake Okeechobee, producing two of the most consequential wind-related dam incidents in United States history. The failures were not caused by coastal storm surge entering the lake; Hurricane-force winds acted over the lake’s long fetch, creating substantial wind setup with damaging waves and runup superimposed on the raised water surface. The 4- to 5-foot sand-and-muck levees had little freeboard and limited erosion resistance. More than 100 people died near Moore Haven in 1926. In 1928, floodwater swept across the low-lying communities and agricultural lands south and southeast of the lake, killing at least 2,500 people, many of them Black and Bahamian migrant farmworkers. Rescue, recovery, cleanup, and burial continued for weeks under difficult conditions. The disasters led to federal flood-control authorization in 1930 and construction of the first major federal Lake Okeechobee levees, which later became part of the 143-mile Herbert Hoover Dike. The incidents show that extreme wind must be treated as a primary dam-loading condition and carried through freeboard, slope-protection, risk, and emergency-planning evaluations. Read More

Mount Polley Tailings Storage Facility (Canada, 2014)

Researcher: Meghan Walter, P.E.
Reviewers: Bill Johnstone, Ph.D., P.E. and John W. France, P.E., BC.GE

In the early hours of August 4, 2014, the Mount Polley tailings storage facility (TSF) in central British Columbia, Canada, failed suddenly and catastrophically, releasing 25 million m3 of water and tailings downstream into Polley Lake, Hazeltine Creek, Quesnel Lake, and the Cariboo River. The released slurry contained heavy metals, causing extensive environmental damage, destroying aquatic habitat, contaminating drinking water, and degrading water quality.

Following the breach of the TSF, an investigation found that the failure was caused by deformation of the northern section of the Perimeter Embankment due to foundation failure, which occurred because of inadequate strength of a continuous layer of glaciolacustrine clays. The glaciolacustrine clays were assumed to be overconsolidated. As a result, designers failed to adopt the appropriate undrained strength interpretation for the glaciolacustrine materials in the foundation and did not recognize that it was susceptible to undrained failure when subjected to increasing loads imposed by the embankments as they grew in elevation. Post-failure investigations found the glaciolacustrine layer to be preconsolidated prior to embankment construction but it became normally consolidated under the loads applied by construction of the Perimeter Embankment. The embankment slope had been constructed to a height of almost 40 m (131 ft) on an unbuttressed slope of 1.3H:1V, making it marginally stable. As the tailings loading increased over time, the weight on the foundation increased until the strength of the glaciolacustrine layer was exceeded and it could not support the load, abruptly and catastrophically collapsing.

Three professional engineers were disciplined in relation to their work at the Mount Polley mine. Two of the three engineers incurred fines of $25,000, were required to pay legal fees, and are no longer permitted to practice engineering in British Columbia. The third individual’s engineering license was suspended for a period of two months, and that individual was required to complete educational courses about tailings storage and management. Read More
 

Mud Pond & Lake Lee Dams (Massachusetts, 1883 & 1965)

Researcher: Lee Wooten, P.E.
Reviewer: Greg Richards, P.E., CFM

The Berkshires of western Massachusetts have been a perilous setting to build dams on at least three occasions.  The Williamsburg Reservoir Dam failure, which killed 139 people in 1874, was the deadliest of these events.  Two other dam failures that occurred on one site in another Berkshire community - East Lee.  At about 5 a.m. on April 20, 1886, the stone dam at Mud Pond failed and released flood waters that swept through East Lee, killing seven residents.  In 1965, Mud Pond was enlarged with another dam, constructed over the same footprint as the previous dam, and renamed Lake Lee.  Eighty-two years after the first dam failure, on March 25, 1968, Lake Lee Dam failed, killing two downstream residents.  The failures had common contributing conditions in addition to their location - poor foundation conditions, steep and narrow floodways, deficient engineering and construction, and little regulatory oversight. Read More
 

DMAD Dam (Utah, 1983)

Researcher: Everett W. Taylor
Reviewer: Seth Thompson, P.E.

The case study details the history and hydrological conditions leading up to the failure of the DMAD Dam spillway during the 1983 flood event. A combination of record snowmelt, full reservoirs, and inaccurate runoff forecasts resulted in floodwaters overwhelming the river system's storage capacity. In June 1983, sustained hydraulic scour at a downstream drop structure caused severe headcut erosion that ultimately undermined the DMAD spillway. The resulting breach released a peak discharge of 30,000 cfs, forced the controlled breach of Gunnison Bend Reservoir, prompted evacuations in Millard County, and inflicted over $6 million in structural and property damage. The disaster highlighted the critical need for modernized infrastructure, coordinated basin-wide water management, and accurate forecasting. Subsequent recovery efforts led to infrastructure reconstruction and legislative reforms. Read More
 

Sella Zerbino (Italy, 1935)

Researcher: Irfan A. Alvi, P.E.
Reviewer: Lee Wooten, P.E.

Supplementing the 154-foot-high (47 m) Bric Zerbino main dam, the Sella Zerbino secondary dam was a 46-foot-high (14 m), 360-foot-long (110 m) concrete gravity saddle dam near Molare, Italy, completed circa 1925. On August 13, 1935, an intense rainstorm produced a flood roughly three to four times the capacity of the spillways and outlet works, all of which were at the main dam. After both dams were overtopped by about 7 feet (2.2 m), the secondary dam and much of the fractured rock saddle beneath it failed, releasing about 16,000 to 20,000 acre-feet (20 to 25 million m³) of water and resulting in at least 111 fatalities downstream. Recent studies indicate that the storm, though severe, was foreseeable, contrary to the “unforeseeable” storm cited to acquit the owner and designers in 1938. In the four decades leading to the failure, a series of human and physical factors interacted and compounded, including grossly inadequate discharge capacity and a secondary dam, which was added late, designed hastily, not designed for uplift or overtopping, and founded on rock which the designers themselves had judged too erodible to serve as a spillway. Read More
 

Gros Ventre Slide and Kelly Flood (Wyoming, 1927)

Researcher: Kevin Ruswick, P.E., CFM
Reviewer: Nathaniel Gee, P.E., Ph.D.

In June 1925, a landslide occurred in northwest Wyoming with a volume estimated of 50 million cubic yards. It may have been triggered by snowmelt, heavy rains, earthquake tremors, or a combination of factors. The slide obstructed the Gross Ventre River creating a natural landslide dam approximately 200 ft high.  Water levels rose quickly behind the slide and reached a maximum level approximately five feet below the crest of the slide forming Slide Lake.  People in the town of Kelly, located approximately 7 miles downstream of the slide, were very concerned about potential failure of the slide dam.  Geologists and engineers from around the country visited the slide dam and concluded it was stable and a permanent feature of the landscape. The pool held steady during the remainder of 1925 and withstood the spring runoff from 1926.  Following a winter with significant snowpack and a warmer than average spring, the landslide dam overtopped and failed on May 18, 1927. The flood killed six people in the town of Kelly. This case study will evaluate the geological conditions leading to the slide formation and human factors leading to the fatal Kelly Flood.  The case study also includes a HEC-RAS dam breach hydraulic model in an effort to simulate hydraulic conditions resulting from the breach of the Gros Ventre Slide. Read More

2026 Lessons Learned

Dams can be targets for sabotage and terrorism.

Researcher: Seth Thompson, P.E.
Reviewer: Greg Richards, P.E.

Dams are typically evaluated for natural hazards such as floods and earthquakes, but they are also critical infrastructure that can be targeted for sabotage, terrorism, and cyber intrusion. Documented attacks worldwide show that aggressors often focus on vulnerable appurtenances such as powerhouses, control buildings, and mechanical/electrical systems, rather than attempting to breach the dam itself. Modern reliance on ICS/SCADA and remote operations further expands risk if cybersecurity practices are weak. Key lessons include integrating deliberate attacks into risk management, limiting sensitive public information, implementing layered physical and cyber security, planning for appurtenance-focused attacks, and strengthening coordination with emergency management and law enforcement. Read More
 

Seemingly minor problems with instrumentation and dam appurtenances can escalate into major incidents and dam failure.

Researcher: Riley Manwaring, P.E.
Reviewers: Nathaniel Gee, P.E., PhD and Eric Halpin, P.E.

Instrumentation, control systems, and other dam appurtenances are sometimes treated as secondary features when compared with the dam embankment, foundation, or spillway, yet past incidents show that these components can be critical to safe dam performance and public safety. Failures involving reservoir level sensors, gate components, outlet works, automated controls, warning systems, and other supporting features have contributed to overtopping, uncontrolled releases, loss of operational control, and, in some cases, dam failure.

These cases demonstrate that problems with seemingly minor or overlooked features often develop over time through deterioration, misalignment, poor calibration, inadequate maintenance, flawed control logic, or gradual divergence between actual field conditions and the assumptions under which systems were originally designed and operated. Therefore, effective dam safety programs must include these systems in design reviews, inspections, testing, monitoring, maintenance, and emergency planning. Recognizing the importance of instrumentation and appurtenant features and treating them as risk-significant components of the overall dam system rather than accessories can reduce the likelihood of major incidents and help prevent dam failure. Read More
 

Coordinating dam Emergency Action Plans with local emergency managers strengthens incident response.

Researcher: Preston Wilson Sr., P.E.
Reviewer: Bill Johnstone, Ph.D., P.E.

Coordinating dam Emergency Action Plans (EAPs) with local emergency managers is critical for effective incident response and community safety. Early collaboration ensures clear roles, communication protocols, and joint exercises, enabling faster decision-making and reducing confusion during emergencies. Local emergency managers bring essential knowledge of vulnerable populations, evacuation routes, and multi-agency networks, which strengthens EAP development and realism.

Key lessons include aligning EAPs with local emergency management objectives, leveraging local expertise, and building relationships before a crisis. Proactive strategies like dam tours, contact verification, and joint exercises foster trust and preparedness. Integrating EAPs with local Emergency Operations Plans (EOPs) ensures seamless coordination.

Effective communication is vital, and using tools like public alert frameworks, Joint Information Systems (JIS), and pre-scripted messages minimizes confusion. Collaboration also improves resource deployment and access to local, state, and federal support during emergencies. Regular EAP exercises help identify gaps in notification, decision-making, and resource coordination, driving continuous improvement.

Ultimately, strong partnerships with local emergency managers enhance preparedness, streamline response operations, and save lives. Read More
 

EAP exercises strengthen dam emergency readiness.

Researcher: Preston Wilson Sr., P.E.
Reviewer: Bill Johnstone, Ph.D., P.E.

Emergency Action Plan (EAP) exercises are vital for transforming written plans into actionable response capabilities. These exercises simulate emergencies, enabling dam owners, operators, and emergency managers to identify vulnerabilities, test procedures, and improve coordination. The Homeland Security Exercise and Evaluation Program (HSEEP) offers a structured framework for designing and evaluating these exercises, which can be discussion-based (e.g., tabletop exercises) or operations-based (e.g., full-scale drills). EAP exercises enhance decision-making, clarify roles, and strengthen partnerships among stakeholders. Dam EAP exercises can address typical dam failure modes such as overtopping, seepage, piping failure, or structural failure due to natural or external hazards. It can also address emerging threats like cybersecurity risks, misinformation, and critical infrastructure interdependencies. Lessons from past events and exercises, such as the 2018 Oroville spillway Incident and the 2020 Michigan dam failures, highlight the importance of clear communication, role alignment, and resource management. Regularly updating EAPs based on exercise outcomes ensures continuous improvement and readiness. Read More