Mount Polley Tailings Pond Breach
Hi and welcome to Failurology; a podcast about engineering failures. I’m your host, Nicole
And I’m Brian. And we’re both from Calgary, AB.
This week in engineering news, Moving Objects with Ultrasound Waves.
Researchers at the University of Minnesota Twin Cities have developed a new method to move objects with ultrasound waves, allowing devices to move without needing a built in power source.
Using the principles of metamaterial physics, scientists developed a way to move larger objects. Whereas before the objects had to be smaller than the wavelength. Metamaterials interact with waves, like light and sound. During the testing, they placed the object on a table and were able to steer it by sound, without touching it.
By placing a specialized tiny pattern on an object's surface, they can push it forward or pull it towards a source like a tractor-beam in the science fiction show Star Trek.
Researchers are planning to test various frequency waves, materials and object sizes.
This technology has a wide range of applications from manufacturing to robotics, by contactlessly actuating things as they move through a process.
Now on to this week’s engineering failure; the Mount Polley Tailing Dam Breach.
I would say this is a more complicated failure than the typical dam failures we’ve covered. It’s also more geotechnical than others because the reports used a lot more dirt words that I had to look up. Sometimes I wonder about other disciplines, but I think it’s clear, dirt is not for me.
Mount Polley is a Canadian gold and copper mine located near Williams Lake, British Columbia. The mine is made up of two open pit sites and an underground component. It’s owned by the Mount Polley Mining Corporation, who is owned by Imperial Metals Corporation.
The mine opened in 1997 and was placed on care and maintenance (in other words, closed) in 2019. It was planned to re-open in 2022 but that doesn’t appear to have happened.
The mined material is processed through a mill where the minerals get floated and concentrated. Everything else, known as tailings, is the leftover waste material. Tailings generally consist of crushed rock, water, trace quantities of metals, and additives used in processing and are often in the form of a wet slurry.
There are three types of conventional tailings dams.
Upstream –where new levels of embankment are built on top of previously deposited tailings, inward from the tailings. This is a dangerous method as the underlying tailings can liquify and undermine new embankments.
Downstream –new embankments are built on top of older ones, outward from the tailings.
Centreline –built on the center with previous embankments.
The types of tailings dams constructed depend on space, safety, and cost. The Mount Polley configuration was known as a modified centerline design with the core inclined slightly in an upstream direction.
Three adjacent embankments confine the Mount Polley tailing pond on the north, east, and south sides, with naturally rising ground on the west. The pond was 4 km2 in surface area shortly before the breach. The embankments were made up of a core, acting as an impervious element, with a filter zone downstream; these were supported by a rockfill zone.
The mine had received approval to raise the crest of the dam by 2.5m. At around 1am on August 4, 2014, a breach occurred in the west end of the northern embankment, while they were still 1m short of the permitted elevation of the crest. The loss of containment was sudden.
The tailings, which consisted of 17 million cubic meters of water and 8 million cubic meters of tailing material, carried felled trees, mud, and debris as it wore away the banks of the Haletine Creek on its way to Quesnel Lake. Widespread water restrictions were implemented right away. Not boil water advisories, as boiling was not sufficient, residents were instructed not to consume the water. Initial tests showed zinc levels above chronic exposure limits for aquatic life.
A Letter of Understanding between the Province and Soda Creek Indian Band and Williams Lake Indian Band outlines a collaborative approach to jointly address all aspects of the tailings breach. The Ministry of Environment led the response for the environmental monitoring, impact assessment, mitigation, and remediation of the area impacted as a result of the Mount Polley Mine tailings breach. This work was undertaken with First Nations, local governments, provincial and federal agencies, and public representatives.
The agreement has five components that will be conducted in accordance with First Nations traditions and scientific knowledge and recognizes the health and safety of the public and workers, including members of the First Nations, are paramount:
A principal's table consisting of the First Nations and the Ministers of Environment, Aboriginal Relations and Reconciliation, and Energy and Mines will oversee a government-to-government response.
A senior officials committee from the three ministries and designates for the First Nations will be responsible for overseeing all of the response activities such as assessing impacts, clean up, remediation planning and decisions related to future requirements to respond to all aspects of the Mount Polley incident.
$200,000 to each First Nation ($400,000 in total) to cover the costs already incurred and future costs related to the tailings pond breach.
The recognition of the important economic contribution of mining to B.C. and the commencement of a dialogue about existing laws, regulations and policies in relation to the mining sector in B.C.
Agreement that the entities responsible pay for all costs and damages incurred in relation to the Mount Polley Mine incident in accordance with applicable legislation.
From the beginning of the tailing pond construction in 1995 until early 2011, one company was the engineer of record, but this changed responsibility in 2011, and was expected to change again after the 2014 construction season.
The embankment's elevations had been raised several times since the early 2000s, totalling about 40m total increase.
The dam embankment was classified as a zoned earth and rockfill dam. The impervious layer composed of glacial deposits
Independent Investigation Panel
Shortly after the breach, an independent review panel was formed to investigate and determine the cause. Human intervention and overtopping were ruled out as causes. Although we will circle back to the role of water in the failure in a bit.
Piping and cracking, where water starts to seep through the embankment layers and erode them from within was heavily reviewed by the panel. We have talked quite a bit about dams on this podcast, both in our regular episodes and our mini failures. Piping and cracking are very common causes of failure of earth dams and the following factors were of concern to the panel.
Modified centreline tailings dams, while within precedent, are disposed to longitudinal cracking
Following a previous construction phase, the core width was reduced to 5 m, which is thin for the planned hydraulic head; again, this has precedent but requires careful filter and transition design and construction
The filter and transition were particularly thin and required meticulous care to be constructed as intended
Details of filter and transition construction in as-built drawings indicated departure from intended design
Much of the as-placed filter material failed to meet applicable filter criteria and requirements for internal stability of its grading .
The core had been overtopped in one location for a brief period in 2014, resulting in softening and enhanced deformability
The core was not contained by the steep rockfill shell in as stiff a manner as might have been possible
A cavity was detected in the core remnant of the left abutment of the breach that was the result of internal erosion
Observed flow to the seepage collection system exhibited a transient spike on April 22, 2013, of the kind sometimes characteristic of internal erosion
Despite all these concerns, the panel didn’t find any evidence that the break was caused by piping or cracking resulting in uncontrolled internal erosion.
Observations from the Surface Investigations provided clear evidence for shearing, bodily lateral displacement, and rotation of the embankment that resulted in the breach. The panel concluded that the August 4th, 2014 failure was caused by shear failure of the dam foundation materials when the loading imposed on them exceeded their capacity. The dam failed rapidly and without precursors.
A shear surface in the remnants of the dam core and by deformation in the weaker glacially deposited layer of silt and clay 8-10 m below the original ground surface, support this failure cause.
One of the native soil layers was localized to the breach area. It went undetected, in part because the subsurface investigations were not tailored to the degree of this complexity. But neither was it ever targeted for investigation because the nature of its strength behavior was not appreciated. The problem with this layer was that as the dam grew higher and the loading increased this layer’s strength behavior would change and it became weaker. This was not accounted for in the design of the dam.
As well, the dam was unprecedentedly steep, and while this was justified by design analysis, it was unreasonable. There was another area of the embankment that had similar characteristics, but it was additionally supported by a buttress and therefore did not fail.
Other contributing factors include:
The dam raising methods were evaluated a year at a time, bordering on ad hoc, and lacked foresight in planning.
Instrumentation was relied upon as a substitute for definitive input parameters and design projections. But for both practical and strategic reasons, the dam was ill-suited to this approach. The steep slopes and constant construction prevented installation of instruments at optimal locations. As well, the instrumentation program was incapable of detecting critical conditions because the strength behaviors were recognized.
While the dam didn’t overtop, water played a part in the failure. High water levels at the time of the breach had some influence on dam stability. As well, the high water required less of a drop in the dam crest before overtopping occurred, which would have possibly allowed some time for intervention to prevent complete loss of the tailings. A lower water level would have also reduced the amount of sediment and water erosion downstream.
The panel found that the regulator and inspections were within reason and did not contribute to the failure. Suggesting that since the flaw in the foundation design was unknown, it was unpredictable.
So there you have it, the Mount Polley Tailing Pond Breach. An undetected soil layer and inexperienced design choices led to a breach of the tailing pond in August 2014 and serious consequences for three licensed professionals.
For photos, sources and an episode summary from this week’s episode head to Failurology.ca. If you’re enjoying what you’re hearing, please rate, review and subscribe to Failurology, so more people can find us. If you want to chat with us, our Twitter handle is @failurology, you can email us firstname.lastname@example.org, you can connect with us on Linked In or you can message us on our Patreon page. Check out the show notes for links to all of these. Thanks, everyone for listening. And tune in to the next episode where we’ll talk about the 2021 Texas Power Outage. Bye everyone, talk soon!