Canada’s worst mine disaster grew from a dam built in the wrong place

Maybe you saw the news about the expert review panel’s conclusion that undetected weakness in glacial deposits beneath a tailings dam caused the disastrous breach and waste release at the Mount Polley copper mine in British Columbia last August.

If so, you were perhaps reading the Duluth paper, which carried a brief Reuters story under the headline, “Investigation finds design flaw caused Canada mine dam spill.”  

Amazingly, you still have to squint to find coverage in U.S. media of what is now generally described in the Canadian press as the greatest mining disaster in the country’s extensive history, with cleanup costs projected at more than $200 million.

This is so even in a media market like ours, at a time when the prospect of new copper mines at the edge of the Boundary Waters is at the core of so much consternation, conflict and debate.

When legislators are holding hearings to get updated on those projects, as they did Tuesday afternoon in a cursory, heavily pro-industry “informational hearing” by the House’s new committee on mining and outdoor recreation.

When a key architect of the failed B.C. dam has also had significant involvement with the PolyMet project that’s first in line for a Minnesota mining permit.

An editor at Duluth’s News Tribune helpfully added that context to the Reuters piece published Saturday – and hats off to her or him for doing so:

The company that designed, engineered and oversaw the construction of the Mount Polley tailings dam —  Knight Piésold —  also provided the Minnesota Department of Natural Resources and PolyMet with input on the current proposal for the proposed Poly-Met copper mine project near Hoyt Lakes.

Two Knight Piésold employees are listed in the Supplemental Draft Environmental Impact Statement as contributors. And, in a 2013 corporate overview, PolyMet officials described the relationship as “the State of Minnesota has engaged Environmental Resource Management and Knight Piésold to assist in the completion of the EIS.”

Investigations continue

This is not to say that the Canadian panel blames Knight Piésold for the dam’s construction flaws or ultimate failure. It doesn’t really blame anyone – not the designers, not the operators, not the owners, not the regulators or inspectors.  

The picture it paints is of a design that was flawed from the outset, because of subsurface conditions that weren’t sufficiently investigated or understood, and went undetected over its 20-year lifespan.

Which is not to say there weren’t other, obvious problems with tailings storage at Mount Polley. The ever-rising pond was in and out of trouble for years as a result of unsound management practices, some of them noted by the expert reviewers.

It may be that their full findings will have more to say in the way of fixing responsibility: The version of the report released on Friday was heavily redacted by B.C. officials, perhaps to protect ongoing inquiries, perhaps to protect some backsides. (A Vancouver activist group has counted 113 supporting documents held back.)

And the investigations are far from finished.

On Tuesday afternoon, as spokesmen for PolyMet and other Minnesota mining projects were assuring Minnesota lawmakers that state and federal regulators would guarantee the environmental safety of their operations, Canadian cops were serving new search warrants at the Mount Polley mine and the downtown Vancouver offices of its owner, Imperial Metals.

The document seizure, according to CBC, lasted long into the evening.

A history of problems

I think it is important to remember that while the review panel’s conclusions are both new and significant, the history of trouble in tailings and water storage at the Mount Polley mine is a very old story.

Within a week of the breach last Aug. 4, Canadian journalists had combed through documents and dug out the basic narrative:

Imperial Metals opened the mine in 1997, when copper prices were high, then shut it down four years later when they were low. In 2005, rising prices moved Imperial Metals to put the mine back online; in 2009, it proposed to expand production – and tailings storage – beyond its original permits.

One way to do that was to just keep adding height to the dam, an “earth-rockfill” structure formed by piling up dirt, crushed rock and tailings to specified heights and slopes, under the guidance of engineers. Another was to “dewater” the storage pond by pumping some its contents into natural waterways.

The B.C. government signed off on both solutions; had it declined, the pond would have filled to overflowing.

The papers also documented a sharp decline in mine inspections throughout the province since the Liberals won control of the government from the New Democrats in 2001. And plenty of whistleblowers came forth to tell of warnings that went unheeded.

(Among those sounding alarms may have been Knight Piésold, which resigned as the engineer of record at the Mount Polley operation in 2010; after the collapse, it released a letter in which it warned Imperial Metals that the dam was being raised too high, and in a way that meant “it can no longer be considered a Knight Piésold design.”)

Some 15 million cubic meters of water and solids let loose when the dam breached, and because of past concerns about exceeding the pond’s design capacity, the initial suspicion was that the upper portions of the dam had given way.

But the expert reviewers found no evidence of that, nor any that the dam had gone soft in the middle from “piping” of water laterally through its core – perhaps the most common way these dams are known to fail.

They found that the pond was very near, but still below, its authorized height at the moment of collapse. Because the safety margin had grown slight, its condition near the rim was being watched closely; the report includes a somewhat unnerving table of all the checks made on Aug. 1, 2, 3 and 4 that failed to detect any trouble until just before it all gave way.

A question the experts don’t answer, and perhaps can’t, is whether the dam might not have failed anyway, from some other cause, even if its foundation had been placed on a sufficiently solid footing.

A near-disaster last May

On that point I’ll close by quoting a passage, lightly compressed, from Vancouver Sun columnist Vaughn Palmer, who has been covering the story for a long  time and providing, in my opinion, an especially clear perspective. Zeroing in on a near-failure last May that is documented in the report, he writes:

The spur was a heavy run-off from an abnormally high winter snow pack, followed by torrential rains. With the rising waters in the tailings pond threatening to over-top the dam itself, the mine operator put out a call on the 25th to AMEC, its engineering firm of record.

After arriving on the scene the following day, geotechnical engineer Dmitri Ostritchenko found plenty of cause for concern: wet spots on the embankment, seepage here and there, and a pond almost level with the core of the dam.

Two days later he reported by email to the company’s senior geotechnical engineer, Andrew Witte, that the situation had not much improved. “At the end of the day, the freeboard level is basically zero,” wrote the on-site engineer, referring to the gap between the water level and the crest of the dam.

Despite some effort to reduce the amount of water behind the dam, tailings were still being added to the pond because the mine was continuing to operate.

This was too much for Witte. The safe operating standard was nine-tenths of a metre of freeboard at bare minimum. Mining operations had to take a back seat until that was restored. He directed Ostritchenko to remind the company of its obligations.

“That is a dangerous game to play and we need to make sure that our ass is covered by telling them to pump water out of the tailings storage facility. We cannot support the ‘just keep operating in the danger zone attitude.’ Remember, if they lose the dam, the mine can’t operate anyways.”

They didn’t lose the dam — not then, anyway. That wouldn’t happen for another 10 weeks.

 * * *

The full  report, with supplemental material, can be read or downloaded here.