Imagine a big hole in the ground, similar to the one pictured above. Now imagine that pit filled with up to 10 billion tons of tailings — waste that’s derived from mining — that will remain there forever.

It’s a reality that isn’t exactly far-fetched. If a large — and controversial — copper mine is built in Alaska’s Bristol Bay, what remains will become a lasting part of the landscape. These potential dams could be among the biggest in the world, but they’re by no means the first. So what, exactly, are these man-made storage structures you’ve probably never even heard of?

The Basics

Currently, there are about 3,500 tailings dams worldwide.

The contents of the tailings themselves depend on what mineral is being mined, though the process of creating them is generally similar. According to Dave Chambers, an engineer and geophysicist Center for Science in Public Participation, more than half of the material removed from mines is cast away immediately.

The ore-containing material is then processed, “which means grinding it up to a very small particle size, and then using some chemical additives to remove the material you want,” Chambers told FRONTLINE. “That material is usually less than 1 percent of the stuff you’re usually putting through the processing. So you usually end up with a lot of waste.”

The waste is often in the form of a slurry: a fairly even combination of solid waste and water.  A slurry’s specific contents vary based on the type of mining being done and where the mine is located. For example, the slurry from the proposed Pebble Mine in Bristol Bay would likely contain sulfide minerals. 

“The minerals that we’re trying to mine are what are called sulfide minerals,” Chambers explained in our film Alaska Gold. “The most common sulfide mineral is iron sulfide, pyrite. And what happens when you expose pyrite to oxygen and to water is that it breaks down chemically into a weak sulfuric acid. That acid, in turn, will dissolve some of these other accompanying sulfide minerals that contain lead, zinc, cadmium, selenium, arsenic, and a whole host of other things” that can be dangerous for the environment.

There is no ideal tailings dam design because each environment has different needs and obstacles. But there’s some consensus that the type of design known as “upstream” is both the cheapest to build and the most susceptible to failure. This is because, says Chambers, “you can actually use … tailings for physical support,” reducing the cost of building materials. But because the tailings are mixed with water, they can be unstable structurally.

The safest type of dam build, Chambers told FRONTLINE, is what’s known as the “centerline” method. It doesn’t rely on the slurry itself for support: “You actually have to engineer it, so there’s size specification for the material, it has to be compacted, that sort of thing.”

For more, including diagrams, on these methods and others, see this report [PDF].

How Often Do They Fail?

An extensive 2001 study [PDF] by the International Commission of Large Dams and the United Nations Environmental Programme found that, on average, one major tailings dam incident occurs each year, although that figure doubled between 1995 to 2001.

Most of the failures were a result of issues with water balance, construction and a general lack of understanding about how the dams work, according to the study. There were also failures caused by “unpredictable” climate and geological events, such as floods and earthquakes, though the report makes clear that “it can be argued that with today’s knowledge, allowance should have been made for these events.”

Heavy rain can also pose problems. “The most common failure mechanism is related to hydrologic events — that is large storms that basically overwhelm the storm retaining capacities that the dam was designed for,” Chambers told FRONTLINE.

A 2002 paper [PDF] by Michael P. Davies found an increase in failures in the early 2000s matched only by statistics from the early to mid-1930s. “There’s quite a long list of recent tailings dam failures of dams that were built with modern technology,” said Ron Cohen, a professor at the Colorado School of Mines, told us. Here are three examples:

• Stava, Italy: In 1985, tailings dams built to store waste from flourite mining failed due to poor design and extreme water pressure. Two hundred thousand cubic meters (186,462 cubic miles) of tailings flowed more than 4 km (almost 2 miles) downstream at a speed of up to 90 km (almost 56 miles) per hour. The incident killed 269 people and destroyed more than 60 buildings. Take a look at these aerial images of the area before and after the disaster.

• Los Frailes, Spain: Poor design led to this 1998 failure, in which a 50-meter (164-mile) section of the dam’s wall collapsed, sending acidic water containing sulfur, zinc, copper, iron and lead into the Rio Agrio and adjacent farmland. A 1998 study on the incident [PDF] estimated that about 5,000 jobs were lost due to the failure.

• Kolontár, Hungary: In 2010, a tailings dam storing waste from bauxite mining collapsed due to heavy rain. The red toxic sludge from the dam spread over eight square km (more than three square miles), flooding nearby towns. Ten people were killed, and about 120 were injured. The CEO of Magyar Alumínium ZRt, the company in charge of the dam, was briefly arrested and released after the incident. Not soon after, the government took over the company.

But John Shively, CEO of the Pebble Limited Partnership, noted that “in 2010, the third largest earthquake in history rocked the nation of Chile, and the tailings pond at one of the world’s largest copper mines in that country did not fail.”

The WISE Uranium Project has a list of major failures from 1960-2011.

Who Oversees Tailings Dams?

It depends. In the U.S., state regulators have that job, Chambers told FRONTLINE. “I suspect that, worldwide, there are more national oversight bodies, because it’s probably less regulation on a provincial level,” he said.

In a review of the literature on dams [PDF], Chambers notes a finding that “39 percent of the tailings dam failures worldwide occur in the United States, significantly more than any country.” Why? It could have something to do with which countries actually gather information on incidents. The U.S. has “historically reported tailings dam failures more consistently than, say, tailings dams failures in China,” Chambers notes. “But then you look at Europe, and they’re going to be reporting failures as often as we are, too.”

But Knight-Piesold, a science and engineering consulting company that contributed to a series of white papers [PDF] in response to the Environmental Protection Agency’s critical review of the Alaska’s Pebble Mine project [PDF], stresses that “tailings dams for large-scale mining operations are major structures that that have been constructed, operated and closed successfully in many parts of the world.” They specifically call out the centerline structure, as well as one known as “downstream,” as ones that are reliably safe. 

Pebble has yet to submit a final plan for the mine, though the partnership told Alaska’s KTUU that the tailings facilities “will be built to 21st century standards and that the risk of failure, in any given year, is only around 1 in a million.”

What’s the Future?

Tailings dams will continue to be constructed because of the demand for products that contain minerals like copper. They are the only man-made structure of this size that will remain in perpetuity, even after a mine closes. “We can actually stop using even a water-supply reservoir,” Chambers told FRONTLINE. “Or you can drain the water behind it and it poses no risk.”

Knight-Piesold writes that this time-frame is manageable: “Tailings dams can be built to stand indefinitely provided the right procedures, protocol checks and monitoring are in place throughout all phases of a dam life, including design, construction, operation, closure and post closure.”

“We, as a society, have been building these structures for, say, the last century at most,” Dave Chambers points out. “So we really don’t have any long-term experience with them,” he says. “Are we building them properly? The dam failures would suggest we’re not, from an engineering standpoint.”

“We just don’t have answers for a lot of these things.”