This story was originally published by Montana Free Press at montanafreepress.org.
Despite a decades-long effort to remove heavy metals from the Clark Fork River after a flood more than a century ago washed tons of contaminated mining sediment into the river, a new study suggests the Clark Fork could face a similar disaster if a future flood breaches the berm at the Smurfit-Stone site.
The Smurfit-Stone site, a former pulp mill near Frenchtown downstream of Missoula, is home to hundreds of acres of unlined ponds storing toxic waste, essentially buried in the floodplain of the Clark Fork. For years, an earthen berm has stood between the toxic waste and the river, but recent spring runoff eroded the berm and heightened concerns about its long-term integrity and the risks it poses to downstream communities.
To better understand the growing risks and to urge the Environmental Protection Agency to consider the risks of a berm failure, the Clark Fork Coalition and American Rivers commissioned a study from River Design Group, a local engineering and restoration firm. The study used hydraulic modeling of the site to evaluate various berm failure scenarios under current and future climate conditions.
The unlined ponds at Smurfit-Stone are already leeching dioxins and heavy metals into the groundwater and river. The study found that if the berm breaches, thousands of tons of toxic waste including heavy metals, PCBs, dioxins, and furans could spill into the Clark Fork River and travel downstream, potentially reaching as far as Lake Pend Oreille in Idaho.
“Berm failure is not a catastrophic event—it’s inevitable. It will happen.”
“This study was really aimed at pushing the EPA to really evaluate the risk of berm failure seriously,” said Andrew Gorder, legal director for the Clark Fork Coalition, a Missoula-based nonprofit dedicated to protecting and restoring the Clark Fork River basin. “The risk of berm failure has been brought up by every stakeholder who’s participated in the Superfund process but the EPA has consistently said, ‘We cannot account for catastrophic events such as a berm failure when we look at risk assessments.’”
The EPA is responsible for monitoring and overseeing the cleanup of the Smurfit-Stone site. Smurfit-Stone is not officially designated as a Superfund site, but past owners and operators of the mill and property entered an agreement with the EPA to follow the Superfund process and fund the investigation and cleanup. Fourteen years into the process, the EPA is still in the remedial investigation phase, collecting data to understand the contaminants and their risks to human health and the environment. This phase will continue for a few more years before a feasibility study begins to analyze cleanup options. The actual cleanup is likely at least a decade away.
“Berm failure is not a catastrophic event—it’s inevitable,” Gorder said. “It will happen.”
During a recent forum to present the study in nearby Alberton, Allie Archer, the site’s EPA project manager, said she was impressed by the information in the study and that it would be “taken into account as we look at risk assessments and future risk.”
The study
The study examined three scenarios for berm failure at Smurfit-Stone: piping, breaching and overtopping. Sam Carlson, staff scientist for the Clark Fork Coalition, said that to understand the flood risk at Smurfit, we need to look beyond historical records and consider the possibility of much larger floods.
The first scenario, “piping,” happens when water finds a weak spot in the berm, like a pipe or a hole made by burrowing rodents, that it can push through. According to the study, in a 1-in-100 chance flood under current climate conditions (at approximately 66,000 cubic feet per second—only slightly higher than flows observed during recent spring flooding), the Smurfit site would be vulnerable to piping. Piping could move contaminated sediment and flush contaminants through the groundwater.
The second scenario, “breaching,” would happen if floodwaters moved rapidly enough to pick up and transport toxic soil. In a 1-in-100 chance flood, with water flowing at about 100,000 cfs under projected future climate conditions, about one-third of the site would flood and the water could carry organic matter and contaminated sediment downstream. Breaching could also occur at smaller flood levels.
“Some of the worst contamination is associated with organic matter as a result of the bleaching process of wood pulp,” Carlson said. “That organic matter would definitely be mobilized under a scenario like a breaching.”
The third scenario, overtopping, is the least likely but most catastrophic outcome in which a massive flood would spill over the top of the berm and rush through areas storing toxic waste, spreading contamination far downstream. The study estimates a 1-in-500 chance of overtopping occurring (with flows of about 130,000 cfs) under future climate scenarios. The study shows that once the berm starts to fail, it could take as little as two hours for a large breach to occur, making emergency response nearly impossible.
The EPA monitors the berm’s condition and has an emergency response plan, which Archer said includes being on-site daily when the river is at flood stage and being on-site 24 hours if they see water rising to those levels. But Carlson argued that two hours is “not enough time to get the kind of heavy equipment and materials that would be required to reinforce the berm.”
Chris Nelson, a hydraulic engineer with RDG who worked on the study, noted that there are many different ways the berms can fail.
“They can be associated with different flow levels, and they can happen in different locations,” Nelson said. “But the end result is essentially the same. You have a failure of the berm. You have water enter this contaminated area and then getting back out into the river.”
Potential for a catastrophe
Any number of combinations of the scenarios in the study could occur, and the risk of berm failure increases over time due to ongoing climate change.
“Even these rare events, they are very improbable in any given year, but over the timeframes we’re interested in, those chances become fairly significant,” Carlson said.
The terms “100-year flood” and “500-year flood” refer to the annual chance of such floods occurring, not their frequency over a century. For instance, a “100-year flood” means there is a 1-in-100 chance of that flood occurring each year, translating to a 40 percent chance of happening at least once in a 50-year period.
The Clark Fork River has already experienced floods that surpassed predictions, such as the 1908 flood. Historical photos showing water marks on an old grist mill suggest that a flood of similar size may have occurred in 1887. More recently, the 2022 Yellowstone River flood, which had less than a 1-in-1,000 chance of occurring, underscored the need to consider how climate change might increase the frequency and severity of extreme weather events. The Yellowstone flood was caused by heavy rain on melting snowpack.
In less extreme cases, such as the 2018 Clark Fork River flood, emergency measures were required to shore up the berm after contamination plumes, containing heavy metals like arsenic and manganese, were observed leaching into the river.
“We’ve been pushing the EPA since then to take the risks posed by the berm seriously and incorporate that in the ultimate remedy they come up with for the site,” Gorder said.
“What we now think are our catastrophic floods are going to become the new norm in the future.”
Gorder hopes the EPA will use the study’s data in their risk assessments and scrutinize the berm more closely. He cautioned that simply reinforcing the berm would be a temporary fix, as it was never meant for flood protection.
“This is great information, and I’d be happy to be able to discuss this more as we look at protective remedies down the road,” Archer said during the meeting.
The EPA has committed to preparing a climate vulnerability assessment, but Gorder expressed concerns that it might not thoroughly analyze potential failure scenarios or include the latest climate data, based on his review of other EPA climate assessments nationwide.
Despite the looming threat of a berm failure, Lisa Ronald, the Western Montana associate conservation director for American Rivers, said her organization doesn’t want to shortcut the completion of the necessary sampling to inform the cleanup process.
“We’re concerned about the timeliness of the cleanup, but we also want to make sure that we have the data that we need to get the cleanup that is warranted,” Gorder added.
Ronald also warned against a “waste-in-place” approach, which involves leaving the waste at the site with some containment measures.
“What we now think are our catastrophic floods are going to become the new norm in the future,” Ronald said. “The consequences of that just aren’t worth the gamble at Smurfit. The study presents compelling evidence to suggest that a waste-in-place remedy isn’t going to work here. We can’t look at these berms as a protective structure.”
Ultimately, Gorder said, “We’d like to see the most hazardous, toxic material that we know exists onsite dug up and placed in a safe repository that’s out of contact with the groundwater.”
