Tribal agencies' long-held view that some information used in the environmental study for PolyMet's proposed copper-nickel mine was wrong got new life on Thursday when state officials acknowledged that recent water data could lead to changes in the study.
The Minnesota Department of Natural Resources said the current draft of the environmental impact statement for PolyMet's mine and processing facility estimated the nearby Partridge River's flow using data that differs from other, more recent data. The flow rate is an important part of the modeling process used to estimate how contaminants could escape from the site into the watershed that feeds into Lake Superior.
"We think that this is a data-poor project and that more analysis needs to be done," said Margaret Watkins, water quality specialist for the Grand Portage Band of Lake Superior Chippewa. "We think there will be more significant impacts than the EIS presents."
DNR officials said they were aware of the data discrepancy but said hydrologists were still discussing whether it would require a redo of the document's water modeling. The newest data estimates a base flow rate of 1.3 to 1.8 cubic feet per second, whereas the data used in the EIS estimated flow at 0.5 cubic feet per second.
"It's been suggested that this is a fatal flaw in the whole [supplemental draft EIS]. That's absolutely not true," DNR spokesman Chris Niskanen said. "These are the kinds of discussions that sometimes emerge out of an environmental review process like this that are important to making the final draft EIS a better document, and that's precisely where we are right now."
PolyMet officials also downplayed the development in comments late Thursday.
"We've heard lots of different claims and questions and concerns expressed. To conclude that some actions are going to be required is premature and really inappropriate," PolyMet spokesman Bruce Richardson said.
But the accuracy of the Partridge River's flow rate is something the tribes questioned in 2009, when PolyMet's first draft EIS was released.
"The use of flow data on the Partridge River from a site 20 years and 17 miles distant from the proposed project does not provide sufficient information to allow a adequate assessment of the hydrologic and environmental impacts of the project on the Partridge River," the tribes wrote in their comments to the DNR in February 2010.
Those same concerns appear in a section of the latest version of the EIS labeled "major differences of opinion."
"We think that the modeling for water quality and water quantity impacts from this project is inadequate," said Nancy Schuldt, water projects coordinator for the Fond du Lac Band of Lake Superior Chippewa. "And we believe that they didn't use all the existing data that they could have or should have. They didn't require the company to do adequate gauging and water quality sampling to inform their modeling."
Steve Colvin, who is overseeing the PolyMet environmental review process for the DNR, said the latest flow data for the Partridge River comes from a gauge that is located much closer to the mine site. But that gauge has only taken measurements for about a year. Hydrologists prefer using data that's been averaged over a number of years, he said.
"That's kind of the balancing act," Colvin said. "What [the hydrologists are] going to have to do is weigh the fact that we have actual data but it's temporally limited, and is that better than data from far away that's better overall quality but it's farther away?"
Questions about the water model came up during the first two public hearings on the proposal, and they are likely to come up again at the third and final meeting next week in St. Paul. DNR officials said they will respond to that concern and others before issuing a final EIS.
Colvin said re-running the water models with new data was always a possibility, but he declined to estimate how long it would take to do that. First, he said, the experts will have to determine how significant that discrepancy in the data is.
"It's certainly an important parameter, but some of these baseline parameters are more sensitive to changes than others," Colvin said. "If it's a sensitive parameter, then that's something we'd need to pay more attention to than a parameter that's not likely to change outputs much."