Hamilton, the Michigan State professor, said the EPA committee is conducting a number of research projects, including examining the submerged oil's effects on aquatic life, refining a hydrodynamic model to track the oil's movement and analyzing the environmental impacts of dredging and the effectiveness of less intrusive methods of oil removal. He spoke with InsideClimate News as an individual scientist, not as a representative of the EPA.
"My personal interest is to be more ready for the next spill," said Hamilton, who has spent years studying the Kalamazoo's hydrology and water quality.
U.S. imports of dilbit are projected to quadruple in the next decade, and dilbit would be carried on the proposed Keystone XL pipeline, which crosses the critically important Ogallala aquifer. Earlier this month, the Nebraska Department of Environmental Quality concluded that a dilbit spill in the Ogallala would be less serious than a spill into surface water, because groundwater moves slowly within the aquifer. But the agency did not model the impact of a dilbit spill on Nebraska's rivers or lakes.
Spill Forced EPA to Improvise Cleanup Techniques
The debate over the Kalamazoo cleanup underscores how little is known about dilbit and how to remove it from waterways.
To gauge the extent of damage after the spill, the EPA developed a method it calls "poling" to map the amount and location of the oil. Using hand-held poles, workers agitated the sediment in the riverbed to see if oil floated to the surface. They found that the current was sweeping clumps of oil downstream, allowing them to collect in low-lying areas, where they were sometimes buried in up to six inches of sediment.
Results from the latest survey show that oil is pooling near Ceresco Dam, Mill Ponds and Morrow Lake Delta. Oil is also showing up in places that were once oil-free, and the EPA is concerned it could move further downstream during floods.
Enbridge disputes these conclusions.
"Poling is a rough, subjective method to determine the general location of submerged oil without accounting for volume, source of oil or potential for migration," the company said in its letter.
Hamilton acknowledged that the technique is flawed, but says it's the best method they have given the unique challenges of this spill. "Poling remains to this day the only practical method of going out in the field and trying to find submerged oil," he said.
Enbridge also argues that the dredging would do more harm than good.
Last year an InsideClimate News investigation of the spill showed that regulators and scientists have constantly struggled to balance oil cleanup with protecting the ecosystem. The EPA's proposed order acknowledges that dredging is too destructive for some parts of the river. But it also says that after consulting its scientific committee and other experts, it determined that the benefits outweigh the potential damage in the three areas tagged for cleanup.
"The decision has been made, by EPA at least, that we should go after [the oil] while we can," Hamilton said. "Once it gets spread all around it's harder to [clean up]."
Enbridge has used less intrusive methods in many parts of the river, including agitating sediment and collecting any oil that floats to the surface. In its letter objecting to the EPA's plan, it said it prefers to continue with those techniques.
The EPA is studying whether these methods are effective, Hamilton said. But so far, no one has "properly measured what fraction of the oil is collected that way."
Jay Wesley, a fisheries expert with Michigan's Department of Natural Resources, said the three areas targeted by the EPA already have been heavily disturbed by cleanup activities over the past two years, "so some additional dredging…probably won't affect the ecosystem too badly."
Under orders from the EPA, Enbridge used an even more intrusive method in 2011 to clean up Talmadge Creek, a Kalamazoo tributary that received the brunt of the damage from the ruptured pipeline. The creek was so badly contaminated that Enbridge had to essentially rebuild two miles of it.
"They dug out the whole stream and its valley, carted it [away] in trucks and brought in clean materials…It's 100 percent new," Hamilton said.
The dredging the EPA is proposing now would be much less intrusive than that, but the process could still take months and substantially increase the cost of the cleanup, which already totals more than $809 million.
Before any work can begin, Enbridge will have to apply for permits from Michigan's Department of Environmental Quality.
Michelle DeLong, who leads a response team the MDEQ formed to deal with the Kalamazoo spill, said the permitting process would take at least a month.
If Enbridge ends up dredging 100 acres, it would be "a pretty large scale operation," she said. A staging ground would be needed for the heavy equipment, and Enbridge would have to submit a detailed work plan to the MDEQ. Impacted property owners would have an opportunity to comment on the plan and could request a public hearing. If that happened, DeLong said the permitting process could extend to two months.