Landowners fear that a dilbit spill would be more devastating than a spill from conventional crude oil, as was the case during the 2010 Kalamazoo River spill. There, the dilbit initially floated on top of the water, but as the light chemicals evaporated, the bitumen sank to the river bottom. Emergency responders had to improvise new cleanup methods, and that work still isn't done. In October, the Environmental Protection Agency asked Enbridge to dredge an additional 100 acres of the riverbed.
Barbara Bekins, a hydrologist with the U.S. Geological Survey and an expert on aquifer contamination, expects dilbit to behave much more like regular crude if it is spilled into an underground aquifer.
Bekins has spent years studying an aquifer near Bemidji, Minn., where 450,000 gallons of light crude oil spilled in 1979. Initial cleanup efforts removed about three-quarters of the oil. In 1983, the U.S.G.S. began using the Bemidji aquifer, which isn't used for drinking water, to study how oil is remediated through natural processes. Almost 80,000 gallons remain in the surrounding soil and aquifer more than 30 years later.
After the Kalamazoo spill, the light chemicals in the dilbit evaporated because the dilbit was exposed to sun and wind. In an underground aquifer, however, sun and wind aren't factors, so Bekins said the thick bitumen would stay close to the point of release, while the lighter chemicals slowly dissolve into the groundwater. In regions where the water table lies deeper underground, the light chemicals would initially remain stuck to the rocks and soil above the water table. But over time, rain would wash them down into the groundwater.
Nebraska's DEQ used a computer model to analyze how dilbit would affect the Ogallala. They chose to focus on benzene, a cancer-causing agent found in all crude oils.
Stansbury, the University of Nebraska civil engineer, said that was a reasonable decision because benzene dissolves readily in water. Other chemicals found in dilbit are unlikely to travel as far as benzene, though he cautioned that it's hard to be sure since some chemicals are considered trade secrets.
The DEQ's computer model showed that in the case of a 42,000-gallon spill, it would take about three years for a plume of benzene to move 1,000 feet. The plume would have a concentration of 5 parts per billion, the upper limit of the EPA's safe drinking water standards for benzene. The plume could continue to spread further than 1,000 feet, they wrote, but by then the benzene concentration would be less than 5 parts per billion.
Dave Miesbach, a Nebraska DEQ water specialist who worked on the report, said some people mistakenly assume that an aquifer acts like an underground river or lake. But water in the Ogallala aquifer moves only about 300 feet a year on average, because it is trapped in tiny empty spaces between rocks, sand and gravel, and is sometimes blocked by impermeable layers of clay.
Bekins said the DEQ's findings are similar to what scientists discovered at Bemidji. There, the benzene plume grew for the first 10 years, then stabilized. The total area impacted never reached more than 660 feet long, 490 feet wide and 33 feet deep.
The water outside that zone is safe to drink, she said, and benzene levels in the contaminated area continue to drop as the benzene is consumed by naturally occurring bacteria that feed on petroleum products.
Stansbury said the science from the Bemidji spill—along with the modeling the DEQ did on the impact of a 42,000-gallon spill—offers general insight into how an oil spill would impact the Ogallala. But he said the study would have been much more valuable if it had assessed the impact of a worst-case scenario spill, one much larger than 42,000 gallons—and if the agency had modeled the effects along specific parts of the route, taking into account the location of drinking wells, the local geology and other factors.
Stansbury's colleague at the University of Nebraska-Lincoln, Wayne Woldt, has tried for years to get funding for such a study. But Stansbury said it has become "harder and harder to get money from truly unbiased sources."
Funding from industry or environmental groups could be seen as tainted by conflicts of interest, he said. And the research needed for the Ogallala is too specific to attract funding from the National Science Foundation, which supports more general, big picture projects. The EPA sometimes supports such projects, he said, but its budget has been tightened in recent years due to budget cuts.
The EPA did not respond to questions about its funding.
Jane Kleeb, the director of Bold Nebraska and a prominent pipeline opponent, said that given the importance of the Ogallala aquifer, no decision should be made about the pipeline until scientists conduct a detailed study like the one proposed by Stansbury and Woldt.