Ornithologist Dan Cristol stands along the South River, which was contaminated with mercury from the former DuPont plant in Waynesboro, Virginia. Greg Kahn


How Much Should Major Polluters Pay? A DuPont Settlement Provides a Model

A biologist traced mercury from a company spill to contamination in songbirds, and devised a new way to hold polluters financially accountable.

It was just another sweltering summer afternoon gathering blood samples from Shenandoah Valley birds when the news came in. The ornithologist Dan Cristol had been conducting a preliminary assessment funded by DuPont to determine to what degree the company’s pollution of the watershed might have affected the avian community. DuPont was facing potential legal action and had cautiously agreed to one summer of funding for a small team to gauge just how expensive fixing the damages might be. True to his nature, Cristol hadn’t been tentative in his research. He and his students had skulked into stream-bank kingfisher nests, cornered screech owls near bridges, and mist-netted dozens of species of songbirds. Using tiny needles, they’d extracted drops of bird blood before gently releasing their subjects back into the wild. Then they’d shipped their samples to a toxicology lab at Texas A&M, and watched as their funding dribbled away at a rate of $55 per analyzed sample.  

Now as the sun blazed over the South River, a major tributary of the mighty Shenandoah, and waves of heat rose up from the newly mown hayfields, Cristol opened an email from the lab and read the first test results.

“Holy fucking shit,” one of the students cried out.

“I rechecked the numbers about five times to make sure,” Cristol recalls. “We were being funded by the responsible party, so I figured DuPont would look at what we’d found and say, ‘OK, thanks but no thanks, we’ve seen enough.’ I was worried that after this tantalizing glimpse we would not get to learn what was really going on. But to their credit, everyone just kept moving forward and letting us propose to answer each new question that arose.”

Cristol and his students had discovered that the DuPont mercury spill had penetrated much further into the avian food web than anyone had previously expected. Not only was mercury found in fish-eating raptors like Osprey and eagles, but it was present in bluebirds that flitted far away from the contaminated South River; it was in surprisingly high levels in the feathers of the distinctly non-riverine Red-eyed Vireo whose song tells you to look-up way-up tree-top to find it; it was in scrappy Carolina Wrens, whirling Tree Swallows, and reclusive thrushes. Even in the diminutive Blue-gray Gnatcatcher that weighs in at a miniscule third of an ounce.

More importantly, the work had laid the foundation for a novel way to restore North American songbird populations that are declining throughout the country. For Cristol’s research has ultimately perfected a way of holding major polluters accountable for something as profound as it has long been intangible: a means to calculate and seek reparations for bird years lost.

Mercury in Cristol's lab. Greg Kahn

“You couldn’t have a better site to test the effects of methylmercury,” Cristol told me as we stood on a bridge over the South River in the City of Waynesboro and stared southeast at a 177-acre chemical plant. For roughly 50 years this gray smudge of a facility, built into the green hillsides by DuPont in 1928, manufactured something called acetate fibers—which used mercury as a catalyst for the first 20 years of its operation.

DuPont to its credit has never strongly contested that it put significant amounts of mercury into the South River. Ever since mercury was detected in river sediment and floodplain soil around the plant in the 1970s, the corporation has been trying to figure out a way to put its pollution legacy behind it. For years DuPont funded a vaguely missioned “South River Science Team” where state officials and academics monitored mercury levels in fish, with the hope that concentrations would eventually go down. No such decrease was observed. Sampling continued to show levels in some fish higher than 4 parts per million—nearly four times that found in swordfish, which the Food and Drug Administration urges consumers to avoid because of high mercury levels. Nevertheless, officials representing the State of Virginia (technically the key plaintiff in these early proceedings) seemed at a loss as to what to do next.

“They had been completely bamboozled,” says Nancy Marks, a senior attorney for the nonprofit Natural Resources Defense Council’s (NRDC's) litigation team. Under Marks’s direction NRDC filed an intention to sue DuPont in the early 2000s to move the issue forward from monitoring to mitigation. “[DuPont’s] remedy was to have a hundred year monitoring program. But we knew the mercury in the river was sky high. And DuPont was the only obvious source.” Unlike other American watersheds that have been host to numerous polluters, the Waynesboro DuPont plant is all the mercury-discharging industry the South River basin ever had. The bulk of the mercury in the ecosystem and any harm it may have caused birds is undeniably DuPont’s fault.

“For us it was a no-brainer,” Marks recalls. “It was a very strong case.” The case was made even stronger by the fact that NRDC had just won a big legal victory in a suit in Maine under the same legal theory. The Maine suit had gone to trial, and the mercury polluter, a company called Mallinckrodt, Inc., if ordered by the court to pay for remediation, will likely owe hundreds of millions of dollars (this in addition to many millions of dollars spent on legal fees). While DuPont declined to comment about this aspect of the case, it seems likely that the prospect of a Maine-sized case going to trial gave corporate officers pause. Soon after the NRDC action, the company entered into a consent decree with the river’s trustees, the State of Virginia, and the U.S. Fish and Wildlife Service (USFWS).

But by trying to reach a settlement over the South River, DuPont started unraveling another knot when it agreed to finance the research that would determine the value of its liability. That’s where Dan Cristol entered the picture. If the company was really ready to pay to restore the birds of the South River drainage, how would that cost be calculated? It was obvious that immediate and quantifiable harm had been done to, say, sport fish and the people who had lost the opportunity to catch and eat them. But when the bird blood results revealed methylmercury present throughout the avian community, even in songbirds that lived far away from the river, Cristol realized they had a chance to build a much more expansive case.  

First they had to figure out how birds that had nothing to do with the river were getting so much mercury into their systems.


“There’s nothing like getting paidto get up early and catch birds on a beautiful spring morning,” Cristol told me as we cruised down Route 340 to check in at 10 of the 50 sampling sites he established during the seven years of his mercury field study. “But by eleven o’clock at night when you’re going out to catch screech owls with your students, hoping local drug dealers aren’t hanging out under the bridge, you’re saying, ‘This is beyond tedious. This is horrible.’” The Salvadoran street gang MS-13 is active in the Shenandoah Valley and several murders have occurred in the area, including at least one along the river’s banks.

Nevertheless Cristol set about designing and implementing a largely undergraduate-staffed research regimen up and down the length of the South River. “We knew the mercury must be coming through what the birds ate. And the only way we could figure out exactly what they were eating was to catch them in the act.” Realizing it was more or less impossible to reliably catch adult birds eating, Cristol found he could measure diet through the next best thing: what adults fed their babies.

Using a method perfected by ecologists in the early 1990s, Cristol and his students sneaked into nest boxes while adult birds were hunting and put tiny plastic zip ties or “ligatures” around the nestlings’ necks. “We had to be really careful,” Cristol recalls. “Too tight and the babies would suffocate. Too loose and the food goes down and you lose your sample. Just right and you get the perfect bug to measure.” It’s a testament to Cristol’s care and sensitivity that not a single bird was harmed by ligature during the years of field sampling. This task was performed hundreds of times in multiple locations. There were all sorts of things in those bird craws. Grasshoppers and grubs. Gnats and houseflies. But there was one unexpected prey item that was causing most of the  problems.

Dan Cristol, a biology professor at the College of William and Mary in Williamsburg, Virginia, spearheaded the research that led to the discovery of how mercury pollution affected songbirds in the Shenandoah Valley. Greg Kahn

“Thirty percent of their diet was big spiders,” Cristol says. “And those spiders were bringing in 70 percent of their mercury.” Spiders are alpha predators of the insect world. They eat big bugs that have in turn eaten smaller bugs, which had rooted in the mercury-laden river sediment. Just as swordfish and sharks end up as storehouses for all the mercury their prey contains, as well as all the mercury all the prey of their prey contained, so too do spiders end up “biomagnifying” mercury in the environment and concentrating it in their flesh. And they were then dumping this immense toxic load onto songbirds.

It was a revolutionary discovery. “It was just a one pager in Science, but it was a game changer. [Dan] moved the needle on [understanding] how mercury behaves in the environment,” Fish and Wildlife’s John Schmerfeld told me. Henceforth, when mercury contamination cases are considered (and there are dozens pending) litigators will be more likely to look beyond the immediate contamination site and consider the wildlife populations in the wider surrounding environment. It’s a difference that could change the nature of mercury settlements in many cases around the country.

Now that they knew how waterborne mercury was making its way into terrestrial bird blood, the next step was to establish just how many birds mercury had harmed and how badly. This number—the dollar amount needed to restore bird numbers to their pre-DuPont levels—would inform the price tag of the settlement. Again, a disciplined, time-consuming regime was part of the solution, as was some pretty heated negotiation for access to private land. “People we asked for access came in two flavors,” Cristol remembers. “One worried we might be jackbooted government thugs coming to take away their property rights. The other thought we were working for DuPont and trying to poison them.” They needed permission to put up row after row of birdhouses along the polluted river, as well as in nearby uncontaminated tributaries, in order to establish reference populations of Tree Swallows.

By comparing swallows in mercury-contaminated stretches and unaffected areas, Cristol and his students were able to show that reproduction was indeed being affected. Overall they found a 20-percent decline in offspring in high mercury areas. In other more sensitive species, like Carolina Wrens, they observed that as mercury levels increased to three parts per million the birds were more likely to abandon their nests altogether.

The research team was also able to understand how wide a swath methylmercury had cut through the South River drainage. “With river contamination, length is easy,” Cristol would tell me. “Width is more difficult.” That mercury was in the river sediment was obvious. But because they had proof of terrestrial insects with high mercury levels they could begin to fan out into the floodplain to look for effect. Mercury becomes methylated (and thus can permeate cell membranes) by interaction with anaerobic bacteria in areas that are frequently moistened. The flat fields that flooded regularly on either side of the South River turned out to be methylation factories. In many cases mercury concentrations in the birds and bugs were found to be worse many miles downstream than right next door to the polluting facility. In the end with his nest boxes and insect collections Cristol was able to determine that mercury had affected more than 11,000 acres, a much wider swath than had previously been thought. Two more years and 900 bird surveys later, he knew the densities of every species of songbird in the Shenandoah Valley.

With that kind of data in hand, parties to the DuPont settlement could then plug numbers—including bird density and range, contamination levels, and reproductive success—into a model developed in the early 2000s. The original model evaluated the damage a barge company had inflicted on a coral reef in Florida. But the model’s basic math can be applied to a range of different damaging agents and ecosystems. “Once you have an idea of what the injury is you can put that into the model and then on the back end, eventually the theoretical equivalent in money is coughed out,” says Schmerfeld.

But there was one more piece of the puzzle that had to be solved in order for the data to stand up in court. Correlation is not causation as is so frequently said in scientific circles. To establish causation, a different kind of experiment had to be initiated. An experiment that, for a passionate bird lover like Cristol, would prove to be the most painful of  all.


The epidemiology around methyl mercury is still evolving, but the pollutant, at the concentrations found in the contaminated stretches of the South River, has the diabolical tendency to profoundly interfere with life’s processes rather than kill outright. It is formed when anaerobic bacteria bind a “methyl group” of carbon and hydrogen atoms to a mercury atom. The methylation process transforms relatively inert inorganic mercury into something that is more easily assimilated and can even pass the “blood barrier” to the brain. Methylmercury’s tendency to bind to sulfur-containing proteins that are central to the nervous and metabolic systems can cause multiple malfunctions. Birds can lose efficiency in capturing prey as well as something behavioral scientists called “nesting tenacity.” It can alter immune response, spur autoimmune disorder, change expression of reproductive hormones, and limit an animal’s ability to respond to stress.

But while the health effects of mercury poisoning on vertebrates are clear it is extremely difficult to pin the loss of birds on a single mercury pollution event. Mercury was probably causing nesting failures in Dan Cristol’s field research subjects but other confounding factors such as an uptick in invasive predators or some other unknown pollutant could also have played a role. To strengthen the case for actual causation Cristol had to conduct a laboratory-based phase where captive animals were intentionally subjected to the isolated and punishing effects of a high mercury diet. It was clear that Cristol had found this aspect of his research the most troubling. “I could not ethically justify it,” he told me as he opened the door to a repurposed cattle barn on the campus of William and Mary, “if I didn’t think that their imprisonment in cages was going to save a lot of birds in the  wild.”

A juvenile wolf spider. These arachnids accounted for 70 percent of the mercury found in birds in the Shenandoah Valley. Greg Kahn

Inside the lab hundreds of Zebra Finches peeped and fluttered and pecked away at piles of rainbow-colored birdseed that the lab techs have taken to calling “fruity pebbles.” A black or orange mark on a birdcage label indicated whether the subjects inside were controls eating clean fruity pebbles, or experimentals given food dosed with methylmercury to the same concentration as a swordfish steak. Cristol and his students pursued this line of research for six years with an equal amount of rigor as their field studies. They repeated the nesting trials they’d done along the South River. They tested the birds’ memories by hiding food in one of 10 feeders and then examining the birds’ efficiency at re-finding the food an hour later. They probed the finches’ stress-regulation abilities as indicated by the levels of the hormone corticosterone. They even examined their songs and compared them with the lower-pitched song distortions they’d observed in the field. In truth they tested so many different vectors, including heredity and song learning across multiple generations, that it’s beyond the scope of this article to list them all.

The meticulous approach comes back to Cristol’s central passion: to preserve the lives of birds. “This is the weight of evidence,” he told me, closin