What is killing the coho?
Researchers are trying to determine which chemicals in stormwater are contributing to the deaths of large numbers of coho salmon in Puget Sound. It has prompted a larger question: What exactly is in stormwater anyway?
Jenifer McIntyre holds out a large Ziploc bag full of a mysterious black substance. “What do you think is inside?” she asks.
I take the bag and heft it, squeeze it. The material is fine and feels dry, lightweight, oddly synthetic. I hazard a guess. “Is it some kind of dust or dirt or something?”
McIntyre shakes her head. “Nope!”
I squeeze the bag again. “I don’t know, compost? Coffee grounds, maybe?”
“Nope.” McIntyre grins. “It’s ground-up car tires.” She takes the bag back and gives it an affectionate pat. “I’ve been waiting awhile to get this, so it’s a good day.”
McIntyre, an aquatic ecotoxicologist with Washington State University (WSU), is part of a broad coalition of scientists at groups including WSU, NOAA Fisheries, the U.S. Fish and Wildlife Service and the University of Washington working together to solve a longstanding mystery. The ground-up tires are for use in an upcoming study to see what chemicals might leach out of them during a rainstorm, as rain turns into urban stormwater. The results might help clarify what is killing large numbers of coho salmon in the waters of Puget Sound, a condition known as "pre-spawn mortality."
Stormwater may be Puget Sound’s most well-known pollutant, and at the same time its least known. While the state has called stormwater Puget Sound’s largest source of toxic contaminants, scientists are still having a tough time answering two basic questions about it: What is stormwater, exactly, and what does it do?
Let it rain
Every year, the Puget Sound region receives up to forty inches of precipitation, most of it as rain. In the past, which is to say before the I-5 corridor became the bustling urban matrix it is today, much of that rain seeped into the soil or collected on leaves and grass and then evaporated back into the atmosphere; less than one percent was thereafter left to trickle into the Sound as surface runoff. But as humans altered the drainage basin of Puget Sound, so, too, did we alter the fate of the rains. Now, with more than 350,000 acres of impervious surfaces—streets, roads, highways, parking lots, building roofs, and so on—between twenty and thirty percent of precipitation turns into surface runoff. This translates into more than 370 billion gallons of stormwater per year pouring into Puget Sound.
As modern stormwater sluices downhill, it gathers whatever is in its path. By the time it becomes soundwater, it is a formidable toxic stew. According to a 2015 report from the Washington Department of Ecology, at least 33 pollutants have a 50% or greater detection frequency in stormwater, meaning that they are found in at least half of samples. The list includes almost everything from fecal coliform to polycyclic aromatic hydrocarbons, or PAHs, which are known carcinogens. On top of those pollutants, 16 others are found in at least 20% of samples, and hundreds of other chemicals also are present.
All of these pollutants and toxins can have profoundly negative effects on Puget Sound’s biota, such as aquatic insects and, especially, salmon runs, several of which are federally listed as threatened. Nathaniel Scholz, a biologist with NOAA, and colleagues from several government agencies showed in 2011 that between 60 to 100 percent of coho salmon returning to some lowland urban streams in Puget Sound die before spawning. More recent work found that juvenile and adult coho salmon die within hours of exposure to untreated runoff from the 520 bridge between Seattle and the eastside of Lake Washington. And in a new paper in Ecological Applications, biologists from NOAA and the Washington Department of Fish and Wildlife found that across 40 percent of coho’s Puget Sound range, returning spawners are being especially hard-hit in urban areas, primarily due to stormwater.
A complex mixture
So which of the potentially thousands of chemical compounds found in stormwater might be killing the coho? That question is behind McIntyre’s research today at the Washington Stormwater Center in Puyallup, and it is why she has waited so eagerly for the bag of tire shavings.
Among the biggest suspects, not surprisingly, are the millions of cars that pass nearby, shedding potentially toxic substances such as synthetic rubber from tires, motor oil, windshield washer fluid, transmission fluid, brake dust and automobile exhaust. Those and other substances are being tested at the lab.
If in fact the problem is cars, McIntyre wonders: “Could we find a vehicle pollutant source that is responsible for most of the toxicity?” And if so, could that help regulators deal with the issue?
For the study of the ground up tires, McIntyre will run water through the grounds — a process not unlike making a cup of coffee — and see whether the contaminants that leach out match the toxicity of urban stormwater runoff. Early results have been intriguing. Coho salmon have died when exposed to the leached chemicals, but other chemicals may have similar results and it’s too early to say what role automobile tires might have in pre-spawn mortality of coho in the wild.
“Stormwater is a complex mixture,” McIntyre says. It can vary by location (urban, rural), habitat type (forest, farmland), and season (winter, summer). As to its exact chemical makeup, she demurs. “That’s probably a better question for someone like Ed Kolodziej,” she says.
While McIntyre is doing the biology — by researching how salmon react to certain chemicals — Ed Kolodziej is looking deep into the chemistry of stormwater. When I find Kolodziej, he is at work in a backroom in the Department of Civil and Environmental Engineering, which serves as his makeshift office on the UW Seattle campus. He is just here for the day; he spends most of his time at the UW Tacoma Center for Urban Waters where he works with a series of specialized instruments that can measure the presence of molecules in a water sample in the parts per billion [Editor’s note: The Center for Urban Waters is the parent group of the Puget Sound Institute, which publishes Salish Sea Currents].
Kolodziej is using a process known as high resolution mass spectrometry to understand stormwater’s convoluted chemical makeup. If a particular chemical is in the water, the instruments at the lab are likely to find it. The proverbial needle in a haystack? No problem. But what if you don’t know exactly what you are looking for? That’s more difficult, and it's where Kolodziej’s work may differ from that of other researchers.
Conventional diagnostic methods are best at finding known chemicals. Typically, when facing a sample, a researcher will come up with a list of toxicants they think they’re likely to find, called targets, and then test for those substances.
What Kolodziej is helping to develop is a method for testing urban stormwater that uses liquid chromatography coupled to high-resolution quadrupole time-of-flight mass spectrometry. (This goes by the equally formidable acronym LC-QTOF-MS.) He and his co-authors described the process last August in a paper in Environmental Sciences: Processes & Impacts. “Basically, I think of it as someone going fishing for a specific kind of fish, versus a trawler that pulls everything in and then sorts through the catch,” says Kathy Peter, Kolodziej’s postdoctoral scientist and one of the study’s co-authors. “When you run a stormwater sample, you might see 1,000 or 2,000 features, and each feature is a chemical. Some of them will be natural, but some will be synthetic compounds that you need to test.”
This sort of non-target analysis is good at uncovering what Kolodziej calls the unknown-unknowns of stormwater. “What are the emerging contaminants?” he says. “We’re good at building analytical methods for things we know about, but there’s tons of stuff we don’t know about.” To help assemble a wide array of urban runoff samples, he has enlisted citizen scientists. If someone sees a salmon die in a stream, they can take water and tissue samples. Kolodziej can then analyze the water that salmon was swimming in, and try to figure out what killed it.
Even as Kolodziej and his colleagues have identified possible toxins, their precise origin remains as murky as the stormwater itself, at least in the published literature. “But,” he says, “cars and trucks seem to be the biggest culprits.” Motor vehicles are always shedding little bits of themselves as they whiz down roads and highways: flakes of brake, tire dust, droplets of motor oil, antifreeze and gasoline. All of it adds up. The question is how might these different substances interact with one another. Can they become greater than the sum of their parts?
As for what to do about stormwater, McIntyre and her colleagues at the Washington Stormwater Center are testing a variety of possible solutions. With coho and the lethal stormwater cocktail streaming off the 520 bridge, McIntyre was able to reduce the runoff’s toxicity simply by running it through a vertical soil treatment column: essentially, a barrel full of sand, shredded bark, and compost. After that, the coho were basically fine. (An interesting quirk is that chum salmon—or, as McIntyre calls them, “zombie fish” — were essentially untroubled by what killed the coho. “They just swam right through it like nothing,” she says.) And in a greenhouse down the road, Ben Leonard, one of her graduate students, is testing different lengths of swale for the extra removal of metals, running gallons of stormwater over a mix of Dutch clover and red fescue. His goal is to learn what a minimum effective length of swale might be, so Washington Department of Transportation engineers know how much to plant next to roads.
“It’s an issue of horizontal versus vertical, and how long stormwater stays in contact with the media,” Leonard says. “Horizontal is better for roadways, but vertical is good for, say, a road next to a steel refinery next to a river where salmon spawn.”
All of these fixes may one day solve what at present seems like an intractable problem. Once everyone has a better idea of the contaminants in stormwater, people can start to recommend changes in a policy sphere. Source control is always better than treatment after the fact, as Kathy Peter points out. “You don’t want to be managing a problem like this in perpetuity,” she says.