Coho are killed by pollution as soon as they hit their natal streams in an estimated 40 percent of their range in Puget Sound.

Share story

The sweet seep of autumn rain is bringing coho salmon back home to their natal streams all over the Puget Sound basin — where too often they encounter a bitter truth: pollution in a shocking 40 percent of their home range so bad it can inflict a swift death.

The culprit is stormwater, and it is causing death rates so high, some populations of wild coho are at risk of local extinction, researchers found.

Yet there also are surprisingly simple and cheap solutions at hand, the researchers wrote in their paper, published today by the Ecological Society of America in the scientific journal Ecological Applications.

Researchers at NOAA’s Northwest Fisheries Science Center in Seattle worked with collaborators, including the U.S. Fish and Wildlife Service, local tribes and the Wild Fish Conservancy, to survey 51 sites from 2000 to 2011 in streams all over the Puget Sound basin. They used the survey data with a new computer model to map predicted coho death rates. The results show that in an estimated 40 percent of their range in the Puget Sound Basin, 10 to 40 percent of coho salmon die before they can even spawn because of pollution.

The researchers hope their survey and mapping methods can be taken up by others in the region and beyond, including citizen science and restoration groups, to help communities grapple with growth and its impact on salmon.

“It can be extremely useful for urban planning to know that if you develop this area and in an old-school way you are going to take out a lot of coho,” said Blake Feist, lead author on the paper.

The stormwater problem has long bedeviled the region.

Millions of dollars have been spent to upgrade the physical environment of urban streams, from adding big logs to slow flows and restore a more natural streambed complexity, to restoring streamside vegetation. Even more money from taxpayers, utility-rate payers and businesses is going into tighter regulation and treatment of so-called point-source pollution — dirty water from factories, sewage-treatment plants, combined sewer overflows and other sources.

But an insidious pollutant remains: stormwater. Every time it rains, a witch’s brew of pollutants from every impervious surface — roofs, parking lots, roads and more — sluices into storm drains. Most of it passes untreated into Puget Sound.

Researchers have known since at least the late 1980s that stormwater and salmon don’t mix. “Whatever it is, it has been killing coho for a long time,” said Nathaniel Scholz, a research zoologist in the Ecotoxicology Program at NOAA’s Northwest Fisheries Science Center, who co-authored the paper and published a series of other influential papers beginning in 2011 on so-called pre-spawn mortality.

Afflicted females full of eggs, back from their long journeys in the sea, ripe and ready to spawn, instead become disoriented when they encounter their home streams, such as Seattle’s Longfellow Creek. They gasp. Swim weakly at the surface. Lose their equilibrium. And within just a few hours, die.

Ironically, some projects to benefit coho and other species by restoring urban streams remain traps where coho go home to die.

By pinpointing the region’s coho die-off hot spots, researchers hope the map can be useful to local governments, communities and urban planners as development and redevelopment continue to reshape the landscape in Seattle and beyond.

Green zones on the map show where coho are still doing well and should be protected from degradation.

Other areas and especially red zones are trouble spots, where no further degradation of the freshwater habitat — such as tree clearing — should occur and a surprisingly simple fix might make a difference where sites are redeveloped, or yet unbuilt.

In earlier lab experiments, written up in 2016, scientists found that filtration of stormwater through a matrix of soil, bark chips and gravel was remarkably effective. While 100 percent of hatchery coho exposed to highway stormwater died, none of the fish in the same stormwater after it was passed through the filter showed any symptoms of distress.

Researchers are still searching for just what it is in unfiltered stormwater that is so toxic to coho (chinook and chum are unaffected). Correlation of traffic volume to increased rates of coho death by stormwater points to a possible culprit: motor vehicles.

In their research, scientists found that so-called coho pre-spawn mortality is related not only to the percentage of impervious surface in their home stream environment, but specifically to the volume of traffic.

Local roads did not see the same coho death rate. But the stew collected by researchers from the onramp to SR 520 headed toward Interstate 5 at Montlake, for example, killed 100 percent of the hatchery coho exposed to it. The more cars stoking the stormwater brew, the higher the death rate.

“It doesn’t prove that cars driving on roads causes pre-spawn mortality,” Feist said. “But it sure does point a finger at it.”

Is it tire dust? Crankcase oil? Brake-pad dust? Soot settled out from tailpipe exhaust?

Scholz said researchers are going to use this year’s rainy season to keep probing the question. The subject of the team’s focus now is tire dust, a fine black grime that leaches a combination of chemicals into streams in stormwater.

Every time it rains.