One of the most prominent facts you may think you know about stormwater in Puget Sound isn’t true. Remember when the Department of Ecology touted the image that the torrent of pollutants washed every two years into Puget Sound was comparable to an Exxon Valdez oil spill disaster? Turns out, we now learn from Ecology, that’s simply wrong. How that big error came about is an important story that needs to be told and learned from.
If there is any saving grace, it’s that better information, as it becomes available, can strengthen and properly target both our public programs and private actions to make us more effective environmental stewards.
The Exxon Valdez imagery came out of an ongoing stormwater study conducted by the Department of Ecology, Phase 1 of which was completed in 2007. One of the big estimates in Phase 1 got a lot more attention than it deserved, given the caveats Ecology should have stressed. The estimate was that 52 million pounds of petroleum and other toxics — called “mass loadings” — are carried every year into Puget Sound.
“Runoff is like a slow-moving oil spill,” said Ecology’s special assistant for Puget Sound in November 2007. It wasn’t all petroleum hydrocarbons but a mix of toxic materials, as Ecology knew at the time. But a reporter latched onto the now-regrettable bright idea: Make the size of that big number meaningful to people by its comparability to an Exxon Valdez catastrophe. The image spread like an oil spill itself everywhere across the water.
Two problems. First, because many things other than petroleum are harmful in stormwater and their harmfulness is not necessarily proportional to their mass or weight, the Exxon Valdez and oil-spill imagery carried with it an entirely misleading message of what made up the entirety of the runoff threat. Ecology should have foreseen that trap, but it fell for the oil-spill image anyway. What Ecology publicists couldn’t know at the time, however, because relevant information has only been recognized and noted in subsequent phases of the study, is that the contribution of petroleum hydrocarbons to the overall problem of toxics in stormwater is dramatically smaller than first thought.
And that highlights the second problem. All the underlying data was weak in the first place. When Phase 1 was undertaken, nobody had ever systematically and directly measured the amount of toxics that flow into Puget Sound. So the entire exercise in estimation was calculated from assumptions. In simple terms, the Phase 1 calculation depended on assumptions of how much water flowed into Puget Sound (the flow model) and assumptions about the nature and amount of toxics the runoff water contained, gallon for gallon. Simple basic idea behind a very complicated spreadsheet: multiply flow model volumes times toxics concentrations (fractions of ounces per gallon, for lay understanding) to get total pounds of "loadings."
In Phases 2 and 3 of the study, after the image of a big oil spill was put out, Ecology stepped up to refining the flow model with actual measurement of real-world flows into the Sound from various real-world typical land uses. Revelation: The multipliers that had been used earlier from the flow model were significantly too large.
The new work found even worse trouble in the Phase 1 concentration assumptions that Ecology had gathered from generic national estimates or measurements taken from distant places. When Ecology finally had the benefit of actual measurements from actual places around Puget Sound under actual local conditions, the numbers were at major variance with the generic numbers the Phase 1 calculation had relied upon. At least the new numbers resulted from samples Ecology carefully drew from specific types of land — urban residential and commercial, agriculture, and forest lands — from actual river flows, and from different seasons of the year, allowing big gains in the ability to understand relationships between particular types of places and particular types and quantities of toxics in runoff.
As all this work came together, Ecology realized how badly out of whack the picture of the Exxon Valdez and a big oil spill was. Based on the new opportunity to make estimates from actual sample measurements, flows were lower and toxic concentrations were different from the driving components of the Phase 1 estimate.
The Phase 3 report containing the newest calculations is now circulating for peer review and has not been released. One reviewer who has seen the study described the reduction from earlier estimates of mass loadings, especially for petroleum, as “dramatic.” An Ecology scientist used the phrase “a lot, although petroleum is still very significant.” Stay tuned for what fraction of an Exxon Valdez spill the new estimate for toxics might be (although Ecology will be more cautious about lumping all toxics into a single estimate) and for petroleum hydrocarbons and other materials as toxic components. That is, if Ecology (or anyone who relied upon the picture Ecology drew) ever has anything to say about the Exxon Valdez in this context again.
But does that mean stormwater goes away as a big problem? Emphatically not. It just means our rhetoric about and approaches to the problem are going to have to catch up with the facts.
Stormwater presents various threats of severe damage to water quality and living things. Needs for protection will still remain, long after the Exxon Valdez image fades as a metaphor. But in the future, Ecology’s credibility will need to be better protected by bringing more circumspection to characterizations that rest on palpably inadequate data.
Here are some points to keep in mind. First, it’s always been a mistake to think that “loadings” measured in pounds were the sole crux of the issue, or that putting a single tall black hat squarely on petroleum could help the public to keep track of the cast in a complicated plot. There are lots of different toxic materials; even the many that the Phase 3 study focuses on are just the tip of the iceberg. But toxicity — the threat from toxics — can’t be evaluated just by weight; specific toxic substances vary in what they damage and how, and like must substances that poison, concentration is a key consideration in harmful dosage.
Another point is that the harmful effects of stormwater sometimes have little to do with toxics as we generally think of them. In some areas, especially from tracts of land in agricultural or forestry use, simple sediment (and sometimes bacteria from farm animals) carried away in storm runoff can clog streams, disrupt wetlands, and pour into rivers. The brown smudge across Elliott Bay after the heaviest rains two weeks ago came largely from far up the Green River watershed, not from urban runoff. Sometimes, however, it’s just the sheer erosive force of nothing but ordinary rain runoff from cleared land that sluices destructively through streambeds and banks, disrupting the living things in or on them. Rural areas and natural upland habitat can be even more sensitive than urban areas to these stormwater risks and consequences.
What kinds of areas are at risk from stormwater impacts? And how do the various kinds of impacts (toxics being just one) affect natural systems unique to each location — a marshy wetland in east King County, an oyster bed in Shelton, a salmon stream in Sedro-Wooley, a bulkhead on Seattle’s Elliot Bay waterfront? The kind of data Ecology is finally preparing to present allows a much more precise way of thinking about stormwater protections than the simplistic inference from the Exxon Valdez comparison — that arresting the oil drippings on roads and parking lots is the emblematic stormwater remedy.
Like what you just read? Support high quality local journalism. Become a member of Crosscut today!