This afternoon, on Day 72 of the Gulf spill, the Environmental Protection Agency (EPA) released the results of its first round of toxicity testing on the chemical dispersants being applied to the crude. Let’s break it down good news/bad news, because that seems kind of bloggy.
Good news: the EPA’s initial tests found that the dispersants Corexit 9500—the main dispersant BP has used in the Gulf—and other chemical brands ranged from “practically nontoxic” to “slightly toxic.” (OK, so “good news” is a relative term, but I’m trying my best to be more optimistic here.) Here’s what the EPA said in a statement:
EPA’s results indicated that none of the eight dispersants tested, including the product in use in the gulf, displayed biologically significant endocrine disrupting activity. While the dispersant products alone – not mixed with oil – have roughly the same impact on aquatic life, JD-2000 and Corexit 9500 were generally less toxic to small fish and JD-2000 and SAF-RON GOLD were least toxic to mysid shrimp. While this is important information to have, additional testing is needed to further inform the use of dispersants.
Here’s how the EPA defined toxic: lab workers determined what concentration of parts per million of the chemical—eight brands were tested—was needed to kill half the fish or shrimp in the EPA’s sample. (The EPA used shrimp and a small fish called inland silverside.) Corexit 9500—which is banned in Britain for use in oil spills because of concerns about its toxicity—was lethal to half the shrimp or fish at 130 parts per million, and therefor got the label “practically nontoxic. Dispersit SPC 1000, the worst-ranking chemical, killed half the fish or shrimp at just 2.9 parts per million, which is enough to qualify you as “moderately toxic.” You can find the full report here in PDF.
Paul Anastas, the EPA’s assistant administrator for research and development (and the father of green chemistry), said in a conference call with reporters this afternoon that the agency recognized that dispersants still carried side effects—and that they would continue to caution BP in their use. He noted that since the EPA Administrator had expressed concern about a month ago about the amount of dispersants BP had been using on the oil spill, the company had reduced the its volume by 70%. (Of course, during the same time period BP managed to partially cap the well and reduce the rate of the oil flow, which should have reduced the need for dispersants.) But the conclusion for the EPA, at least for now, was clear: using dispersants seemed to be better than not using them. “Oil is the number one enemy here,” Anastas said.
Bad news: where do I begin? There’s the fact that the EPA has only now completed the first round of toxicity tests on dispersants—on the 72nd day of the oil spill, after well over 100 million gallons of crude has spilled into the ocean, after BP has already poured nearly 1.6 million lbs. of chemical dispersants into the Gulf. Imagine if this is the way we tested drugs—let millions of people take it, and then come out with our safety recommendations. It certainly is a good thing that it turned out that BP’s favored brand, Corexit 9500, wasn’t overly toxic—would have been rather embarrassing if that didn’t turn out to be true.
Now, the EPA and industry might argue that Corexit 9500 and other dispersants have undergone toxicity tests before. Dispersants have to be tested to be included on the National Contingency Plan Product Schedule—the EPA’s list of accepted dispersants for oil spills. But that doesn’t mean they’ve actually been tested by the EPA, or by anyone independently. As Marian Wang points out over at ProPublica, the EPA has just been rubber-stamping safety tests carried out by industry:
The toxicity tests are part of the listing criteria only in the sense that they must be done and the paperwork must be filed with the EPA. But there’s no maximum toxicity for products to be included on the EPA’s list. That means the 14 products, once deemed to be above a certain threshold of effectiveness, could be placed on the list as long as toxicity tests were done and the paperwork was complete, regardless of what the toxicity tests found. (To be applied to an oil spill, however, a product that is authorized must still be approved by the federal on-scene coordinators and regional response teams.)
In other words, the EPA was taking the manufacturers’ word for it that their chemicals were safe—and only began testing independently a month into the spill. (Shades of the Mineral Management Service here, which let oil companies fill out environmental impact assessments themselves in pencil.) The toxicity definition used in this round of EPA tests is one the agency created on the spot—there was no toxic standard for dispersants before.
In addition, the EPA has yet to test the chemicals independently to see how effective they actually are at helping to clean up the oil—something that’s important, given that everyone recognizes that there’s a cost-benefit equation with dispersants. At least we now have an idea of the cost to the aquatic environment—though not fully, as the EPA hasn’t yet tested how toxic an oil/dispersant mix might be—but we don’t yet independently know the benefit. That might come later. Anastas stressed that the EPA planned additional rounds of tests—although given that the spill might well have ended by early August, when the relief wells are scheduled to be completed, it’s hard to say what benefit they might bring.
Indeed, the dispersants debacle is one more sign of just how unprepared both industry and the government were for a spill of this magnitude. The government let BP pour hundreds of thousands of gallons of chemical dispersants into the Gulf—and use them deep under water, something that had never been tried before—without any independent verification that this was safe. Maybe that’s acceptable in the early stages of a disaster—but on Day 72? That’s ridiculous.
Ultimately, though the EPA can’t bear that much blame. Our system for regulating environmental chemicals is wholly dysfunctional and completely in the favor of the industry. The EPA has only independently tested a relative handful of the tens of thousands of industrial chemicals approved for use now—otherwise, we depend on industry’s assurances. The situation is so badly skewed that the EPA was able to claim victory when it forced Nalco, the manufacturer of Corexit, to reveal the dispersant’s ingredients, which had been largely kept secret for business confidentiality, something that is regularly claimed by chemical companies. Think about that: in the midst of the greatest environmental disaster, the government had to force a company to reveal the ingredients of a chemical being poured into the oceans, and it still took weeks. As I wrote in a long story for TIME earlier this year, we need change in our chemical regulation system:
If scientists were slow to arrive at that conclusion, Washington has been even slower. The Toxic Substances Control Act (TSCA), the 34-year-old vehicle for federal chemical regulation, has generally been a failure. The burden of proving chemicals dangerous falls almost entirely on the government, while industry confidentiality privileges built into the TSCA deny citizens and federal regulators critical information about how substances are made and what their effects are. In the years since the TSCA became law, the Environmental Protection Agency (EPA) has been able to issue restrictions on only a handful of chemicals and has lacked the power to ban even a dangerous carcinogen like asbestos.
Until that happens, it’ll be mostly bad news for us.