Not so fast: Why dispersants EPA ranks as “practically non-toxic” are still a concern

Richard Denison, Ph.D., is a Senior Scientist.

To judge by the headlines and leads of stories that ran on the websites of the New York Times (“The E.P.A. on Dispersants: Cure Is Not Worse Than the Disease”) and the Washington Post (“Oil dispersant does not pose environmental threat, early EPA findings suggest”) reporting on an EPA conference call held Wednesday, you’d think the first round of test results on dispersants conducted by EPA answered all outstanding questions and gave their use a clean bill of health.

Hardly.  (Despite the misleading headlines and leads, the rest of these stories were more nuanced and more accurate.)

As I reported in a post Wednesday, the new acute toxicity data were from tests conducted on the dispersants by themselves, rather than mixed with oil – which is what the environment sees.  Moreover, the new data did little more than confirm already available data showing that currently listed dispersants exhibit relatively low acute toxicity to fish and shrimp, and that by themselves they are less toxic than oil by itself.

So despite the hoopla, the new data from this first round of testing are of very limited utility in answering any of the more profound questions surrounding the use of dispersants. 

Those questions include:

  • Do dispersants increase the acute toxicity of oil to aquatic organisms?

If the previous data are any guide, we can expect the next round of EPA testing now underway – not on dispersants alone, but on mixtures of dispersants plus oil – to show significantly elevated levels of acute toxicity.

That means that EPA’s finding reported Wednesday that Corexit 9500 when tested alone ranks as “practically non-toxic” to fish and “slightly toxic” to shrimp will likely be elevated for the dispersant-oil mixtures to higher rankings, perhaps to “moderately toxic” or even “highly toxic.”

  • What are the effects of repeated or sustained, rather than short one-time, exposures to dispersants and dispersed oil?

The new data were derived from what is called a static test protocol, in which organisms are exposed to a concentration of dispersant that is not replenished.  Moreover, the tests are very short:  48 or 96 hours duration.  Yet organisms might well be exposed repeatedly or on an ongoing basis, something not addressed by the new data.

  • Do dispersants when added to oil cause sublethal effects, and if so, at what concentrations?

EPA’s classifications reported Wednesday are driven solely by tests looking for acute lethality:  What concentration of a dispersant is needed to kill half of the organisms within the test’s short duration.

It is far more likely that the adverse effects of most concern will be more subtle, for example, reductions in growth or reproduction of aquatic organisms, including microorganisms such as plankton.

By the way, less discussed by EPA are data it released Wednesday on cytotoxicity – which measures the ability of a chemical to kill individual cells.  In several cell types and assays, Corexit ranked among the most cytotoxic of the eight dispersants tested.

  • What effects does dispersed oil have on other organisms that live or feed not only in the water column but on the bottom or in sediments, on corals and other more sedentary organisms, on coastal and estuarine organisms, etc.?

EPA has not yet conducted any tests on these types of marine organisms.

  • Does dispersant use facilitate or impede volatilization, dissolution and degradation of oil, and how does it affect the composition and concentration of various degradation products?

A core assumption driving dispersant use is that, by breaking oil into small droplets, the rate of dissolution and degradation will be enhanced.  The data on this question, however, are decidedly mixed.  Some studies have actually found that dispersants can impede degradation of oil as a whole or of specific chemical components of the oil.  See pages 165-181 of the 2005 report, Oil Spill Dispersants: Efficacy and Effects, by the National Academy of Sciences.

  • Does dispersant use facilitate or impede bioavailability of oil and its toxic components?  How does it affect bioaccumulation and hence the potential for food chain contamination?

Data on this question are also very limited, according to the 2005 NAS report (see pages 215-219).

  • Are chemical dispersants more effective at dispersing and breaking down oil than physical dispersion processes?

And perhaps the ultimate question:

  • Where and under what conditions should chemical dispersants be used in lieu of mechanical methods for containing and collecting spilled oil?

For more reading on this entire subject, I do strongly recommend the National Academy’s 2005 report, which summarizes the available data and proposes a research agenda for supporting more scientifically sound decision-making on dispersant use.  What stands out from reading that report is how many fundamental questions about all aspects of dispersant use, including those above, remain unanswered.

What is even more sadly ironic is that many of those same questions were raised in an earlier NAS report, Using Oil Dispersants in the Sea.  The date on that report: 1989.

I’ll end with this question:  This time around, will we finally commit to finding answers to these questions, so that we have at least some of the answers in time for the next oil disaster?

This entry was posted in Environment, Health science and tagged , , . Bookmark the permalink. Both comments and trackbacks are currently closed.


  1. jim wadell colo.
    Posted July 6, 2010 at 10:12 pm | Permalink

    If its possible to dump junk in the pipe line, in an attempt to plug the leak. It may be possible to bind a tire & tube,or build a sort of inflatable plug ,track it to a certain spot in the pipe and inflate it

  2. Posted July 7, 2010 at 2:55 am | Permalink

    Please view this researcher's comments on the toxicity of corexit and the toxicity test parameters that declared corexit 'safe.'