Selenium Standards Misinterpret Key Research

NOTE: The following is one of a series of case studies produced by the Union of Concerned Scientists' Scientific Integrity Program between 2004 and 2010 to document the abuses highlighted in our 2004 report, Scientific Integrity in Policy Making.

In late 2004, the Environmental Protection Agency (EPA) proposed new selenium standards that scientists say are much too high. The new standards were based on what scientists say was a faulty reading of key research into the effects of selenium on fish populations. Rather than correct its initial misinterpretation of the data, however, the EPA dug in its heels behind new standards that scientists said could be devastating to fragile stream ecosystems.

Selenium is a naturally occurring element that is discharged into the environment through mining activity. While selenium is an important micronutrient, it can be toxic to fish, birds and other wildlife in even small concentrations.¹ The EPA has set a drinking water standard for selenium of 0.05 parts per million (ppm).

There was widespread consensus within the environmental community that a different standard was needed for protecting aquatic life. Since selenium poisoning in fish was not closely related to concentrations in water, aquatic toxicologists agreed that new standards should instead measure the element's concentration in fish tissue. When the EPA released a draft of the new, proposed standards scientists were infuriated that the agency had set the selenium standard for concentrations in fish tissue at a much higher equivalent level, 7.9 ppm.
The EPA based its proposed new standard on a misinterpretation of a study by U.S. Forest Service aquatic toxicologist Dennis Lemly.² Lemly himself said the EPA contractor drafting the new standards had arrived at the number in error. Lemly was quoted as saying "the tissue levels of selenium the EPA is proposing would kill half the fish in my study, which is clearly not sufficient to protect aquatic life. The idea is preposterous."³  The new standards wrongly assumed that the study was based on a sample size of 210. While the study began with 210 fish, Lemly had removed fish during the study. In an interview Lemly explained that "this error made it appear that, at a given selenium tissue level, a smaller percentage of fish had died than actually did."4

The EPA's standards were problematic in that they assumed an acceptable level of selenium was one which killed "only" 20 percent of fish. Yet as Greenwatch reported, "In the past the EPA has kept the sub-lethal effects of a contaminant under 10 percent."5

Lemly joined four U.S. Department of Interior (DOI) experts in a detailed critique of the draft new standards, recommending that the permissible selenium level in tissue be set instead at 5.85 ppm."6 At least 90 other scientists and professors signed a letter of protest against the proposed new selenium standards.7

At first the U.S. Fish and Wildlife Services used the Endangered Species Act to block the draft of the new selenium standards and give the EPA a chance to correct its mistake. In the meantime, the coal-mining industry used Lemly's study "to try to influence West Virginia to adopt weaker selenium standards."8

When the draft was finally released in 2004, the 7.9 ppm standard level remained.9 Carlos Delos, selenium proposal manager for the EPA, defended the standard, noting that the new draft calls for additional monitoring if tissue concentrations reach 5.85 ppm in the fall. Agency scientists found this guideline confusing.10

Scientists and environmental groups are concerned that America's aquatic ecosystems will suffer as a result of the administration's refusal to base its selenium standards on the best scientific information. In the words of Aaron Colangelo of the Natural Resources Defense Council, "relaxing the standard would allow polluting industries to save millions of dollars while leaving wildlife to pay the ultimate price."12

1. Environmental Protection Agency, “Consumer Fact Sheet on Selenium (pdf),” accessed December 8, 2006.
2. Lemly, A. D. 1993. Metabolic Stress During Winter Increases The Toxicity of Selenium To Fish. Aquatic Toxicology 27: 133, accessed December 8, 2006 (subscription required)
3. “EPA Proposal to Weaken Water Quality Spurs Outrage,” Greenwatch Today, 18 April 2005, accessed 18 September 2006
4. Rebecca Renner, “Proposed Selenium Standard Under Attack,” (pdf) Environmental Science and Technology, 2 March 2005, accessed 18 September 2006
5. Greenwatch.
6. Skorupa, J. P., T. S. Presser, S. J. Hamilton, A. D. Lemly, B. E. Sample.  EPA’s Draft Tissue-Based Selenium Criterion: A Technical Review. Draft 2004.
7. Letter to EPA Administrator Stephen Johnson, April 18, 2005, accessed December 8, 2006
8. Greenwatch.
9. EPA Proposed Standards, “Notice of Draft Aquatic Life Criteria for Selenium and Request for Scientific Information, Data, and Views,” Federal Register, December 2004, accessed December 8, 2006
10. Renner.
11. “EPA Proposes to Lower Selenium Standards for Rivers and Streams,” Natural Resources Defense Council, 4 March 2005, accessed 18 September 2006



We Need Your Support
to Make Change Happen

We can ensure that decisions about our health, safety, and environment are based on the best available science—but not without you. Your generous support helps develop science-based solutions for a healthy, safe, and sustainable future.