J. Environ. Monit., 1999, 1 67N US Focus EPA’s selenium revision heads into murky waters As the US EPA revises its standards for selenium in fresh water, a heated debate rages between government scientists, who say that existing standards must be tightened because selenium contamination is widespread, and industry scientists who believe that it is an environmental oddity. Fish, crustaceans and birds appear to be uniquely sensitive to even modest increases in environmental concentrations of this element.To set new standards, EPA must find a way to take into account an emerging understanding of selenium’s complex geochemistry. Water monitoring misleading In 1987, EPA set a chronic standard for selenium based on studies of a reservoir called Belews Lake in North Carolina. Belews Lake is split into two parts.In one part, where dissolved selenium concentrations averaged 10 mg L-1, bioaccumulation in aquatic food chains caused massive reproductive failure in the reservoir’s fish. In the other part, with average concentrations below 5 mg L-1, no eVects were seen. So EPA set the chronic standard at 5 mg L-1. Since then the picture has changed.At Belews Lake much lower concentrations, 0.2–4 mg L-1 of dissolved selenium, have been found to cause fish deformities and other problems.1 In addition, researchers now understand that a standard to protect fish based solely on dissolved selenium has little scientific basis. This is because fish are mainly exposed to selenium through diet. Simply analysing for dissolved selenium ignores the metal’s complex biogeochemistry.As a result, EPA is revising its standards for acute and chronic exposure in fresh water.2 But this new information leaves EPA with a problem—what to measure?Water concentrations are related to eVects, but it is a non-linear relationship that also varies from one site to another—it might be amenable to modelling, or a statistical screen.Other candidates are selenium concentrations in fish tissues or sediments which are more directly related to exposure. But such measurements are still relatively new and there is disagreement about whether scientific knowledge could support regulations. So EPA brought together selenium experts in May 1998 for a workshop to evaluate its options.3 Reseachers now understand that seleniummust be converted to an organic compound before it becomes toxic.Seleno-methionine [(CH3Se(CH2)2CH(NH3)CO2H] is thought to be the key chemical form. This conversion is a function of the species of selenium in the water column, the water chemistry and the types of microorganisms in the sediment. The organic compound then cycles between sediments and the food chain as organisms accumulate selenium.But the details of this picture are far fromclear.3The type of ecosystemhas a large eVect on selenium cycling. Fast flowing rivers, slow moving rivers and lakes, saline systems and cold northern streams may all diVer in response to seleniuminput because they are divergent in water chemistry, organic productivity, sediment residence times and the types of microorganisms they support.To acknowledge these diVerences, EPA will have to craft site-specific guidelines. Site-specific guidelines are EPA-approved methods for modifying standards to account for individual site characteristics. To get a better handle on these relationships, the experts advised EPA that an analysis of selenium contamination in water should at a minimum include: dissolved versus particulate selenium which can then be diVerentiated into selenate (SeO-) and selenite (HSeO+SeO-).Peptide- and protein-bound forms of selenium are important because they are critically related to the potential for chronic eVects. The protein-bound forms should be specifically included in the analysis of selenium in the particulate fraction, since this is the primary step towards bioaccumulation.But even such a complete analysis is likely to prove misleading about whether selenium contamination is aVecting aquatic life—that is the advantage of tissue or sediment measurements. Tissue or sediment monitoring? The biggest advantage of tissue measurements is that tissue integrates all exposures, whether from food or water. The best tissues in which to measure selenium appear to be fish ovaries or eggs as concentrations have been linked to reproductive eVects in some species.3 But there needs to be a larger data set that encompasses interspecies variability in ovary concentrations.To date, studies have measured total selenium in tissues, but a more accurate representation could be obtained by measuring proteinor peptide-bound forms of organoselenium, because the incorporation of selenium into protein is a trigger for biological eVects.Sedimentary Selenium µg g-1 (dry weight) Adapted from Van Derveer and Canton 199768N J. Environ. Monit., 1999, 1 US Focus Sediment is the dominant sink for selenium and sedimentary organic materials are an important dietary source for aquatic invertebrates.But the literature relating sediment-based criteria is sparce. Fish data indicate that toxic eVects may occur when total sedimentary selenium concentrations exceed 4 mg g-1 (dry weight). It might be possible to relate water column measurements to sediment concentrations.3 Widespread or rare? The selenium revision has also heated up debate about the extent of selenium contamination.Department of Interior scientists who work on sites aVected by selenium say that the problem is widespread and want to see the current standard, 5 mg L-1, cut in half. Industry scientists who do not see eVects in their areas, strongly oppose this suggestion. Deformities in young birds or fish that result from exposure to too much selenium are widespread, according to Joseph Skorupa, a US Department of Fish and Wildlife biologist in Sacramento, CA.Government scientists conducted screening surveys in 14 western states and found that out of 161 sites sampled for bird eggs, 79 sites yielded bird eggs with enough selenium to expect toxic eVects, he said, adding that more surveys would reveal more sites. A lower standard is necessary, he emphasized, to protect fish and birds.On the other hand, industry scientists contend that most of the problems for fish and birds have arisen from selenium in soils which is naturally high in the western US. Adverse eVects to wildlife from selenium contamination are very limited in other areas, according to William Adams, enviromental director at Kennecott Utah Copper Corporation in Magna, UT. Like many metals, selenium is a necessary micronutrient, essential for life at small doses, but toxic in larger amounts. In humans, selenium deficiency and toxicity are recognized, and this element is believed to have a role in cancer prevention.There are regions, such as places on the Canadian Shield, where selenium concentrations are so low that algae grow better when selenium is added to the water.But for fish there is at least a 10-fold diVerence between insuYcient and toxic concentrations.3 In most field situations, it does not appear that insuYciency presents a problem. Since it is clear that the characteristics of specific sites are very important in determining what level of selenium contamination causes harm to fish and wildlife, many observers believe that EPA will develop a two-part strategy. Such a strategy would use a screening measurement, perhaps water concentrations, that would trigger tissue or sediment monitoring.This means that the revision will have to place a great deal of emphasis on the site specific guidelines. The agency is aiming to produce a draft for a chronic freshwater standard and site-specific guidelines by Spring 2000, and final recommendations by Fall 2000. Notes 1 A. D. Lemly, T eratogenic eVects of selenium in natural populations of freshwater fish, Ecotoxicol. Environ. Safety, 1993, 26, 181 2 United States Environmental Protection Agency, OYce of Water, Revision of Selenium Aquatic L ife Criteria, Fact Sheet EPA-822-F-98-003, 1998. URL: http://www.epa.gov/ost/Rules/selenium/ factsh.html 3 United States Environmental Protection Agency, Report on the Peer Consultation Workshop on Selenium Aquatic T oxicity and Bioaccumulation, EPA-822-R-98-007, 1990. URL: http://www.epa.gov/OST/ selenium Rebecca Renner Science writer and editor based in the US Tel: +1 570 321 8640; Fax: +1 570 321 9028, e-mail: applepie@sunlink.net