8281 - Laserfiche WebLink

Home Browse Search

COMMENT AR Y 229 invertebrates to 2-5 j,lg/g. Based on a protocool for aquatic hazard assessment of selenium in five ecosystem compo- nents (water, sediments, invertebrates, fish eggs, and bird eggs; Lemly, 1995), there was still a moderate hazard at Belews Lake in 1996 due to ecosystem loading prior to 1986. This moderate hazard is due to the recycling of selenium in the aquatic food chain. The selenium contamination event at Belews Lake can also be considered a pulse event, in that adverse effects have been documented 10 years after sel- enium input had ceased (Lemly, 1993b, 1997). Because of the long-term effects of pulse events from selenium exposure in aquatic ecosystems, site-specific water quality criteria must consider the effects of pulse events (Lemly, 1998). Another example of aquatic ecosystem loading of sel- enium is given by Maier et al. (1998). They reported that application of seleniferous fertilizer to a deer forage range in California resulted in a pulse of selenium entering a stream and briefly raising the water concentrations from < 1 j,lg/L to 10.9 j,lg/L at 3 h post-application. Selenium concentra- tions in aquatic invertebrates in the stream increased from 1.67 j,lg/g before application to 4.74 j,lg/g 3 h after applica- tion and remained elevated after 2 (4.02 j,lg/g), 4 (4.99 j,lg/g), 6 (4.21 j,lg/g), 8 (4.30 j,lg/g), and 11 (4.54 j,lg/g) months post- application. This is the first study to determine that a short pulse event can load an aquatic environment quickly and that the selenium can be conserved in the ecosystem. These concentrations of selenium in the food chain are a potential ecotoxic problem based on the results of laboratory and field studies (Maier and Knight, 1994; Lemly, 1993a, 1996). 4. CRITIQUE OF A SEDIMENT-BASED SELENIUM CRITERION Recent articles by Canton and Van Derveer (1997) and Van Derveer and Canton (1997) concluded that the chronic water quality criteria for the protection of fish and wildlife from the bioaccumulative effects of selenium should be expressed on a particulate basis, such as sediment selenium concentration or a measure of the organic content of sedi- ment where selenium would be expected to accumulate. In an example in their second article, they present a sediment selenium model and use it to derive a site-specific chronic dissolved selenium standard of 31I1g/L, using a sediment selenium toxicity threshold of 2.5 j,lg/g and a site-specific mean sediment total organic carbon of 0.5%. This site- specific standard contrasts sharply with the current USEP A criterion of 5 j,lg/L. 4./. Belews Lake. North Carolina First, in criticizing the use of waterborne criteria, Canton and Van Derveer (1997) state that recent "recommendations of 2 j,lg/L or less... are based primarily on data from two sites ...: Belews Lake and Kesterson." There are several field studies from other locations, which suggest that a rec- ommendation of 2 I1g/L is appropriate. The field locations are Martin Lake, Texas (Lemly, 1985a; Sorensen, 1988), Hyco Reservoir, North Carolina (Lemly, 1985a; Gillespie and Baumann, 1986; Woock and Summers, 1984), and Chevron marsh (Richmond), California, Salton Sea, Califor- nia, and Swedish lakes (reviewed by Skorupa, 1998). Second, in reviewing the justification for the current USEP A national water quality criterion, the literature re- view of Canton and Van Derveer (1997) did not include several important publications, which influenced their con- clusions. For example, they state .. ... Belews Lake... im- pacts to fish populations . . . seemingly resulted from elevated dietary Se." They also state, .. ... waterborne con- centrations that reportedly eliminated many of the fish spe- cies of Belews Lake." In point of fact. there is a substantial literature base to indicate that selenium caused the observed adverse effects at Belews Lake (Cumbie and Van Horn, 1978; Duke Power Company, 1980; Finley, 1985; Lemly, 1985a, b, 1993b; Couglan and Velte, 1989). Later, the authors state "This resulted in elevated Se in sediments and subsequently in the food chain, although water column con- centrations were relatively low, in the range of 10 j,lg/L." This statement does not take into account a substantial amount of literature reporting that waterborne uptake of selenium by primary producers and zooplankton occurs, thus loading the water-column food chain with selenium (reviewed by Ihnat, 1989; Maier and Knight, 1994). Plank- tonic food chains in Belews Lake contained toxic concentra- tions of selenium derived from waterborne uptake (Cumbie and Van Horn, 1978; Lemly, 1985a, b). Further, from a bio- logical standpoint, 10 j,lg/L is not "relatively low" because reviews of several field studies have found adverse effects in aquatic organisms at water concentrations ranging from 2 to 5 j,lg/L (Maier and Knight, 1994; Lemly, 1996; Skorupa, 1998). Canton and Van Derveer (1997) also stated concerning Belews Lake that "An arm of the lake [Highway 158 arm] that did not receive coal fly-ash effluent did not appear to have any discernible toxic effects on fish." To the contrary, Sorensen et al. (1984) reported adverse effects in fish from this arm of the lake, and Holland (1979) reported selenium concentrations in plankton, benthic invertebrates, and fish from this arm of the lake that were above toxic thresholds. 4.2. Gunnison River. Colorado Canton and Van Derveer (1997) list two citations (Butler et aI., 1991, 1994) to indicate that high selenium concentra- tions were found in the Gunnison River in water, sediment, and tissues where they state that no evidence of biological impacts was observed. These two studies by Butler et al. (1991,1994) were contaminant surveys conducted as part of the NIWQP in the Gunnison River basin-Grand Valley