FINAL REPORT ON THE AQUATIC MERCURY ASSESSMENT STUDY

Abstract

In February 2000, the United States Environmental Protection Agency (EPA) Region 4 issued a proposed Total Maximum Daily Load (TMDL) for total mercury in the middle and lower Savannah River. The initial TMDL, which would have imposed a 1 ng/l mercury limit for discharges to the middle/lower Savannah River, was revised to 2.8 ng/l in the final TMDL released in February 2001. The TMDL was intended to protect people from the consumption of contaminated fish, which is the major route of mercury exposure to humans. The most bioaccumulative form of mercury is methylmercury, which is produced in aquatic environments by the action of microorganisms on inorganic mercury. Because of the environmental and economic significance of the mercury discharge limits that would have been imposed by the TMDL, the Savannah River Site (SRS) initiated several studies concerning: (1) mercury in SRS discharges, SRS streams and the Savannah River, (2) mercury bioaccumulation factors for Savannah River fish, (3) the use of clams to monitor the influence of mercury from tributary streams on biota in the Savannah River, and (4) mercury in rainwater falling on the SRS. The results of these studies are presented in detail in this report. The first study documented more » the occurrence, distribution and variation of total and methylmercury at SRS industrial outfalls, principal SRS streams and the Savannah River where it forms the border with the SRS. All of the analyses were performed using the EPA Method 1630/31 ultra low-level and contaminant-free techniques for measuring total and methylmercury. Total mercury at National Pollutant Discharge Elimination System (NPDES) outfalls ranged from 0.31-604 ng/l with a mean of 8.71 ng/l. Mercury-contaminated groundwater was the source for outfalls with significantly elevated mercury concentrations. Total mercury in SRS streams ranged from 0.95-15.7 ng/l. Mean total mercury levels in the streams varied from 2.39 ng/l in Pen Branch to 5.26 ng/l in Tims Branch. Methylmercury ranged from 0.002 ng/l in Upper Three Runs to 2.60 ng/l in Tims Branch. Total mercury in the Savannah River ranged from 0.62 ng/l to 43.9 ng/l, and methylmercury ranged from 0.036 ng/l to 7.54 ng/l. Both total and methylmercury concentrations were consistently high in the river near the mouth of Steel Creek. Total mercury was positively correlated with methylmercury (r = 0.88). Total mercury bound to particulates ranged from 41% to 57% in the river and from 28% to 90% in the streams. Particulate methylmercury varied from 9% to 37% in the river and from 6% to 79% in the streams. Small temporary pools in the Savannah River swamp area near and around Fourmile Branch had the highest concentrations observed in the Savannah River watershed, reaching 1,890 ng/l for total mercury and 34.0 ng/l for methylmercury. The second study developed a mercury bioaccumulation factor (BAF) for the Savannah River near SRS. A BAF is the ratio of the concentration of mercury in fish flesh to the concentration of mercury in the water. BAFs are important in the TMDL process because target concentrations for mercury in water are computed from BAFs. Mercury BAFs are known to differ substantially among fish species, water bodies, and possibly seasons. Knowledge of such variation is needed to determine a BAF that accurately represents average and extreme conditions in the water body under study. Analysis of fish tissue and aqueous methylmercury samples collected at a number of locations and over several seasons in a 110 km (68 mile) reach of the Savannah River demonstrated that BAFs for each species under study varied by factors of three to eight. Influences on BAF variability were location, habitat and season-related differences in fish mercury levels and seasonal differences in methylmercury levels in the water. Overall (all locations, habitats, and seasons) average BAFs were 3.7 x 10{sup 6} for largemouth bass, 1.4 x 10{sup 6} for sunfishes, and 2.5 x 10{sup 6} for white catfish. This study showed that determination of representative BAFs for large rivers requires the collection of large numbers of fish and aqueous methylmercury samples over at least one complete seasonal cycle from the entire area and all habitats to be represented by the TMDL. The third study concerned the use of clams as an indicator of the influence of SRS discharges on mercury levels in Savannah River biota. Mercury levels were compared in Asiatic clams (Corbicula fluminea) collected from the discharge plumes and just upstream of the mouths of SRS creeks and two creeks located downstream from the SRS. Asiatic clams were selected for study because their relatively sedentary behavior made them a better indicator of local mercury exposure than mobile organisms. « less