Environ. Sci. Technol. 2009 43, 8759–8766
Mercury Concentrations in Surface Water and Harvested Waterfowl from the Prairie Pothole Region of Saskatchewan BRITT D. HALL,* LAUREN A. BARON, AND CHRISTOPHER M. SOMERS Department of Biology, University of Regina, 3737 Wascana Parkway, Regina, Saskatchewan, S4S 0A2 Canada
Received August 17, 2009. Revised manuscript received October 7, 2009. Accepted October 12, 2009.
Mercury cycling in prairie ecosystems is poorly understood. We examined methylmercury (MeHg) concentrations in whole water from 49 diverse prairie wetlands and lakes in Saskatchewan. We also determined total Hg (THg) concentrations in waterfowl harvested by hunters for consumption. Average whole water MeHg concentrations ranged from 0.02 to over 4 ng L-1 and were higher in water from wetland ponds compared to those in lakes. High MeHg concentrations in prairie wetlands present the possibility of increased Hg concentrations in biota inhabiting these and other similar systems. We therefore measured THg in 72 birds representing 13 species of waterfowl that commonly use prairie aquatic habitats. A large range in THg concentrations was observed among individual birds, with values ranging from below the detection limit to over 435 ng g-1. When waterfowl were classified according to diet, we observed clear evidence of THg biomagnification with increasing proportion of animal prey consumed. THg concentrations in waterfowl collected by hunters did not exceed consumption guidelines of 0.5 mg kg-1 developed for fish. This is the first study that has reported MeHg concentrations in water from the prairie pothole region of southern Saskatchewan.
Introduction Anthropogenic emissions of inorganic mercury (HgII) have increased long-range transport of Hg and subsequent pollution of remote areas (1). On a global basis, the major sources of Hg to the atmosphere are emissions from coal-fired energy production and waste combustion (2), fluxes that are likely to increase with further global industrialization (3). Generally, HgII enters aquatic ecosystems from the atmosphere as direct deposition on surface waters (4) and runoff from terrestrial areas (5, 6). Once present in anaerobic regions of aquatic systems (e.g., wetland soils, lake sediments), HgII can be converted through microbial activity into methylmercury (MeHg, 7,8). In situ production of MeHg via microbial methylation is the main source of MeHg in most aquatic systems (9, 10). The amount of MeHg produced in an aquatic environment will depend on factors that control microbial population growth or metabolic function (7, 11, 12) as well as the amount of HgII available for methylation (13, 14). Environmental factors that promote HgII methylation such as the availability of organic carbon and sulfate and tem* Corresponding author e-mail:
[email protected]. 10.1021/es9024589 CCC: $40.75
Published on Web 10/20/2009
2009 American Chemical Society
perature are particularly important in wetland sediments (8, 15), suggesting that prairie wetlands have an elevated potential for high methylation rates and, therefore, high MeHg concentrations. Methylmercury is a neurotoxin that is easily bioaccumulated by humans and wildlife that consume aquatic organisms such as insects or fish (16, 17). Consumption of excess Hg by pregnant or breastfeeding women and children is of special concern because MeHg is potentially dangerous for children and the developing fetus (18). Health concerns related to consuming fish and wildlife with elevated Hg concentrations continue to be widespread as evidenced by the number of Hg consumption advisories currently issued in North America (19, 20). Mercury deposition and subsequent MeHg production may be of special concern for wildlife and humans in the prairie pothole region in central North America, which is characterized by a high density of small, diverse, wetland ponds that are critical wildlife habitats. Because these ecosystems provide cover and nesting sites for many game and nongame wildlife species as well as being some of the most important waterfowl breeding habitat remaining in North America (21) high MeHg concentrations could pose significant health and environmental risks to prairie wildlife that consume aquatic organisms inhabiting those ecosystems. Despite this possibility, there have been few studies examining Hg cycling in these important prairie ecosystems. In other ecosystems, studies have shown that piscivorous animals such as mink (Mustela vison), river otters (Lutra canadensis), and loons (Gavia immer) are vulnerable to increased MeHg accumulation (22-25). Elevated MeHg concentrations have also been measured in insectivorous song birds such as tree swallows (Tachycineta bicolour) and red-winged blackbirds (Agelaius phoeniceus) (26) and nonwetland species (25). Because birds and other vertebrates obtain the majority of MeHg from their diet (27, 28), waterfowl may accumulate high levels of MeHg if they are consuming plants and invertebrates from wetlands with high rates of MeHg production. For example, in 2005 three common duck species [northern shoveler, cinnamon teal (Anas cyanoptera), and common goldeneye (Bucephala clangula)] in Utah were found to have elevated Hg concentrations, and as a result the state of Utah issued one of the first Hg consumption advisory in North America for ducks harvested from lakes and wetlands with no known direct source Hg contamination (29). Waterfowl that use high MeHg wetland habitats can be an important source of food for people living on the prairies. Consumption of these animals by hunters and their families could be an important vector of neurotoxic MeHg to these communities. However, our current lack of knowledge regarding Hg levels in pothole systems precludes evaluation of risks to humans and wildlife. The goals for this study were to (1) measure MeHg concentrations in water from diverse prairie wetlands and lakes and (2) determine Hg levels in a variety of waterfowl that commonly inhabit these types of systems. Addressing this topic will provide information necessary to protect wildlife and human populations inhabiting these habitats.
Methods Sample Collection and Analysis. Water. Our first goal was to quantify MeHg concentrations in water from lakes and ponds across southern and central Saskatchewan (Figure 1), which is centrally located in the prairie pothole region of North America. Field sampling was conducted at 49 lakes and wetland ponds between May 2006 and August 2007 (Table VOL. 43, NO. 23, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
8759
FIGURE 1. Location of lakes and wetlands sampled for methylmercury in whole water. Eleven lakes and wetlands ponds were scattered across southern Saskatchewan. The remainder of lakes and wetland ponds were concentrated in three regions: Estevan, Humboldt, and at or adjacent to the St. Denis National Wildlife Area (SDNWA). Coordinates for each site are provided in Table S1 of the Supporting Information. S1 of the Supporting Information). We sampled a wide range of habitat types, including highly saline lakes, freshwater lakes, and a number of wetland ponds with varying conductivities, which we used as a proxy for salinity (Table S2 of the Supporting Information). Wetlands were classified as systems having representative wetland vegetation or maximum depths of less than 0.5 m. Nineteen wetland ponds were located on the St. Denis National Wildlife Area (SDNWA; 52° 12′ N, 106° 05′ W) and were classified into Type 3, 4, and 5 wetlands using the Circular 39 classification system of Shaw and Fredine (30). Surface water sampled for MeHg was collected using trace metal-free techniques from open water areas of the wetland ponds. In shallow areas less than 0.75 m deep, samples were taken by hand by wading into the open water area, taking care to avoid sampling any plume created. In waters deeper than 0.75 m, samples were taken by hand from a fiberglass canoe or inflatable boat. All surface water samples were taken in either Teflon or sterile fluoro-carbon polymer bottles for MeHg analysis and precleaned glass bottles for total mercury (THg; all forms of Hg) analysis. Samples for MeHg were immediately placed on ice and frozen within 48 h. Total Hg samples were preserved using trace metal-grade concentrated HCl to 1% of total sample volume. In order to prevent contamination during sampling, all water samples were taken using clean-hands-dirty-hands sampling protocols (15, 31). Samples for MeHg were distilled, ethylated, and analyzed by cold vapor atomic florescence spectrometry (CVAFS, 32, 33, and 34). Detection limits for MeHg in water were between 0.01 and 0.05 ng L-1. Matrix spike recoveries for MeHg were generally >80% and >90%. Total Hg samples were analyzed using CVAFS after BrCl oxidation and SnCl2 reduction (35). Total Hg analysis had a detection level of 0.2-0.3 ng L-1 at a blank level of 0.3-0.4 ng L-1. Lakes and wetland ponds were sampled for conductivity, pH, dissolved oxygen concentrations, and temperature using a YSI probe. Samples for dissolved organic carbon (DOC) and sulfate concentrations were taken and analyzed using standard methods (36). Specific UV absorbance (SUVA) was calculated by dividing the UV absorbance measured at λ ) 254 nm on an Agilent 8453 UV-visible spectroscopy system by the DOC concentration (37). Wavelengths of 254 nm are associated 8760
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 23, 2009
with the aromatic moieties in a sample and can be indicative of the quality of DOC sampled (38). Differences in THg and MeHg concentrations in water from wetlands compared to lakes were determined using Mann-Whitney rank sum tests. Forward stepwise multiple linear regressions were performed on log transformed data to determine which water chemistry variables predicted MeHg concentrations. Statistics for water data were done using SigmaStat 3.5. Waterfowl. Breast muscle tissue samples from 13 species of waterfowl harvested during the fall of 2006 and 2007 were donated by local hunters. The majority of the samples were obtained from dabblers such as mallards and geese inhabiting wetlands in the vicinity of Regina and Moose Jaw. The higher trophic level birds (bufflehead, scaup, and northern shoveler) were harvested from the north end of Last Mountain Lake (51°20′ N, 105°15′ W). A small piece of muscle (∼1-3 cm3) from the breast of each bird was removed by each volunteer, placed in a ziploc bag, and frozen until analysis. Volunteers were provided with clean razor blades and gloves and were cautioned against the possibility of contamination of samples. In the laboratory, a subsample of the tissue was cut from the inner portion of the sample to ensure minimal contamination. Total Hg was analyzed by CVAFS after digestion following EPA Method 1631 (35). Briefly, a small piece of tissue (∼0.3-0.6 g) from each sample was digested in 70:30 nitric acid:sulfuric acid at 125 °C for 2 h, oxidized with BrCl, and further digested overnight at 60 °C. Mercury in an aliquot of the digest (0.05 mL for carnivore samples and 2 mL for herbivore samples) was detected using a Tekran 2600 Hg detector. Recoveries of Hg from a certified standard reference material (National Research Council DORM-3) run concurrent with samples were between 100% and 102%. Acid blanks averaged 0.3-0.4 ng L-1, and detection limits were 0.1-0.3 ng g-1. Average Hg recoveries from acid blanks and sample duplicates spiked with a known concentration of a Hg standard were 97-130% and 94-127%, respectively. Approximately 10-15% of the tissue samples were analyzed in duplicate. All tissue data are presented on a wet weight basis. Samples were analyzed for THg rather than MeHg because nearly all of the mercury in muscle tissue of higher organisms
TABLE 1. Categorization of Feeding Groups Using Published Diet Data, Range of Total Mercury (THg) Concentrations, and Sample Size in Different Species of Waterfowl range of THg concentrations (ng g-1 ww) feeding group
species
common name
numbers of samples analyzed (n)
this study
Braune and Malone (48) diet ref
snow goose Canada goose
9 2
BDL -1.96 1.32-2.13
