Environ. Sci. Technol. 2010, 44, 1860–1867
Local to Continental Influences on Nutrient and Contaminant Sources to River Birds CHRISTY A. MORRISSEY,† JOHN E. ELLIOTT,‡ AND STEPHEN J. ORMEROD† Catchment Research Group, School of Biosciences, Cardiff University, Cardiff, United Kingdom, CF10 3AX, and Pacific Wildlife Research Centre, Environment Canada, 5421 Robertson Rd., Delta, British Columbia, Canada, V4K 3N2
Received October 9, 2009. Revised manuscript received January 14, 2010. Accepted January 19, 2010.
Stable isotope and contaminant analyses revealed differences in nutrient sources and contaminant pathways in two species of dipper, Cinclus mexicanus and Cinclus cinclus in western Canada and western Britain. We collected dipper prey and eggs from relatively pristine British Columbian rivers contrasting in the presence of salmon-derived nutrients, and rural Welsh rivers with varying food-web complexity associated with stream acidity. Enriched δ13C and to a lesser extent δ15N in American dipper eggs and prey confirmed that streams with migrating Pacific salmon were enriched with marine-derived nutrients, but overall contaminant concentrations did not differ strongly in dipper eggs between rivers with and without salmon. However dipper contaminant profiles, particularly PBDEs, mercury and ΣDDTs, were related to δ13C, reflecting the marine influence and greater fish consumption at salmon sites. Irrespective of catchment influences and despite feeding at a higher trophic level, American dipper eggs (n ) 17) contained lower levels of organohalogens than Eurasian dippers (n ) 37), but with similar PCB (153 and 138) and PBDE (47 and 99) congeners dominanating. Eurasian dipper eggs from circumneutral streams contained more dieldrin, ΣDDT and Σhexachlorocyclohexanes while Σchlordanes, mirex, ΣPBDEs and ΣPCBs predominated at acid sites. Our data reveal how dippers indicate contaminant levels and sources under contrasting conditions at scales ranging from local to intercontinental, but local environmental conditions apparently alter feeding ecology and exposure pathways even in these closely related species.
Introduction Monitoring birds’ eggs for persistent contaminants in freshwater, marine, and terrestrial habitats has revealed spatiotemporal patterns over broad geographic scales (1-3). However, egg contaminants are affected by local differences in prey availability and individual foraging ecology (2, 4, 5), especially since food resources vary in time and space even in similar habitats (6). Understanding such variability is fundamental in tracing how birds acquire environmental contaminants during egg formation. * Corresponding author phone: 029 20 875871; fax: 029 20 874305; e-mail:
[email protected]. † Cardiff University. ‡ Environment Canada. 1860
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 5, 2010
Birds’ eggs are often assumed to reflect local contaminant conditions, but variations in nutrient acquisition, prey selection and breeding strategy (dietary versus maternal sources) during egg production can be important (7-9). In general, there is limited knowledge of how birds obtain essential nutrients during egg formation, and of what prey items are taken during this critical phase (10, 11). More importantly for monitoring, spatiotemporal variation in prey use and availability could influence contaminant exposure during egg production. Changes in food web structure may impact contaminant availability and uptake in predators by altering the existing prey (8, 12). For example, Pacific salmon are important vectors of marine derived nutrients and contaminants to freshwaters. Being anadromous and semelparous (die after spawning), salmon deposit their eggs and carcasses thereby allowing use by stream consumers (13). While marine derived nutrients from spawning salmon enhance biodiversity and stream productivity, there is evidence of greater contaminant burdens in various riparian biota and predators associated with spawning salmon (5, 14-16). In other environments, acidification can affect food web structure with consequences for avian population densities, diet, breeding performance, and contaminant exposure (17). The mechanism is largely through reduced macro-invertebrate prey abundance and an absence of key taxa required for reproduction (18, 19), but also increased bioavailability and toxicity of metals such as aluminum, cadmium, mercury, and leadsparticularly when calcium intake is depressed (17, 20). Since prey biomass and composition shift strongly in response to acidification, usually with a loss of items rich in calcium, changes in predator diet and exposure to contaminants are of potential concern. Dippers (Cinclidae) are passerines adapted to fast-flowing rivers and streams on five continents. They are important biomonitors for freshwater pollution, since they are strictly aquatic, occupy small territories year-round, and are sensitive to changes in environmental quality (21, 22). With clutches of four to five eggs, dippers have been successfully used to represent contaminant conditions at the breeding site (5, 23). In a companion study, we confirmed female dippers employed an income breeding strategy, shifting their diet during egg production, which directly affected the egg composition both in terms of nutrients and contaminants (10). We hypothesized that varying resource availability in food webs of contrasting character affects avian foraging strategies during egg-laying and ultimately egg contaminant composition. Here, we compared foraging and egg contaminants in two species: the Eurasian dipper (Cinclus cinclus) on rural upland Welsh streams with varying acid-base status and the American dipper (Cinclus mexicanus) on remote western Canadian (British Columbia) rivers contrasting in the presence or absence of migratory salmonids. These closely related species have similar ecology, but the contrasting nature of their specific habitats, food availability, and contaminants in these different geographical locations provides a unique opportunity to assess the importance of local environmental conditions for trophic and contaminant pathways in birds. Uniquely, the study design also permitted a general assessment of levels of persistent organic pollutants in river birds across continents.
Experimental Section American Dipper Sampling. In 2006, we selected four catchments in southwestern British Columbia, Canada based on the presence or absence of anadromous Pacific salmon 10.1021/es903084m
2010 American Chemical Society
Published on Web 02/04/2010
and steelhead (Oncorhynchus spp.) as vectors of marinederived nutrients and contaminants to rivers. Two rivers, the Chilliwack (49° 5′ N, 121° 49′ W) and the Silverhope (49° 15′ N, 121° 23′ W) have up to six species of anadromous fish and annual returns in the range of 130 spawners/km (7500 fish) (Silverhope) to 9600 spawners/km (540,000 fish) (Chilliwack) (Supporting Information (SI) Table SI 1). In contrast, the Upper Skagit (49°10′ N, 121°04′ W) and Similkameen rivers (49° 22′ N, 120°30′ W) have been devoid of anadromous fish for decades because of dams or changes in land use which prevent salmon migration. The neighboring river catchments share similar bedrock geology of Cenozoic sedimentary and volcanic rock (BC Geological Survey). Those without salmon historically supported spawning runs and were otherwise similar to the salmon bearing sites. We identified breeding territories of American dippers by following pairs through nest-building until incubation. A single egg from each clutch was taken randomly after incubation had begun: Chilliwack n ) 6, Silverhope n ) 5, Skagit n ) 3, Similkameen n ) 3. Whole eggs were weighed, measured (length and width) and frozen. Eggshells were later removed and the egg contents were transferred into glass jars chemically cleaned with acetone and hexane, then stored at -40 °C prior to analysis. Eurasian Dipper Sampling. In 2007 and 2008, Eurasian dipper nests were located on tributaries throughout the upper reaches of the dominantly rural catchments of the rivers Wye (52°01′ N, 03°17′ W) and Tywi (51° 55′ N, 03° 54′ W) in Wales, United Kingdom. The tributaries within these catchments vary in water chemistry due to the geology of the bedrock and soils from which the rivers drain thereby altering macroinvertebrate communities (24, 25). We measured pH and hardness (mg CaCO3/L) at each dipper nest site during summer base flow (maximum pH/hardness) and winter high flow (minimum pH/hardness) before averaging the measurements. Average total hardness and pH were positively correlated (r ) 0.93, p < 0.0001) but hardness was used as a proxy for categorizing streams as circumneutral (range 43.3-101.4 mg CaCO3/L), intermediate (range 11.4-39.8 mg CaCO3/L) or mineral-poor (range 5.8-10 mg CaCO3/L) based on data from Edwards (26). We monitored Eurasian dipper nests (n ) 37; 14 circumneutral, 12 intermediate, 11 mineral-poor) along 22 streams. A single egg was collected randomly from each clutch during the first week of incubation. Faecal samples to identify diet composition were also taken from each Eurasian dipper territory during egg-laying. This involved scraping three to six fresh dipper faecal pellets from exposed rocks or sediments within the territory and placing them into a polyethylene vial filled with 70% ethanol. Samples were later deflocculated using 0.1 M NaOH solution for 24 h, filtered and resuspended in ethanol. Invertebrate prey remains were counted at family level and fish at species level using established methods under 12-56× stereomicroscope (27, 28). Prey Sampling. For both locations, stream invertebrates were collected using a standard 3 min kick sample (29) spread across riffle and margin habitats in each dipper territory during breeding in late April or early May. Invertebrates were live-sorted to order in the field and transported in sealed polyethylene bags filled with streamwater. Samples were later rinsed with distilled water, drained, and frozen at -20 °C. Invertebrate families from the orders Ephemeroptera, Plecoptera, Trichoptera, and Diptera were obtained in all study catchments in British Columbia and Wales. Invertebrates were identified to family for information about the community composition but pooled and analyzed by order. At the British Columbia sites, we also collected juvenile salmonids including fry and parr of coho salmon (Onchorhynchus kisutch), steelhead and rainbow trout (Onchorhynchus mykiss), and Dolly Varden trout (Salvelinus malma)
using dipnets or with “Gee” minnow traps baited with salmon roe. Gee traps were left for several hours or overnight and fish
