1. Stable hydrogen isotope (dD) methods for tracking animal movement are widely used yet often produce low resolution assignments. Incorporating prior knowledge of abundance, distribution or movement patterns can ameliorate this limitation, but data are lacking for most species. We demonstrate how observations reported by citizen scientists can be used to develop robust estimates of species distributions and to constrain dD assignments. 2. We developed a Bayesian framework to refine isotopic estimates of migrant animal origins conditional on species distribution models constructed from citizen scientist observations. To illustrate this approach, we analysed the migratory connectivity of the Virginia rail Rallus limicola, a secretive and declining migratory game bird in North America. 3. Citizen science observations enabled both estimation of sampling bias and construction of bias-corrected species distribution models. Conditioning dD assignments on these species distribution models yielded comparably high-resolution assignments. 4. Most Virginia rails wintering across five Gulf Coast sites spent the previous summer near the Great Lakes, although a considerable minority originated from the Chesapeake Bay watershed or Prairie Pothole region of North Dakota. Conversely, the majority of migrating Virginia rails from a site in the Great Lakes most likely spent the previous winter on the Gulf Coast between Texas and Louisiana. 5. Synthesis and applications. In this analysis, Virginia rail migratory connectivity does not fully correspond to the administrative flyways used to manage migratory birds. This example demonstrates that with the increasing availability of citizen science data to create species distribution models, our framework can produce high-resolution estimates of migratory connectivity for many animals, including cryptic species. Empirical evidence of links between seasonal habitats will help enable effective habitat management, hunting quotas and population monitoring and also highlight critical knowledge gaps.

Chlorinated hydrocarbon concentrations in eggs of fish-eating birds from contaminated environments such as the Great Lakes of North America tend to be highly intercorrelated, making it difficult to elucidate mechanisms causing reproductive impairment, and to ascribe cause to specific chemicals. An information- theoretic approach was used on data from 197 salvaged bald eagle (Haliaeetus leucocephalus) eggs (159 clutches) that failed to hatch in Michigan and Ohio, USA (1986-2000). Contaminant levels declined over time while eggshell thickness increased, and by 2000 was at pre-1946 levels. The number of occupied territories and productivity increased during 1981 to 2004. For both the entire dataset and a subset of nests along the Great Lakes shoreline, polychlorinated biphenyls (SigmaPCBs, fresh wet wt) were generally included in the most parsimonious models (lowest-Akaike's information criterion [AICs]) describing productivity, with significant declines in productivity observed above 26 microg/g SigmaPCBs (fresh wet wt). Of 73 eggs with a visible embryo, eight (11%) were abnormal, including three with skewed bills, but they were not associated with known teratogens, including SigmaPCBs. Eggs with visible embryos had greater concentrations of all measured contaminants than eggs without visible embryos; the most parsimonious models describing the presence of visible embryos incorporated dieldrin equivalents and dichlorodiphenyldichloroethylene (DDE). There were significant negative correlations between eggshell thickness and all contaminants, with SigmaPCBs included in the most parsimonious models. There were, however, no relationships between productivity and eggshell thickness or Ratcliffe's index. The SigmaPCBs and DDE were negatively associated with nest success of bald eagles in the Great Lakes watersheds, but the mechanism does not appear to be via shell quality effects, at least at current contaminant levels, while it is not clear what other mechanisms were involved.

The relationship between regional reproduction rates of bald eagles (Haliaeetus leucocephalus) and concentrations of p.p'-dichlorodiphenyldichloroethylene (p,p'-DDE) and total polychlorinated biphenyls (PCBs) in blood plasma from nestling bald eagles was assessed. Blood was analyzed from 309 nestlings from 10 subpopulations of eagles across the Great Lakes region. Geometric mean concentrations of p,p'-DDE and total PCBs were inversely correlated to the productivity and success rates of nesting bald eagles within nine subpopulations. Nestlings eight weeks of age and older had significantly greater geometric mean concentrations of total PCBs and p,p'-DDE than nestlings less than eight weeks of age. The ability to use measurements of p,p'-DDE and total PCBs in nestling blood to determine the potential impact of these contaminants on adult nesting on a regional scale was demonstrated.

In 2000, a pair of bald eagles (Haliaeetus leucocephalus) nested successfully along the shorelines of Lake Ontario in North America for the first time since 1957. However, it is a continuing question whether bald eagles will be able to reproduce successfully as they return to nest on Lake Ontario. Great black-backed gulls (Larus marinus) and herring gulls (L. argentatus) were selected as surrogate species to predict contaminant levels in eggs of bald eagles nesting on Lake Ontario. Because of the suspected overlap in the diets of great black-backed gulls and bald eagles (i.e., fish, gull chicks, and waterfowl), the two species probably occupy a similar trophic level in the Lake Ontario food web and, thus, may have similar contaminant levels. Fresh great black-backed gull and herring gull eggs were collected from three study sites in eastern Lake Ontario in 1993 and 1994 and analyzed for contaminants. Average contaminant levels of dichlorodiphenyldichloroethane (p,p'-DDE), total polychlorinated biphenyls (PCBs), and dieldrin in great black-backed gull eggs were 12.85, 26.27, and 0.27 microg/g, respectively. The mean ratio of contaminant levels in great black-backed gull eggs to contaminant levels in herring gull eggs for these three contaminants was 2.09 (range of means, 1.73-2.38). Predicted levels of contaminants in bald eagle eggs in Lake Ontario would be expected to be similar to the mean levels reported for great black-backed gull eggs. As a comparison, contaminant levels in bald eagle eggs collected from other Great Lakes nesting sites were compared to mean levels reported for herring gull eggs collected from nearby sites in 1986 to 1995. The mean ratio of contaminant levels in bald eagle eggs to contaminant levels in herring gull eggs from these sites for DDE, total PCBs, and dieldrin was 2.40 (range of means, 1.73-3.28). These ratios are very similar to those reported using great black-backed gull eggs, illustrating the apparent similarity in trophic status shared by the two top predator species at these Great Lakes sites. Predicted levels of contaminants in bald eagle eggs at Lake Ontario are similar to levels reported for bald eagles breeding at other Great Lakes sites, suggesting that bald eagles may be able to breed on the shores of Lake Ontario. However, it is unclear at this time what level of breeding success should be expected, given that productivity at other similarly contaminated Great Lakes sites may be below that required to sustain a successful breeding population. The absence of an inland bald eagle population from which bald eagles may begin to colonize the shorelines of Lake Ontario may be delaying initiation of nesting site selection; other factors such as habitat and prey availability would likely not limit reproductive success.

Canada goose (Branta canadensis) harvest management depends on reliable estimates of harvest composition, and established genetic methods provide an alternative to traditional methods. We expanded upon previous genetic studies by comparing the utility of 6 nuclear microsatellite loci and mitochondrial (mtDNA) control region sequences for discriminating among giant (B. c. maxima) and interior (B. c. interior) populations in Ohio (USA) Canada goose harvests at both individual and population levels. Subspecies and populations exhibited greater differentiation in mtDNA (FST = 0.202) than microsatellites (FST = 0.021), as would be expected based on differences in effective population size. Neither microsatellites nor mtDNA alone were sufficient for estimating harvest composition at the subspecies or population level in simulations and empirical blind tests using individuals of known origin; however, a combined microsatellite + mtDNA dataset yielded accurate and precise harvest derivations at the subspecies level. Both population‐level mixed stock analysis and individual‐level assignment tests provided accurate results, but a large proportion of birds could not be assigned with confidence at the individual level. We applied mixed stock analysis and the combined microsatellite + mtDNA dataset to Ohio's 2003–2004 harvest and found that interior populations accounted for 4.9% (95% CI = 1.7–8.0%) of the statewide early season and 9.3% (95% CI = 6.9–11.6%) of the regular and late‐season harvested sample. These results suggest that maximum likelihood harvest derivations are highly dependent on the choice of genetic markers. Studies should only employ markers that exhibit sufficient variation and have been shown through simulations and empirical testing to accurately discriminate among the subspecies or management populations of interest.