Northern Prairie Wildlife Research Center
1. Has Reproduction or Survival of Scaup Changed Sufficiently to Cause Population Declines, and if so, what is the cause(s)?
Basic rates associated with population dynamics -- births, deaths, immigration, and emigration -- were briefly reviewed. Factors affecting birth rates include number of eggs produced, egg fertility and hatchability, nest success, renesting rates, and duckling survival. Birth rate may be affected by population age and sex ratios, body size and condition, timing of breeding, renesting effort, and environmental variables, both biotic (e.g., food, predators) and abiotic (e.g., weather conditions). Mortality differs among age and sex cohorts, during different components of the annual cycle, and may be attributable to different annually varying factors such as hunting, disease or catastrophic events. Fidelity of yearling and adult female lesser scaup to a breeding location is strong (see Breeding ecology: annual survival and philopatry above). Little is known about reproductive or survival rates of lesser scaup except for those breeding in a few prairie-parkland and boreal forest areas. Thus, there are several hypotheses to account for population declines based on changes in vital rates. For clarity, however, we grouped them into breeding and nonbreeding hypotheses.
Breeding Season Hypotheses
Productivity may decline because of changes to boreal forest habitats, contaminants, or nutrition. Potential mechanisms are manifold, but most likely are lowered duckling survival, reduced nest success, or both, assuming that previous models and observations of duck population dynamics apply (Johnson et al. 1987, Carlson et al. 1993, Flint et al. 1998). It is unknown whether relatively small changes in survival or nest success, acting singly or in combination, would cause population declines observed in BGS data; these analyses have not been performed. Female condition and clutch size also may be important, so contaminant or nutrient acquisition hypotheses need to be considered. Alternatively, survival of breeding females may have declined. Modification of boreal forest habitat may have altered predator-prey relationships or increased natural mortality in other ways. We think it unlikely that subsistence hunting of scaup has increased substantially.
Rather than separate reproduction or survival into various components, a preferred approach is to study individually marked female scaup so that reproduction and survival can be determined concurrently. Capturing and marking females before breeding would allow assessment of the contaminant hypotheses because their contaminant levels could be measured or randomly-selected females could be dosed with contaminants suspected to impair reproductive success. Moreover, marked birds can be resighted in future years, providing further data about survival and breeding site fidelity. Below is a phased approach for studying reproduction and survival of scaup on selected breeding areas:
Phase 1: Pinpoint areas where scaup populations have declined most, where they have remained stable, and where they are increasing.
Phase 2: Conduct preliminary study(ies) to determine feasibility (logistics, costs, etc.) of acquiring data from individually marked (breeding success, survival, diet [using stable isotopes of blood], molting) or unmarked (nutrient acquisition, diet, molt) scaup.
Phase 3: Develop a preliminary model of population dynamics using published data and information acquired in phase 2.
Phase 4: Randomly select study sites based on information obtained in phases 1 and 2, involving replicate sites in each trend category, or as a minimum, contrasting areas of decline with those that are stable or increasing.
Phase 5: Obtain data for 4-5 years on each site. Each year, (1) predict and then measure population change, then (2) refine and update the population dynamics model, while (3) testing hypotheses about underlying causes of population change.
Nonbreeding Season Hypotheses
Studies of survival, diet, and nutrient or contaminant acquisition on nonbreeding areas must be done concurrently with breeding studies to provide better data about causes of scaup population declines. It is crucial to simultaneously evaluate whether nonbreeding season mortality has risen and, if it is due to hunting or another cause. It also is essential for correctly interpreting results of breeding season studies (e.g., roles of nutrients or contaminants).
Changes to molting, migration/staging, or wintering areas may affect survival or future reproduction. Likewise, impacts of hunting or natural mortality may influence population size. The role of hunting in adult female survival is uncertain but may have increased in relative importance because harvest of adult scaup (including females) has remained stable in the Mississippi flyway (where most scaup are shot) despite population declines and a lower ratio of immatures:adults in the harvest. Finally, strong fidelity of breeding scaup populations to the same migration and wintering areas could cause declines on breeding grounds if mortality or contamination of food resources on migration or wintering areas has risen.
Investigations are needed at important migration or wintering sites. Specific research questions about the nonbreeding period are examined below, followed by a general approach to address them. Contaminant problems are covered elsewhere (see Issue 3), but there is potential for collaborative research.
How does annual and overwinter survival of scaup vary with age-sex cohort, and what role does hunting play? Banding scaup in boreal forest locations and wintering areas would eventually answer this question. There are many benefits to this approach, but there will be costs and delays in acquiring data, and it would not yield information on natural causes of mortality. One alternative is to mark large numbers of scaup on migration and/or wintering areas with satellite transmitters and determine survival rates and causes of mortality; this, too, is costly but would yield precise results quickly. Another favored possibility is to combine the 2 approaches, using radio-telemetry to rapidly provide reliable information about smaller samples of scaup in specific areas and using band recovery analyses to derive estimates from larger samples marked over a broader geographic area. Preliminary analyses of harvest data (Afton and Anderson, in review) indicate that studies of adult females would be most valuable. Feasibility of using satellite transmitters should be pursued, as they would provide more complete information about molting, staging and wintering locations, local movements and mortality.
What is the fidelity of breeding scaup to specific migration sites and wintering areas, and how does this relate to overwinter mortality, including hunting losses? Declines of scaup in mid-boreal areas of western Canada may be related to fidelity to winter or migration sites if mortality on wintering areas exceeds productivity (and adult survival) on breeding areas. Answering philopatry and mortality questions may be accomplished by annually banding scaup in the same breeding and wintering areas and recapturing birds over time. The advantages are that other information can be acquired from captured birds, other species could be banded, and exact locations of marking are known, but disadvantages include cost and long delays before sufficient data are available for analyses. Previous banding data could provide preliminary information on winter philopatry.
Alternatively, philopatry could be evaluated by applying stable-isotope techniques to feathers of hunter-shot or trapped scaup, thereby delineating affinities of breeding scaup (from the deuterium isotope signal) to specific migration or wintering areas (from where the bird was shot/captured). This approach can quickly provide answers to questions about (1) general breeding origin (primarily latitude) of hatch-year scaup and possibly about female fidelity to specific migration and wintering sites, (2) general molting areas (latitude) of adult male scaup, and (3) breeding philopatry of scaup shot by hunters in different migration and wintering areas. The main advantage of this approach is that answers would be obtained relatively quickly for certain cohorts (e.g., breeding origin of hatch-year scaup; molting location of adult males). Disadvantages include cost (initial refinement of the existing isotope model and development of cheaper analytical methods), and lack of specific information about breeding origins. However, the information obtained on breeding origins likely would be sufficiently precise to test the main question posed above.