Stopover duration of fall‐migrating dabbling ducks

Abstract

The amount of time migrating birds spend at stopover sites, or stopover duration, partially determines an individual's access to resources, the environmental conditions encountered, and the exposure to predation, which in turn affect survival and fecundity. As such, migratory behaviors such as stopover duration can have a considerable effect on populations of migrants and plans for their conservation. This is especially true for migrant waterfowl, which are explicitly conserved through Joint Venture (JV) partnerships under the North American Waterfowl Management Plan. Although waterfowl are one of the most heavily studied taxa, little is known about their stopover behavior due to the scope of migration. Consequently, conservation plans of many mid-migration JVs either omit estimates of stopover duration or rely on antiquated data to estimate energetic requirements. We used weather surveillance radar to identify and enumerate ducks emigrating from an important stopover area in central Illinois. By using radar data in combination with data from weekly aerial inventories, we estimated an average stopover duration for fall-migrating dabbling ducks (tribe Anatini) of 28 days (SD ¼ 12) over 8 years (). Our estimate was similar to the historical estimate of 28 days , which serves as the primary reference for the Upper Mississippi River Great Lakes Region JV conservation plan. In addition to a corroborative mean, we also found considerable inter-annual variation in stopover duration. Estimated annual stopover duration was correlated positively with an index of annual foraging habitat quality (Spearman's rank correlation; r s ¼ 0.83), suggesting ducks may have assessed local conditions and adjusted the spatiotemporal course of fall migration. If the stopover behavior of fall-migrating ducks is flexible and forage-dependent, it is possible ducks allocate their time among sites in a somewhat ideal and optimal fashion, which could substantially affect the way resources are allocated within the spatial context of a JV region. ß 2012 The Wildlife Society.

Mallard survival, movements, and habitat use during autumn in Illinois

“…However, we documented relatively low use (18.7%, diurnal; 18.1%, nocturnal) of designated refuges by mallards in our study area. Although we did not measure food density on refuges in our study area, data from aerial waterfowl inventories and our observations suggested that foraging habitat was generally poor on these areas (O'Neal et al , Hagy et al ). Flooding from the Illinois River in August and thereafter killed or prevented maturation of moist‐soil plants or affected infrastructure used for managing hydrology and moist‐soil vegetation.…”

Quantitative and Qualitative Approaches to Identifying Migration Chronology in a Continental Migrant

“…Numerous techniques have been used to quantify the timing and duration of migration in waterfowl, including radar [19] and radio-telemetry [20]. Although the study of migration chronology has been historically limited spatially and temporally, new technologies and methods have enabled researchers to develop quantitative techniques to define migration at the individual level [26], [27], [43].…”

“…Waterfowl migration has been examined using a variety of methods, including counts [18], radar [19], and very high frequency (VHF) radio-telemetry [20]. Despite these efforts to document migration chronology, past studies have been limited spatially (counts, VHF radio-telemetry), temporally (radio-telemetry, banding), or taxonomically (radar) by available technology.…”

How will predicted land‐use change affect waterfowl spring stopover ecology? Inferences from an individual‐based model

“…), whereas another study in North America estimated a mean stopover duration of 12 days (Krementz, Asante & Naylor ). Finally, a study that examined stopover duration of dabbling ducks using weather radar in Illinois, USA, documented a mean value of 28 days (O'Neal, Stafford & Larkin ). Mean stopover duration for present conditions in our model was 19·53 days, which is within the range of values observed in these empirical studies.…”