Ecologists and conservation biologists are interested in explaining why animal abundance and reproductive success vary among habitats. Initial motivation for this research arose from concerns for Northern Pintail (Anas acuta) populations in North America. Unlike many prairie-nesting dabbling duck populations, pintails failed to increase during periods of excellent wetland conditions, and remained below conservation goals. Low pintail populations have been linked to degraded landscape conditions on the Canadian prairies. Current habitat management for pintails aims to protect and create larger areas of perennial cover either by encouraging better management of grazing lands, by converting cropland to grassland, or by promoting adoption of fall-seeded crops like winter wheat. The central premise is that larger areas of natural grassland cover will attract breeding pintails to nest earlier in the season in low-predation-risk habitat. I studied pintail nesting ecology near the Milk River Ridge, Alberta, 2004-2006, in terms of a life-cycle perspective, from spring arrival and settling on breeding areas, to assess age and quality of nesting females, to determine timing and investment in reproduction, and finally to measure nesting success. A gradient in presumed high (grassland) to low (agriculture) habitat quality provided a landscape template for testing habitat selection models. Pintail breeding pair densities were 1.5-3 times higher in grassland than agricultural landscapes in all three years, regardless of regional population size, with pairs occupying grassland landscapes at higher densities immediately upon arrival in early spring. Northern Shoveler (A. clypeata), gadwall (A. strepera) and blue-wing teal (A. discors) had similar settlement patterns as pintails, but mallard (A. platyrhynchos) pair density was higher in agricultural areas. Relatively more, older female pintails were captured at nests in grassland landscapes whereas yearling females were encountered more often in agricultural areas, a pattern that was not detected in female shovelers. This response suggests that older female pintails may be better able to recognize and settle in higher quality grassland habitats. Body mass of pintail females did not vary among years, decreased seasonally, and was positively related to body size index and incubation stage. Furthermore, pintail body mass did not differ between grassland (650 ± 24 g), ecotone (678 ± 27 g) and agriculture (672 ± 33 g). In female shovelers, body mass varied among years (555 ± 29 g in 2004, 481 ± 18 g in 2005, 508 ± 21 g in 2006), and increased with nesting date. Shoveler body mass did not differ between grassland (519 ± 32 g), ecotone (519 ± 44 g), or agriculture (507 ± 35 g). Nest initiation dates did not vary by landscape for pintail, shoveler or mallard, but all species nested earlier in 2006 versus 2004. In pintail, shoveler and mallard, clutch size was negatively related to nest initiation date. Pintail and shoveler clutch sizes were generally larger in a wet year with abundant wetlands (2006) when compared with a dry year (2004), but no landscape differences were detected. Mallard clutch size did not vary by year or landscape. Female reproductive timing and investment (in terms of clutch size) were unrelated to upland habitat characteristics, counter to a hypothesis that predicts larger pintail clutch sizes in agricultural landscapes. However, pintail and shoveler invested in larger clutches in 2006, a wet year with abundant wetlands, possibly due to greater abundance of aquatic foods. Finally, nest survival rates of duck species, except mallard, tended to be higher in grassland landscapes and lower in agricultural landscapes. Pintail nest survival was consistently higher in grassland than in agricultural landscapes and was highest in 2006 when wetland conditions were excellent. Shoveler and blue-winged teal nest survival rates did not vary strongly with landscape, but were also higher in 2006, whereas mallard and gadwall nest survival estimates did not vary with landscape or year. Overall, pintails settled at higher densities in grassland landscapes where breeding success was higher (indexed by nesting success). This suggests that pintails respond appropriately to cues that enable them to recognize suitable habitat, at least in regions where large contiguous areas of grassland habitat remain. Furthermore, assuming that findings for pintails reflect those of other grassland bird species, large remnant areas of intact natural grassland seem particularly in need of protection or restoration, and management regimes that maintain their habitat integrity. By integrating applied and theoretical aspects of pintail reproductive ecology, I attempted to provide deeper insights into the processes that could shape behavioral decisions by breeding pintails and other duck species. Older pintails may occupy wetlands in higher quality grassland habitat early in spring, forcing subordinate or later-arriving individuals into poorer quality habitat (i.e., where nesting success is lower); however, mechanisms involved in this putative process are unknown. Overall, results suggest that grassland restoration or enhancement (e.g., managing grazing intensity) could improve reproductive success of pintails and possibly other grassland bird species.

According to theory, habitat selection by organisms should reflect the associated probability of survival or reproductive success. Understanding habitat selection, at multiple scales, is of interest not only from a theoretical perspective, but from an applied perspective for species conservation. Northern pintails (Anas acuta) are migratory, temperate-nesting birds that breed in greatest concentrations in the prairies of North America. Declining populations suggest that habitat loss and changing land use may have decoupled formerly reliable fitness cues from selection of suitable nest habitat. I used data from 62 waterfowl nesting study sites in prairie Canada (1997–2009), to examine whether nest survival, a primary fitness metric, at nest and habitat patch scales, was predictive of habitat selection at corresponding scales. In addition, I used systematic long-term annual pintail population monitoring data (1961–2009), and recruitment indices (juvenile:adult female ratio) from hunter harvest, to examine adaptive habitat selection among landscapes within the Prairie Pothole Region (PPR). The influences of breeding population density and landscape composition were examined at all scales. At nest and patch scales, pintail nest survival varied with nest initiation date, nest habitat, pair density, and landscape composition. Nest habitat preference reflected patterns in nest survival suggesting nest habitat preference is adaptive. Preference was generally low for habitats with low nest survival (e.g., spring-seeded cropland) and high for habitats with high nest survival (e.g., idle grassland). Differences in preference among habitats weakened at high breeding density and in landscapes with more grassland. Population-level recruitment tended to be greater when pintails settled in landscapes that were wetter than normal, contained more grassland, and were moderately variable in local elevation. Pintails were strongly associated with wetter than normal landscapes but shifted into cropland-dominated landscapes and flatter landscapes when populations were high. My results indicated that pintails express adaptive habitat associations with density-dependence acting through buffer mechanisms. Finally, I use the results of the above analyses to, 1) model and map the estimated long-term average spatial abundance of pintail pairs across the PPR as a function of landscape-level covariates, and 2) construct a deterministic model predicting pintail productivity given habitat and landscape attributes. These models allow conservation efforts to be targeted to affect the most birds, and they allow estimation of the demographic response to conservation actions.

Several factors have contributed to impeding the progress of migratory waterfowl spatial modeling, such as (1) waterfowl’s reliance on wetlands, (2) lack of understanding about shifts in distributions through time, and (3) large-scale seasonal migration. This doctoral dissertation provides an array of tools to address each of these concerns in order to better understand and conserve this group of species. The second chapter of this dissertation addresses issues of modeling species dependent on wetlands, a dynamic and often ephemeral habitat type. Correlation models of the relationships between climatic variables and species occurrence will not capture the full habitat constraints of waterfowl. This study introduces a novel data source that explicitly models the depth to water table, which is a simulated long-term measure of the point where climate and geological/topographic water fluxes balance. The inclusion of the depth to water table data contributes significantly to the ability to predict species probability of occurrence. Furthermore, this data source provides advantages over traditional proxies for wetland habitat, because it is not a static measure of wetland location, and is not biased by sampling method. Utilizing the long-term banding bird data again, the third chapter examines the behavior of waterfowl niche selection through time. By using the methods developed in chapter two, probability of occurrence models for the 1950s and the 1990s were developed. It was then possible to detect movements in geographic and environmental space, and how movements in these two spaces are related. This type of analysis provides insight into how different bird species might respond to environment changes and potentially improve climate change forecasts. The final chapter presents a new method for predicting the migratory movement of waterfowl. The method incorporates not only the environmental constraints of stopover habitat, but also includes likely distance and bearing traveled from a source point. This approach uses the USGS’ banding bird database; more specifically, it relies on banding locations, which have multiple recoveries within short time periods. Models made from these banding locations create a framework of migration movement, and allow for predictions to be made from locations where no banding/recovery data are available. / text