Dynamique spatio-temporelle des populations de canards barboteurs et de leur habitat
|Advisor:||Cumming, Steven G.|
|Abstract:||The main objective of my thesis was to quantify the spatial variation in duck populations in North America and their main breeding habitat and assess the importance of this variability. I first present a study case on the Mallard (Anas platyrhynchos) that aim at assessing at how using informative priors on population parameters effects the conclusions drawn from a spatially explicit Gompertz population model. I compared the results from a naïve model and from a model where I constrained the intrinsic growth rate (r) to biologically realistic values. The naïve model lead to the estimation of unrealistic growth rate and shorter return time to equlibirum than those estimated by the informed model. The effects of the extrinsic factors were however comparable across both model. I subsequently used a spatially-varying coefficients model to assess the spatial variation in the ecological drivers of wetlands abundance in the Prairies pothole region (PPR). Overall, fall and spring precipitation were the most important climatic drivers in ponds abundance in the west while winter and summer precipitation and the most important driver in the east. Based on my previous results, I developed a multivariate extension of the Gompertz population model to assess the synchrony and the spatial variation in populations dynamics of four species of dabbling duck: the Mallard; Northern Pintail (Anas acuta); American Wigeon (Anas americana); and Green-winged Teal (Anas crecca). Northern Pintails and to a lesser extent Mallards showed a pattern of negative correlations among populations in the PPR and populations in the Western boreal forest of northern Canada and Alaska supporting the contention that some individual will “overfly” the PPR during drought years. Finally, I assessed the spatial and temporal variability in harvest rate of American Black Duck (Anas rubripes) on their Canadian breeding ground with direct recoveries from banding data. Juveniles recovery probabilities were correlated with the hunter effort but did not decreased after the implementation of new, stricter, harvest regulations in the early 1980’s. Harvest rate along the Saint Lawrence River system and in the Atlantic Provinces remain particularly high and should be closely monitored to avoid overexploitation.|
|Document Type:||Thèse de doctorat|
|Open Access Date:||23 April 2018|
|Collection:||Thèses et mémoires|
All documents in CorpusUL are protected by Copyright Act of Canada.