Methods for refining waterbird colony persistence research and their application to Great Lakes colonial waterbirds.

University of Minnesota M.S. thesis. May 2012. Major: Conservation biology. Advisor: Dr. Francie Cuthbert. 1 computer file (PDF); x, 78 pages, appendices 1-3.After decades of research on waterbird colony persistence, or the re-occupation of a colony site in successive breeding seasons, many environmental correlates of colony persistence have been identified. However, little is understood about what truly drives waterbird colony persistence at a given site; without identifying causative factors, a full understanding of the form and function of colony site use patterns is impossible. To improve understanding of the biology of colonial waterbirds as well as conservation and management efficiency, greater research attention should be directed towards multivariate analyses of predictors of colony persistence and towards experimentation to test theoretical predictions about mechanisms determining colony site use. The former type of research allows scientists to identify the most important variables among a suite of potential predictors, while the latter is the only legitimate way to determine true causative relationships. A series of decadal surveys of Great Lakes waterbird colonies provides an opportunity to illustrate the method of multivariate analysis to inform biological and conservation understanding. I used a hierarchical occupancy model analyzed within a Bayesian statistical framework to determine the most important predictors of site colonization and colony persistence probabilities for the ten most common species in the U.S. Great Lakes colonial waterbird community. Species varied considerably in their responses to site location, human disturbance, and species richness at colony sites. Sites with large colonies and those not susceptible to flooding were most likely to be reoccupied in the following census period, suggesting that sites with these characteristics should be conservation priorities. However, the most effective conservation strategy will take advantage of knowledge about species-level variation to tailor conservation and management activities to the particular species of interest