ACCOUNTING FOR DETECTION HETEROGENEITY AND HOST

Cornell University 2007 During the course of my dissertation research, I made use of capture-recapture methods to investigate local house finch (Carpodacus mexicanus) demography and movements in the context of understanding seasonal Mycoplasma gallisepticum (MG) infection dynamics. Capture-recapture design, estimation, and modeling explicitly accounts for variable detectability of individuals and provides a framework for making multi-model inference, thereby incorporating inherent model selection uncertainty (via Akaike information criterion) into the inferential process. The biological focus throughout my research has generally been centered on the relationship between local spatial scale host population structure, movements, and host-pathogen dynamics. Broadly, my work illustrates the importance of accounting for animal detection probabilities when estimating epidemiological statistics and parameters. I also highlight the importance of considering different forms of animal movements (either biologically induced or as a consequence of sampling design) with respect to understanding dynamics in the finch-MG system (specifically), but also applicable t