Laboratory- and field-based approaches for evaluating connectivity in a dynamic coastal environment: applications for management and conservation

摘要

Connectivity of marine populations, defined as the magnitude of discrete population units interconnected through dispersal, has important implications for the ecology and management of commercially harvested species. Sustainable management requires consideration of the spatial-temporal structure of exploited populations. Connectivity measurement requires accuracy in providing relevant spatial information. My thesis bridges laboratory and field based approaches to provide integrated and reliable estimates of connectivity. Using controlled laboratory experiments, I determined that the interaction of temperature and salinity influenced composition of juvenile Atlantic cod otoliths, thus questioning whether otoliths can reconstruct environmental history when environmental variables are studied in isolation. Utilizing a field survey, I demonstrated that otolith chemistry differences could discriminate among juvenile cod from adjacent bays and coasts of origin. Assignment of residuals derived from laboratory model predictions and field observations improved discrimination, illustrating underlying fine-scale biocomplexity in otolith chemistry, and potential influence of environment on assignment at small spatial scales. These results demonstrate the utility of otolith chemistry as a tool to evaluate contributions of sub-populations to Atlantic cod stocks, and, highlight limitations imposed by environmental variation at scales less than 100 km. In a second series of experiments that focused on larval American lobster, I demonstrated that swimming ability and vertical position in the water column varied significantly among ontogenetic stages and did not did not increase linearly with development. Through a series of common garden experiments, I demonstrated biogeographic variability in swimming ability and the influence of environment. Variability in swimming apparently reflects ambient conditions of the pelagic habitat of origin. Utilizing a biophysical model incorporating observed swimming behaviours, I demonstrated that larval behaviour significantly influenced the magnitude, direction, and duration of dispersal, and that this influence varied both spatially and temporally. These results provide a biological-behavioural context to parameterize bio-physical models and an approach to improve accuracy of dispersal models and advance understanding of connectivity.udBy improving aspects of design and testing assumptions, these analyses provide a template for future use of otolith chemistry and biophysical modelling, punctuating the need for calibration and validation of the assumptions of each strategy when applied to dynamic field conditions.