Water resource management under uncertainty

Abstract

This study investigates methods for approaching the management and conservation of freshwater mussel species in the lower Flint River Basin (LFRB), Georgia. The research focused on developing predictive models of the status, distribution, and dynamics of freshwater mussel species in relation to site-, watershed-, and species-level characteristics. The objectives were to: (1) develop methods for correcting mussel collection data for biases associated with incomplete detection and misidentification of species; (2) develop predictive models of mussel species occurrence that accounted for detection and misidentification biases; (3) develop dynamic multi-state, multi-species occupancy models to estimate metapopulation dynamic rates and improve understanding of the factors contributing to changes in the status and distribution of mussel species; and (4) synthesize knowledge gained from objectives 1-3 into a comprehensive framework useful for developing a decision tool that predicts the response of mussel populations to alternative water resource management actions. Research methods involved field data collection combined with existing mussel collection data from sites located throughout the LFRB. The results indicated generally low species detection probabilities and a substantial risk of misidentification for many LFRB mussel species. Additionally, although misidentification rates generally declined with observer experience, the risk of misidentification varied substantially among observers with identical experience. Models also suggested that mussel species presence was strongly and negatively related to the presence of impoundments and climatic drought, although the effects of drought and reach isolation were largely restricted to small- and mid-order tributaries. Models of mussel species meta-demographic rates indicated an elevated risk of local population extinction in the presence of short-term summer floods, generally low rates of colonization, and reduced local recruitment to existing populations in the presence of below-average low winter flow conditions. Results from this dissertation add to a growing body of literature regarding the influence of natural and anthropogenic factors on the status and dynamics of freshwater mussel populations and species. This knowledge could be useful in the development of management and conservation strategies aimed at balancing human and ecological water resource requirements, particularly if used in conjunction with ongoing monitoring activities under an Adaptive Resource Management framework.