Abstract
Stock assessment methods have been proposed for data-poor fisheries that rely mainly on cost-effective abundance-at-size data and some basic demographic knowledge on growth, maturity, and longevity. However, even these simple data requirements are often unmet. Additionally, coral reef fishes are often monitored using disparate fisheries-independent and -dependent survey methodologies that cannot easily be combined into single datasets. The goal of this dissertation was to develop a series of methods that allow for the efficient assessment of the sustainability of the coral reef fisheries of the Hawaiian Archipelago. To achieve this goal, the specific objectives of this dissertation were to: (1) develop a predictive model to generate standardization factors for species with sufficient fishery-independent survey observations and implement this statistical methodology in an automated, computer-based procedure, and; (2) develop a new approach for the generation of life history parameters that facilitates stock assessments for species with no published values. Quantitative tools were developed to standardized underwater visual survey data collected by several methodologies, produce missing life history information key to assessments, and conduct the stock assessments of 27 coral reef fish species, including 8 species for which no life history information previously existed. The methods and findings of this dissertation provides new tools for confronting data-poor situations, including, but not limited to, coral reef fisheries. The work presented here also provides the first archipelago-wide stock assessment of coral reef fishes in the tropical Pacific Ocean.