The Effect of Environmental Variation at Differing Spatial and Temporal Scales on Stock Assessment Inputs and Outputs with a Focus on King Mackerel in the U.S. Gulf of Mexico

The sustainable management of living marine resources requires a basic understanding of the populations’ dynamics and its interaction with the surrounding ecosystem. Many organizations advocate for research toward Ecosystem Based Fisheries Management (EBFM) and at the base of the hierarchical approach to EBFM is the Ecosystem Approach to Fisheries Management (EAFM) that focuses on how variability in habitat and climate affect the dynamics of a single species. The goal of this dissertation was to enhance the EAFM by providing examples of how specific aspects of the ecosystem such as climate and habitat can be evaluated and included into the stock assessment process to support EAFM. I focused on the impact of defined spatial and temporal scales of the population dynamics – ecosystem interaction (i.e., fish-environment, Chapter Two) and the inclusion of ecosystem information into stock assessment supporting analyses such as indices of abundance and density distribution mapping (Chapter Three), as well as directly into integrated stock assessment models (Chapter Four). To demonstrate the approaches taken in this dissertation the empirical analysis focused on ecosystem components related to the larval stage of king mackerel (Scomberomorus cavalla) in the U.S. Gulf of Mexico, while simulation testing was completed using a simplified hypothetical species. Dissertation results indicated that environmental covariates will have varying impacts on stock dynamics at different spatiotemporal scales. For instance, for king mackerel, wind speed at different spatiotemporal scales led to different responses in perceived larval abundance. Therefore, the use of large scale annually and regionally averaged environmental covariates would distort the perception of the population’s dynamics over time and space. This conclusion was further supported by the work of Chapter Three where geostatistical spatiotemporal models were explored using differing spatiotemporal scales of sea surface temperature to estimate density in the GOM. ...