Identifier

etd-07032012-132140

Degree

Doctor of Philosophy (PhD)

Department

Oceanography and Coastal Sciences

Document Type

Dissertation

Abstract

The move towards an ecosystem approach to fisheries management requires baseline information on the biotic communities and an understanding of the interactions among species. The two objectives of this study were to describe the demersal fish community (DFC) associated with the northwestern Gulf shrimp trawl fishery, and to utilize a multispecies Lotka-Volterra model to examine possible community level effects of fishing. Community level effects include predator-prey interactions and the responses of fish in the same community to fishing pressure. The summer and fall Southeast Area Monitoring and Assessment Program (SEAMAP) bottomfish trawl surveys were used to identify spatial and temporal indicators in the northwestern Gulf. Cluster analysis and nonmetric multidimensional scaling analysis identified four distinct demersal fish communities (DFCs) from the summer survey data and three distinct spatial DFCs from the fall survey data. Indicator species analysis identified two different dominant communities within each DFC during the summer and fall survey periods. No statistically significant temporal trends were identifiable in the SEAMAP data over the selected time period (1986-2007). However, the community metrics from the DFCs were used to create initial parameters for a multispecies Lotka-Volterra simulation model. The Lotka-Volterra model was developed to study the implications of estimating single-species biological reference points from multispecies data and to determine the effect of fishing on other species in the same community. Three series of simulations were created to explore connectance, rebuilding plans, and multispecies fisheries in the Lotka-Volterra model. Across all simulations, the single-species FMSY rates were estimated between 10-75% of the expected values from the Lotka-Volterra model, indicating that fishing mortality rate buffers are needed to account for predator-prey interactions. No trends were detected as to the magnitude of the buffer for the species examined across all levels of community connectance. The magnitude of biomass changes in unfished species decreased as community connectance increased. The projected time for a species to rebuild was longer in the multispecies model versus the single-species model in four of five case studies. Indirect effects were important in all simulations and these results indicate a need for ecological buffers when determining single-species fishing mortality rates.

Date

2012

Document Availability at the Time of Submission

Release the entire work immediately for access worldwide.

Committee Chair

Powers, Joseph E.

Share

COinS