James Thorson: About me


Advisor: Jim Berkson School: Virginia Tech, Department of Fisheries and Wildlife Sciences Supervisory Committee: Brian Murphy, Don Orth, Clay Porch Title: Can Catchability be Multi-Species? Exploring Commonalities in Catchability Changes Between Similar Species and Fishing Gears in the Gulf of Mexico Summary Stock assessment is important to a variety of fisheries management objectives. Many stock assessments use catch-at-age analyses such as virtual population analysis (VPA) or statistical catch-at-age (SCA) models, which often incorporate catch-per-unit-effort (CPUE) as an index of stock abundance. These CPUE time-series data are generally fitted by estimating a constant catchability coefficient, assumed to scale CPUE to abundance. Although the catchability coefficient incorporates many non-stationary effects including anthropogenic, biological, environmental, and management factors, technological changes are especially likely to cause increasing trends in catchability. When assessment models mistakely utilize a constant catchability to model an increasing trend, the trend in catchability may mask decreases in stock abundance. Ensuing "hyperstable" estimates lead to overly optimistic management policy. My project will investigate the efficacy of estimating changes in catchability by pooling information across related gears and species of reef fish in the Gulf of Mexico. It will consist of both simulation modeling and statistical analysis components. Simulation modeling will be used to construct biologically plausible catch-at-age and CPUE data across a range of reasonable scenarios. These data will be used to (1) estimate the distribution of catchability estimates under competing scenarios, and (2) quantify the ability of analyses to detect catchability changes given these same model scenarios. Statistical analysis will also be applied to real-world catch-at-age and CPUE data from selected reef fish in the Gulf, both to determine: (1) what degree of catchability change is supported by fishery-dependent data for these species, and (2) if analysis of several species at a time improves estimation of catchability over assuming constant catchability.

Advisor: Jim Berkson

School: Virginia Tech, Department of Fisheries and Wildlife Sciences

Supervisory Committee: Brian Murphy, Don Orth, Clay Porch

Title: Can Catchability be Multi-Species? Exploring Commonalities in Catchability Changes Between Similar Species and Fishing Gears in the Gulf of Mexico

Summary

Stock assessment is important to a variety of fisheries management objectives. Many stock assessments use catch-at-age analyses such as virtual population analysis (VPA) or statistical catch-at-age (SCA) models, which often incorporate catch-per-unit-effort (CPUE) as an index of stock abundance. These CPUE time-series data are generally fitted by estimating a constant catchability coefficient, assumed to scale CPUE to abundance. Although the catchability coefficient incorporates many non-stationary effects including anthropogenic, biological, environmental, and management factors, technological changes are especially likely to cause increasing trends in catchability. When assessment models mistakely utilize a constant catchability to model an increasing trend, the trend in catchability may mask decreases in stock abundance. Ensuing "hyperstable" estimates lead to overly optimistic management policy.

My project will investigate the efficacy of estimating changes in catchability by pooling information across related gears and species of reef fish in the Gulf of Mexico. It will consist of both simulation modeling and statistical analysis components. Simulation modeling will be used to construct biologically plausible catch-at-age and CPUE data across a range of reasonable scenarios. These data will be used to (1) estimate the distribution of catchability estimates under competing scenarios, and (2) quantify the ability of analyses to detect catchability changes given these same model scenarios. Statistical analysis will also be applied to real-world catch-at-age and CPUE data from selected reef fish in the Gulf, both to determine: (1) what degree of catchability change is supported by fishery-dependent data for these species, and (2) if analysis of several species at a time improves estimation of catchability over assuming constant catchability.