Evaluation of Stock–Recruitment Curves for White Shrimp in Georgia

Abstract

Georgia's food shrimp fishery is currently managed on the premise that a relationship exists between the abundance of spring-spawning white shrimp Litopenaeus setiferus and the abundance of fall recruits. Spawners are caught primarily during May and June, and fall recruits are caught from August through January. The efficacy of this management approach is unknown because the relationship between the two segments of the shrimp fishery has never been evaluated. We applied Beverton–Holt and Ricker spawner–recruit curves to fishery-independent and fishery-dependent indices to determine whether spawner abundance could be used to reliably predict the abundance of fall recruits. The commercial landings (lb) from May and June combined were used as the fishery-dependent spawner index, and monthly catch-per-unit-effort (CPUE) values observed during April, May, and June were used as fishery-independent spawner indices. Potential spawner– recruit models were evaluated for each set of indices and the resulting mean square error (MSE) values were compared. The best-fitting model was a Beverton–Holt curve that incorporated June CPUE as the spawner index. Unfortunately, this model was not very predictive. Multiple-regression techniques also were used to select environmental variables that enhanced the fit of the spawner– recruit model. Three two-variable models were evaluated for predictive fit, and the best model (r ² = 0.447) included September salinity and October water temperature. Our results suggest that the abundance of white shrimp spawners is a poor predictor of the abundance of fall recruits. Inclusion of environmental factors, such as seasonal salinity and water temperature, can improve our ability to predict recruit abundance.