A stochastic simulation model of brown shrimp (Penaeus aztecus Ives) population dynamics in Galveston Bay, Texas, is described, validated, and used to evaluate the effects of management alternatives and changing environmental conditions on shrimp dynamics. The model is composed of submodels representing: (1) recruitment, (2) growth, (3) natural mortality, (4) fishing mortality, and (5) emigration ... [Show full abstract] of brown shrimp. The model predicts significant changes in total annual harvest from the food shrimp, bait, and recreational fisheries resulting from (1) closure of the bay system to all fishing except during the spring and fall open seasons, (2) two-week postponement of the opening and closing of the open seasons for the food shrimp fishery, (3) a 2.5°C increase and (4) a 2.5°C decrease in mean water temperature, (5) an 80% increase and (6) an 80% decrease in fishing effort. No significant change in the total annual harvest is predicted when the food shrimp fishing season is extended from May 15 through December 15. Sensitivity analysis suggests that field experimentation designed specifically to test the hypothesis of a 60-day time lag between brown shrimp recruitment into the bays and exposure to the fishery should receive high priority. Simulation results are discussed within a management framework.