A life history model for assessing alternative management policies for depressed chinook salmon

Population and fishery dynamics are modelled to assess performance of alternative harvest policies for chinook that spawn in western North America. The sensitivity of model performance to variations in stock productivity and patterns of environment variation are explored to evaluate the trade-off between risk of extinction and benefits from harvest. Results indicate maximum economic benefit and low risk of extinction is not likely under the present harvest policy. Even with a conservative harvest-threshold policy, wherein harvest rates are reduced at low stock size and are zero if the stock size declines below an abundance threshold, the risk of extinction is significantly reduced compared to fixed harvest rate policies. Simulated abundance forecast errors within the historical range degraded model performance only slightly. With the constraint that risk of extinction be held below an acceptable level, socio-economic indicators reveal an optimal han;est-threshold policy at a fractional harvest of surplus abundance of about 0.45 and a threshold near 100 female spawners per stock. Based on coded-wire tag results for a major hatchery population, the mean historical harvest rate (1983-1991) was 0.6. (C) 1999 nternational Council for the Exploration of the Sea.