Determining Optimal Stocking Rates Using a Stock–Recruitment Model: An Example Using Walleye in Northern Wisconsin

Abstract

We propose that stock–recruitment models can be used to estimate optimal stocking rates. Data to estimate the optimal stocking rates can be obtained in a relatively short amount of time by sampling similar populations over a few years. Whether the goal of stocking is endangered species recovery or supplementation of recreational fisheries, accurately determining the optimal stocking rate is of ecological and financial importance. As an example, we applied this approach using a Ricker stock–recruitment model to walleye Sander vitreus stocking in northern Wisconsin lakes. Using June stocking data and fall age-0 survey data for 39 lakes over a 14-year time period, we found that the stocking rate resulting in the greatest number of age-0 walleyes was 60 age-0 walleyes/ha. Similarly, using June stocking data and fall age-1 survey data in 18 lakes over a 9-year time period we found that the stocking rate resulting in the greatest number of age-1 walleyes was 75 age-0 walleyes/ha. About 16% of the variation in fall age-0 walleye density was explained by the stocking rate, and 28% of the variation in fall age-1 walleye density was explained by the stocking rate. Density-dependent survival was apparent and significant, as low and high stocking densities resulted in lower age-0 and age-1 juvenile walleye densities. The lake surface area explained a significant amount of the residual variation in the age-1 stock–recruitment relationship such that stocked walleyes generally survived at lower rates in large lakes than in small lakes, which may be due to a greater diversity of predators in larger lakes. Based on our analysis, we recommend stocking small fingerling walleyes at a rate of 75 fish/ha in northern Wisconsin lakes.