Biotic and abiotic factors that influence Walleye recruitment in stocked lakes in Michigan

Abstract Objective In recent decades, widespread evidence of declining Walleye Sander vitreus recruitment in the Midwestern United States has raised concerns about diminishing ecological benefits and fishing opportunities, which in turn has resulted in natural resource agencies using stocking to supplement populations. Success of Walleye stocking events vary; thus, evaluating the external factors that influence recruitment success of Walleye resulting from stocking or natural production is critical for improved management. Methods We evaluated factors influencing the relative recruitment in 98 lakes stocked annually with Walleye by integrating 20 years of stocking data with fisheries assessment and abiotic data. We indexed recruitment as the relative abundance of Walleye cohorts from natural and stocked sources combined and used boosted regression tree modeling to identify variables that are associated with varying recruitment. Result Growing degree‐days at age 1, shoreline development index, mean depth, and degree‐days at age 2 were the most important variables. The boosted regression tree model explained 17% of observed variability in recruitment. A positive relationship between growing degree‐days and Walleye recruitment was observed, whereas shoreline development and lake depth had a negative relationship with recruitment. Secondarily important predictors were lake surface area, spring warming rate, Northern Pike Esox lucius abundance, and degree‐days at age 0. In general, Walleye recruitment was positively associated with these predictors, with the exceptions of spring warming rates >0.3°C/day and being exposed to >3500 degree‐days at age 0. Conclusion Our results indicate that thermal variables in addition to lake characteristics are important recruitment drivers in stocked waters and highlights the potential for resource managers to use this information to prioritize stocking efforts in temperate lakes.