Understanding walleye (Sander vitreus) spawning behavior is important for managing walleye fisheries, but such information is limited for Appalachian reservoirs. We assessed spawning movements and spawning locations for a reestablished walleye population in Cheat Lake, West Virginia. We tagged fifty-two walleye with acoustic telemetry transmitters to evaluate environmental correlates associated with pre-spawn movements and to deter- mine spawning locations. Using an information-theoretic approach, we compared candidate logistic regression models to determine which environmental variables best explained upstream movements to spawning areas. The two models with the most support both included additive effects of year and water temperature, with sex also included in the second of these models. Water temperature had a significant positive relationship with pre-spawn movements in each model. Other environmental covariates such as river discharge and water elevation were not significant predictors of upstream pre-spawn move- ments. Walleye made pre-spawn upstream movements in late winter/early spring to spawning areas in the headwaters of Cheat Lake during periods of el- evated water temperatures (75 % of movement events occurred at water temperatures >4.1 C) where spawning occurred in shallow (<1.5 m), rocky habitat. Male walleye generally made upstream pre-spawn movements earlier than females. Our results also suggested the timing of walleye spawning with respect to water-level fluctuations could influence reproductive success due to stranding of eggs or reducing suitable spawning habitat. Knowledge of pre-spawn movement patterns and spawning locations could aid management of this recovering population. Benefits to management may include the prediction of spawning timing and locations for broodstock surveys and influences of water-level fluctuations and other environmental stressors on spawning success.
