Impacts of Anchor Tag Loss on Walleye Management in the Winnebago System

Rates of tag loss should be estimated and accounted for when using mark–recapture surveys and angler tag returns to estimate fish population abundance and exploitation rates. Walleyes Sander vitreus sampled during April 2010 spawning assessments conducted in the Winnebago system, Wisconsin, were marked with anchor tags and upper caudal fin clips to estimate tag loss rates during three time intervals: 0–11 d posttagging, 0–90 d posttagging, and 1 year posttagging. Tag loss was negligible (<0.5%) within the first 11 d but increased to 4.7% within the first 90 d and to 21.9% after 1 year. After we corrected for the tag loss occurring within the first 90 d, estimates of population abundance decreased 4.3% for females and 4.4% for males, while estimates of exploitation increased 4.9% for both sexes. Accounting for 21.9% annual tag loss led to more severe decreases in the estimates of population abundance (20.0% for females and 21.2% for males) and increases in the estimates of exploitation (28.0% for both sexes) than not accounting for any tag loss. Higher exploitation rates resulted in higher estimates of fishing mortality and lower estimates of natural mortality. When used in yield‐per‐recruit models, the estimated natural mortality rates derived from exploitation rates that accounted for tag loss resulted in higher yields through the implementation of larger minimum length limits to maximize yield. We recommend that managers evaluate the effectiveness of their tagging operations by estimating tag loss and assessing the potential impacts of quantified loss on management metrics. We also caution against the application of population models without an accurate estimate of tag loss and the associated impacts on estimates of population abundance and exploitation and, in turn, fishing and natural mortality rates. Received May 22, 2013; accepted June 11, 2013