Sensitivity of Exploitation Estimates to Tag Loss Estimation Methods in Idaho Sport Fisheries

The objective of this study was to evaluate the sensitivity of exploitation estimates to tag loss estimation methods for six species of sport fish in Idaho. A discrete tag loss model and four instantaneous tag loss models were evaluated using both empirical data and a simulation approach. Empirical exploitation estimates were similar for most species when using the discrete model and all four forms of instantaneous tag loss model. Maximum differences among estimators were less than 2% for five species; however, exploitation estimates differed by as much as 17% for Walleyes Sander vitreus depending on the choice of the tag loss model. The discrete estimator of tag loss, which was computationally much easier to implement than the instantaneous models, resulted in exploitation estimates that were similar to or more accurate than the instantaneous models in the simulation. Exploitation in the simulation was underestimated by more than 1% at a minimum and by greater than 14% at maximum when tag loss was not accounted for. These results suggest that regardless of the choice of tag loss estimator, an attempt should be made to account for tag loss even for species that are expected to have a relatively short average time at liberty and relatively low tag loss over the duration of a study.