Partitioning total size selectivity of gill nets for walleye (Stizostedion vitreum) into encounter, contact, and retention components

Contact and retention components were estimated by indirect methods for walleye (Stizostedion vitreum) in Lake Erie, Dexter Lake (Ontario), and nine Minnesota lakes. In all 11 lakes, the retention function was well described by a unimodal function of fish size/mesh perimeter, and the addition of a contact component significantly improved model fit. A plausible range of models was identified for Lake Erie by considering the contact component to be constant or proportional to mesh size. Total catchabilities were estimated for two lakes by direct methods and encounter probabilities estimated by comparison of direct and indirect selectivities. Encounter probabilities estimated for Dexter Lake increased proportionately to fish length. Encounter probabilities estimated for Mille Lacs had a more complex form and were not well described by allometric power functions such as those sometimes used to describe swimming speed and encounter rates.