Using simulation to understand annual sea lamprey marking rates on lake trout

Abstract

Sea lampreys attack fish, killing some and leaving marks on others. Great Lakes fishery managers rely on observed marking rates to assess the success of the sea lamprey control program and estimate sea lamprey-induced mortality of lake trout. Because marking rates are only observed on survivors of sea lamprey attacks, they may not provide a reliable index of actual attack or mortality rates. To investigate the effect of survivor bias, we developed a simulation model representing a single season (June–December) of sea lamprey attacks. Simulated attack rates varied with month and lake trout size; simulated pierce and lethality rates varied with month alone. Surveyed marking rates were represented by simulated survivors in October; true rates were calculated from all simulated lake trout (dead and alive) in December. Simulation results were subsetted to include only those within the range of marking rates actually observed in the Great Lakes. Type A (piercing) marking rates were a good index of the sea lamprey attack rate and the sea lamprey-induced mortality rate if annual lethality rates were relatively constant. Type B (non-piercing) marking rates were a good index of the sea lamprey attack rate and the sea lamprey-induced mortality rate if annual pierce rates were relatively constant. Due to the uncertainty surrounding the pierce and lethality rates, we recommend that sea lamprey abundance information be incorporated in existing lake trout statistical catch-at-age models via a functional response component relating sea lamprey feeding to lake trout abundance, if possible.