Use of Underwater Visual Distance Sampling for Estimating Habitat-Specific Population Density

Abstract

We contrasted fish abundance estimates generated from mark–recapture and underwater visual distance sampling to determine whether the latter method is a potentially valuable fisheries assessment tool. We further examined whether altering the detection function or habitat stratification and including lake characteristics such as Secchi depth, temperature, and fish density affected distance sampling estimates. Distance sampling produced estimates that were comparable to those of mark–recapture techniques (r ² = 0.60), and the relationship improved considerably when two species that were difficult to sample visually were removed from the analysis (r ² = 0.88). The precision of mark–recapture estimates was significantly better than that of distance sampling. Distance sampling estimates were more similar to mark–recapture estimates when stratified by habitat than when pooled across habitats. The addition of Secchi depth, temperature, fish density, or a combination thereof to a regression equation that included mark–recapture and distance sampling estimates did not improve the strength of the relation. We were able to detect significant among-habitat differences using habitat-specific density estimates provided from distance sampling. Distance sampling is a less-intrusive means to determine fish abundance and microhabitat patterns and provides a way to determine age-0 fish abundance in addition to information (e.g., school size) that is not readily obtained by mark–recapture studies.