Effects of location errors on estimates of dredge catchability from depletion based methods

Depletion-based methods are used to estimate the catchability of a research dredge survey for blue crabs (Callinectes sapidus) in Chesapeake Bay. The experimental design relies on the ability to repeatedly sample the same area, but experiments have not been conducted to determine the effects of sampling location error on catchability estimates. We conducted a simulation study to evaluate the effects of sampling location errors on three catchability estimators (Leslie, Ricker, and Rago). We simulated the distribution of crabs in an area and repeatedly sampled from the area using a range of true values of catchability and four methods to constrain the sampling area: perfect knowledge, buoy deployment, high-accuracy GPS, and consumer-grade GPS. No estimator was best across all scenarios, and in some scenarios no estimator performed particularly well. Error in sampling location generally caused negative bias in the catchability estimates with the amount of bias increasing as location error increased. While the Leslie and Rago methods were relatively accurate when location errors were small, the Ricker method performed poorly because of the constant added to allow zero catches. The Leslie or Rago method performed well when combined with buoys to demarcate the sampling area, and the Rago method performed well with high-accuracy GPS.