Distance-based abundance estimation for abalone

Indices of abundance are an important component of stock assessment models. Unfortunately, conventional attempts to estimate the abundance of abalone are hampered by the patchy spatial distribution characteristic of abalone. An alternative distance-based abundance estimator was evaluated by the simulated sampling of a large natural population of abalone whose relative positions had been accurately mapped. Distance-based abundance estimators can better accommodate the aggregated distribution of abalone. The simulated sampling of the real population of blacklip abalone (Haliotis rubra Leach) was used to examine the accuracy, bias, and sensitivity to changes in abundance of the distance-based approach. In each sample the distances from a randomly selected point and from a randomly selected abalone to the nearest abalone were used as the basis for a compound estimate of abundance. A Monte Carlo procedure was used to obtain percentile confidence limits about abundance estimates. The distance-based method was found to approximate the true abundance of the population and therefore may be useful as an indicator of absolute abundance. The method was also sensitive to changes in abundance. This sensitivity was examined by simulating the effects of fishing at varying rates of exploitation. Simulated reductions at moderate levels of exploitation (20-25% overall reduction) showed that the method was able to detect changes in abundance with reasonable confidence (90%). The resolution of some technical difficulties may enhance the ability of the distance-based method to detect changes in abundance in a real fishery.