A fisher-based, catch-sampling program for a large-scale, benthic molluscan fishery

Obtaining data on the length structure of commercial catches typically requires sub-sampling, but adequate, representative sampling is difficult to achieve. This problem is magnified in fisheries harvesting over large spatial scales or where the morphology among catches is variable. Data from the South Australian abalone fishery were used to determine a sampling program for the practical collection of accurate and representative commercial-catch-sampling data by commercial fishers. The performance of five alternative sampling programs was assessed against a range of simulated true populations. Regular, small samples at the first stage of the fishing process provided the most accurate and precise measures of the length structure of the commercial catch, whilst also having a low impost on fishers. Performance declined substantially at reduced participation rates; confidence intervals at 70% participation rates were typically twice those at 100%. Accuracy and precision were comparable among different fishing areas and true populations, suggesting a single sampling program would provide unbiased data on the length–frequency distribution of the commercial catch, even when fishers target numerous, morphologically different populations. Outcomes of this study are applicable to other fisheries targeting stocks with a variable, fine-scale population structure, and where industry stewardship promotes active participation in data collection and resource management.