Search

• Species misidentification and lack of fishing gear data in official statistics. • Statistical stepwise procedure to estimate landed weight by species and effort. • Relevance of dedicated sampling programs for by-catch species. • Collaboration with fishermen as a source of relevant fisheries knowledge. To assess the exploitation status of marine resources and to formulate scientific advice for fisheries management, reliable fishery data are required. When dealing with data-poor stocks, like skates and rays (Rajiformes) caught by multispecies and multigear fisheries in Portugal, strategies need to be defined to surpass data limitations, like species misidentification and the lack of fishing gear information in landings data. This paper presents a stepwise statistical procedure, involving factor analysis for mixed data and flexible discriminant analysis, to estimate the total number of trips and the correspondent total landed weight by species. Interviews and self-sampling that took place under a pilot study (2011 – 2013) focused on skate and ray fisheries showed that collaborative research is a low cost and valuable tool to improve our knowledge on fisheries and their catches and may contribute to increase the accuracy. Given the persistence of species misidentification errors in the official statistics, the procedure here presented is currently used to report the Portuguese annual estimates of species-specific landed weight. This procedure may be considered to be applied to data deficient fishery bycatch resources for which dedicated sampling programs are not priority. [ABSTRACT FROM AUTHOR]