Optimum sampling levels in discard sampling programs

This paper presents optimum sampling levels in discard sampling programs considering cost and precision objectives simultaneously and explores their dependence on both variables. The analysis is based on the Irish discard program: an onboard-observer voluntary sampling scheme aimed at estimating discard rates in trawl fisheries. Multistage analysis was performed to establish the precision levels achieved in the past, and a cost function was determined to estimate the financial cost of the program. Gear, fishing ground, targeted species, and International Council for the Exploration of the Sea (ICES) divisions were the main factors affecting discarding, together with random effects of the three nested groups considered: haul, trip, and vessel. Reductions in the present budget will imply only marginal decreases in precision, although changes in cost variables can have an impact on sampling levels. On the other hand, increasing the target precision by one-half will imply a considerable increase in sampling and associated cost, which will be difficult if not impossible to achieve. Finally, the analysis by fleet components suggests a marked increase in sampling levels, which emphasizes the importance of clearly stated discard sampling objectives.