Adapting the sampling design of research surveys to improve the biomass estimation of non-target species - the case study of Raja clavate

Abstract

Research surveys are important to evaluate the spatial distribution of fishery re sources and to monitor their abundance. However, the underlying sampling is usually conceived with the focus on specific species and an efficient design may reconcile this objective with the collection of non-target species data. This study evaluates the adequacy of different sampling designs for the IPMA bottom trawl survey conducted along the Portuguese continental coast. It aims to obtain the maximisation of accuracy estimates for the non-target thornback ray Raja clavata, while maintaining the estimates quality for the European hake, Merluccius merluccius, one of the target species. A geostatistical model-based approach was developed considering the semi-continuous nature of the data and the excess of zero values for R. clavata. The proposal of new sampling designs relied on eight optimisation weights and the resulting prediction exactness. Eight survey designs were then evaluated, each adding 11 sampling locations to the survey locations from the original sur vey design. Three designs resulted in higher accuracy for both R. clavata and M. merluccius estimates than those obtained from the random design. The sampling design that maximised the accuracy of R. clavata estimates showed an acceptable trade-off between bias and variance of M. merluccius estimates.