2026/05/16 by Melissa J. Carabott, Barbara F. Nowak, Maree Widdicombe +4 · 1 voice
Biochemistry, Genetics and Molecular Biology · Immunology and Microbiology · Environmental Science · #Myxozoan Parasites in Aquatic Species #Aquaculture disease management and microbiota #Fish Ecology and Management Studies
paper · doi:10.1016/j.aquaculture.2026.744192
openalex publication_date 2026/05/16 · openalex created_date 2026/05/17 · openalex updated_date 2026/06/18
In ranched Southern bluefin tuna ( Thunnus maccoyii , SBT) aquaculture, research sampling relies on small subsamples (10–15 SBT) collected during commercial harvest to estimate fish weight and Cardicola spp. blood fluke burden. This study investigated potential bias associated with small sample sizes commonly used in SBT health research. SBT weights from on-site and simulated sampling were compared to the respective population data sets from four cages operated by a commercial company in Port Lincoln, South Australia at the end-of-season harvest in 2023. To assess the influence of small sample sizes on blood fluke infection, prevalence was simulated at increasing levels for the average cage population. Sampling 10–15 SBT produced estimates that closely approximated population weight in ∼95% of cases. Increasing the sample size offered marginal improvements, with minimal variation in the risk of bias. Estimates of population weight were negatively impacted when sampling exclusively from the first or last 100 fish, however, sampling across 1000 fish reduced the risk of biased results. Sampling 10–15 SBT was sufficient for parasite detection at moderate to high prevalence (25–90%). However, reliability decreased considerably at low prevalence (5–10%), increasing the probability failing to detect infection in simulated iterations. These findings demonstrate the influence of routine subsampling on population level inference, particularly at low infection prevalence, with important implications for the interpretation of research conducted in aquaculture and fisheries systems.