Report on preliminary analyses of FAD acoustic data

Escalle L, Heuvel BV, Clarke R, et al (2019) Report on preliminary analyses of FAD acoustic data. In: WCPFC Scientific Committee 15th Regular Session., Pohnpei, Federated States of Micronesia, p 16

The deployment of satellite and echo-sounder buoys on drifting Fish Aggregating Devices (FADs) has dramatically increased their use by the purse seine fishery, with more than 30,000 FADs estimated to be deployed annually in the Western and Central Pacific Ocean (WCPO). This large volume of echo-sounder readings transmitted every day by buoys on FADs has the potential to be a useful source of information for scientific analysis that could help inform mitigation measures, enhance our understanding of fishery dynamics, and potentially provide independent data on tuna biomass for regional stock assessments. To this end, the current study investigates the type of data available, ‘ground truths’ acoustic estimates, and identifies further avenues of research.
The available data comprise acoustic data from over 5000 buoys deployed on FADs from US-based private sector firms Tri Marine and South Pacific Tuna Corporation in the WCPO in 2016–2018. This included data from two different satellite echo-sounder buoys: Satlink and Zunibal, which present different operational characteristics, such as biomass estimates, depth bins, transmission frequency.
The biomass estimates from echo-sounder buoys were found to be influenced by i) the time of the day, with maximum biomass estimated before sunrise, and ii) the lunar phase, with a slight increase in biomass detected during and just after the full moon.
FAD colonization processes were investigated using the maximum daily biomass estimates after deployment. Biomass estimates showed a significant increase up to around 30 days drifting. To investigate the biomass colonization before a fishing set, catch per set from logsheet operational data were matched with the acoustic dataset using position (≤ 2km) and date/time (same date). In general, high variability was detected and no clear pattern could be identified between catch and echo-sounder biomass estimates. Many factors may influence both the echo-sounder estimated biomass and the catch per set. For instance, it would be relevant to assess the catch/biomass relationships by large areas of the WCPO, as they would present different environmental characteristics. In general, an increasing trend in estimated biomass was detected over the two to five days before a fishing set. Relatively high biomass was noted >15 days before a set for many FADs.
The annual spatial distribution of biomass estimated from buoys was investigated. Although this was influenced by the fishing grounds of the two fishing companies and by the difference in estimated biomass between both echo-sounder buoy brands, it showed higher biomass in the eastern WCPO from 2°S to 10°S and 2°N to 5–10°N. Visual comparison with maps of total CPUE from associated sets showed some similarities in areas with high estimated biomass and high CPUE.