WilDect-YOLO: An efficient and robust computer vision-based accurate object localization model for automated endangered wildlife detection

\n Numerous pelagic species are known to associate with floating objects (FOBs), including tropical tunas. Purse seiners use this behaviour to facilitate the capture of tropical tunas by deploying artificial drifting fish aggregating devices (dFADs). One major recent change has been the integration of echosounders in satellite-tracked GPS buoys attached to FOBs, allowing fishers to remotely estimate fishable biomass. Understanding the effects of this new technology on catch of the three main tuna species (yellowfin tuna, Thunnus albacares; bigeye tuna, Thunnus obesus; and skipjack tuna, Katsuwonus pelamis) is important to accurately correct for this change in catch-per-unit-effort (CPUE) indices used for stock assessments. We analysed catch data from the French purse seine fleet for the period 2010–2017 in the Indian Ocean to assess the impact of this fleet’s switch to echosounder buoys around 2012. Results indicate that echosounders do not increase the probability a set will be succesful, but they have a positive effect on catch per set, with catches on average increasing by ≈2−2.5 tonnes per set (≈10%) when made on the vessel's own dFADs equipped with an echosounder buoy. Increases were due to a decrease in sets below ≈25 tonnes and an increase in those greater than ≈25 tonnes, with a non-linear transition around this threshold. This increase explains the considerable investment of purse seiners in echosounder buoys, but also raises concerns about bias in stock size estimates based on CPUE if we do not correct for this fishing efficiency increase.

Quantifying the increase in fishing efficiency due to the use of drifting FADs equipped with echosounders in tropical tuna purse seine fisheries

Towards a framework for fishing route optimization decision support systems: Review of the state-of-the-art and challenges

Fuel consumption and air emissions in one of the world's largest commercial fisheries.

. This study addresses novel questions on the dynamics of tuna aggregations around floating objects, using echosounder buoys data collected throughout t

Yannick Baidai

Papers

Machine learning for characterizing tropical tuna aggregations under Drifting Fish Aggregating Devices (DFADs) from commercial echosounder buoys data

Machine learning to detect bycatch risk: Novel application to echosounder buoys data in tuna purse seine fisheries

Tuna aggregation dynamics at Drifting Fish Aggregating Devices: a view through the eyes of commercial echosounder buoys