Assessing consistency between fisheries logbook and machine learning-derived VMS data for skipjack tuna fishing effort in the Western Sumatra Indian Ocean

This study addresses the need for accurate fisheries data by assessing the consistency between fishing logbook records and fishing efforts derived from Vessel Monitoring System (VMS) data using machine learning in the Western Sumatra Indian Ocean. The objective of this study was to evaluate the reliability of the skipjack tuna (Katsuwonus pelamis) fishing effort data. We utilized VMS data from to 2014-2023, processed through a vmstofish machine learning function, and compared it with logbook data from Pelabuhan Nizam Zachman Jakarta. The vmstofish function, utilizing the CatBoost model, demonstrated high effectiveness in detecting fishing effort, achieving a recall of 0.983 and an F1-score of 0.931, proving its validity as an alternative data source. Spatiotemporal analysis revealed a significant increase in perfect match rates between VMS-derived and logbook data from 2019-2023 (86.6%), as the impact of e-logbook implementation, indicating improved logbook data quality in recent years. This research provides a robust method for complementing and evaluating fisheries data, offering a more comprehensive understanding of fishing activities crucial for sustainable management, and contributing to blue economy initiatives.