Multi‐Algorithm Species Distribution Modeling for Skipjack Tuna Fishing Ground Detection in the Indian Ocean: A Case Study of Indonesian FMA 572

The distribution of skipjack tuna ( Katsuwonus pelamis ) in the Indian Ocean is dynamic and influenced by oceanographic conditions. This study applies Species Distribution Modelling (SDMs) to predict the potential habitat distribution of skipjack tuna using catch logbook data and marine environmental variables. Four algorithms were employed in the modeling process, including Generalized Additive Model (GAM), Multivariate Adaptive Regression Splines (MARS), Maximum Entropy (MAXENT), and Support Vector Machine (SVM). Model evaluations showed high predictive performance based on Area Under the Curve (AUC) and True Skill Statistic (TSS) values. Sea Surface Height (SSH) was identified as the most significant contributing variable, with an optimal range between 0.2 and 0.6 m. The combination of multiple algorithms improved the accuracy of potential fishing ground predictions. These findings provide a strong scientific basis to support adaptive and sustainable management strategies for skipjack tuna fisheries in tropical marine waters.