PREDICTION OF FISH CATCH IN INDONESIAN WATERS BASED ON CLIMATE FACTORS

Climate change and variability in ocean-atmosphere interactions significantly impact marine ecosystems, including fish catch yields. One of the primary factors affecting fish availability is climate conditions, such as the El Niño-Southern Oscillation (ENSO), Indian Ocean Dipole (IOD), Madden-Julian Oscillation (MJO), Sea Surface Temperature (SST), Salinity, and Current Velocity. This study aims to analyze the relationship between these climate factors and the catch of Skipjack Tuna, Mackerel, and Squid in Indonesian waters.

This study employs a time-series analysis by integrating climate variables with fish catch data. The research methodology consists of three main stages: (1) Data Processing, which involves loading, cleaning, categorizing, and merging datasets, as well as applying log transformations to stabilize variance; (2) Analysis and Modeling, which includes feature selection, lag variable creation, and the application of predictive models such as Pearson Correlation, Random Forest, and Linear Regression, evaluated using K-Fold Cross Validation; and (3) Model Evaluation and Visualization, where model performance is assessed using R² and RMSE, and predictions are compared with observational data through scatter plots and time-series graphs.

The results indicate that current velocity is a crucial factor influencing fish distribution, affecting migration patterns and food availability, ultimately impacting fish catch yields. This study underscores the importance of climate-based analysis in predicting fish catch variations. The findings are expected to contribute to data-driven fisheries management and assist fishermen and policymakers in optimizing fishing strategies to support the sustainability of Indonesia’s fisheries sector.

Keywords: Fish Catch Prediction, Climate Factors, Machine Learning, ENSO, SST, Current Velocity