Forecasting Global Ocean Health from Multidimensional Sustainability Indicators: An Interpretable Random Forest Framework

Introduction:
Forecasting global ocean health is essential for identifying vulnerable marine systems before ecological decline becomes more difficult to reverse. Although the Ocean Health Index provides a widely used measure of marine ecosystem condition and human ocean benefits, its longitudinal structure remains underused for predictive modelling. This study develops an interpretable artificial intelligence framework for forecasting next-year ocean health performance and translating multidimensional sustainability indicators into risk-relevant decision intelligence.

Methods:
The study analyses 341,124 Ocean Health Index score observations covering 220 coastal countries and territories, 19 goals or subgoals, 6 score dimensions, and annual assessments from 2012 to 2025. The framework combines data harmonisation, longitudinal trajectory modelling, K-means country grouping, trajectory-based risk stratification, Random Forest forecasting, feature importance analysis, and country-specific policy diagnostics. The forecasting model used lagged goal and dimension-level predictors, while the contemporaneous overall index was excluded from the predictor set to avoid circular prediction.

Results:
The global Ocean Health Index rose from 71.85 in 2012 to 74.94 in 2019, declined to 69.50 in 2024, and partially recovered to 72.31 in 2025, remaining 2.63 points below its 2019 level. Country trajectories separated into two regimes with mean 2025 scores of 65.07 and 74.47, while risk stratification identified 51 low-risk, 98 moderate-risk, and 71 high-risk ocean health systems. The Random Forest model predicted next year OHI scores with R² = 0.723, MAE = 3.07, and RMSE = 4.03 on the temporal holdout sample. Habitat, carbon storage, and fisheries emerged as the strongest predictors of future ocean health.

Conclusions:
This study advances AI forecasting by demonstrating that multidimensional ocean sustainability indicators contain substantial predictive information about future marine system performance. The framework offers an interpretable basis for ocean health forecasting, risk prioritisation, and policy oriented environmental decision support.