Gorgan Bay, a vital ecological and economic region in the southeastern Caspian Sea, faces escalating threats from climate change and human activities, resulting in diminished water volume and deteriorating quality. This study harnesses cutting-edge machine learning and remote sensing to monitor chlorophyll-a (Chl-a) concentrations, a key indicator of water quality and ecosystem health. Leveraging Sentinel-2 satellite data, field measurements, and advanced regression models, we explored correlations between spectral bands, the Normalized Difference Chlorophyll Index (NDCI), and water depth to predict Chl-a levels. Multiple linear regression (MLR), Random Forest (RF), Classification and Regression Trees (CART), and Gradient Tree Boost (GTB) models were developed to map Chl-a distributions, with various input combinations rigorously tested. The CART model emerged as the top performer, achieving a Root Mean Squared Error (RMSE) of 6.32 μg/L and a coefficient of determination (R²) of 0.82, demonstrating robust predictive accuracy. These findings offer a powerful tool for real-time monitoring of Gorgan Bay’s ecological status, providing critical insights for conservation strategies and sustainable management. By integrating satellite technology with machine learning, this research paves the way for innovative approaches to safeguard vulnerable aquatic ecosystems, captivating researchers and policymakers eager to address pressing environmental challenges.