Enhanced interpretable data driven-based model for dissolved oxygen prediction under multiple environmental factor influences

Environmental context Maintaining sufficient dissolved oxygen in rivers and lakes is crucial for the health of fish, the quality of drinking water, and the overall well-being of ecosystems. This study presents an intelligent, interpretable machine learning-based prediction framework to predict dissolved oxygen variation and accurately identify multiple driving factors. The breakthrough provides water managers with a powerful tool to anticipate water quality changes and protect vital water resources more effectively. Rationale Dissolved oxygen (DO) is a critical indicator of water quality, but accurately predicting DO fluctuations under the combined influence of various environmental factors remains challenging. Methodology This study presents an intelligent prediction framework that integrates interpretable machine learning methods with stacking generalisation techniques to forecast dynamic changes in water quality effectively. Using daily data (2020–2022) from seven stations in the Danjiangkou Reservoir, China, environmental factors were categorised into water quality, streamflow, meteorology and air pollution. A stacked model, combining support vector machine (SVM), multilayer perceptron (MLP), random forest (RF) and K-nearest neighbours (KNN) base learners with a LightGBM meta-learner, was constructed and interpreted using Shapley additive interpretation (SHAP) analysis. The contribution of each feature is accurately determined, and the mode of action of the main factors and the key threshold points are identified. Results The stacked model provides efficient and accurate DO forecasts, outperforming single base models with an average improvement in R2 of 5.22% on the test set, and achieving a 2.29% increase in R2 and a 4.81% reduction in root mean squared error (RMSE) compared to the unstacked LightGBM model. Global interpretation reveals that the water quality parameter category has the most significant impact (up to 70% of total feature importance), with water temperature (WT) being the most crucial factor. Other key contributors include conductivity (EC), ammonia nitrogen (NH3-N), turbidity, surface runoff, atmospheric pressure (AP) and sulfur dioxide (SO2). Discussion The proposed framework effectively captures the synergistic effects of multiple environmental factors on DO. It offers a robust method for forecasting their dynamics, demonstrating the value of categorised environmental analysis for water quality management.