A hierarchical ensemble approach for multi-country PM10 forecasting using LightGBM and residual neural network

Accurate day-ahead forecasting of particulate matter (PM10) concentrations is critical for public health interventions, regulatory compliance, and urban air quality management. However, existing approaches suffer from temporal leakage, single-city limitations, inadequate hierarchical modeling of geographic dependencies, and reliance on single-model architectures that fail to capture complex nonlinear pollution dynamics. This study presents a novel three-stage leakage-free stacked ensemble framew