Nationwide Machine Learning-Ensemble PM Mapping Prediction and Forecasting Models in South Korea with High Spatiotemporal Resolution and Health Risk Estimation-Based Evaluations.

Several tudies developed machine learning-based PM prediction models; however, nationwide models addressing both mapping prediction and forecasting were limited. Further, although the prediction accuracy is different from PM-related health risk estimation, previous studies solely examined the prediction accuracy. This study suggests a method to assess the statistical properties of PM-health risk estimation, which also can be used as a model selection. We used three machine learning algorithms and an ensemble method to construct PM mapping prediction (1 km) and two-day forecasting models majorly using satellite-driven data in South Korea (2015-2022). We performed a simulation study to examine the statistical properties of short-term PM risk estimation using prediction models. Our ensemble spatial prediction model showed better performance than single algorithms (0.956 test ). The range of the values was 0.78-0.98 across the monitoring sites. The average % bias was from 1.403%-1.787% when our mapping models for PM-mortality risk estimation, compared to the estimates from monitored PM. The best of our forecasting models was 0.904. This study developed machine learning models for spatial PM predictions and forecasting in Korea. This study also suggested a method to address risk estimation and model selection concurrently when multiple prediction models were used.