High-Spatial Resolution Ground-Level Ozone in Yunnan, China: A Spatiotemporal Estimation Based on Comparative Analyses of Machine Learning Models

Monitoring ground-level ozone concentrations is a critical aspect of atmospheric environmental studies. In recent years, the escalating problem of ozone pollution has been particularly notable in Yunnan Province, a region in western China with significant development. Given the existing limitations of satellite data products and the discontinuity of ground observations, there is a pressing need for high-precision models to accurately simulate ground-level ozone to assess surface ozone pollution. In this study, we have compared several widely utilized ensemble learning and deep learning methods for ground-level ozone simulation. Furthermore, we have thoroughly contrasted the temporal and spatial generalization performances of the ensemble learning and deep learning models. The 3-Dimensional Convolutional Neural Network (3-D CNN) model has emerged as the optimal choice for evaluating the daily maximum 8-hour average ozone in Yunnan Province. This selected model adeptly integrates temporal and spatial information derived from satellite data, meteorological data, and other relevant auxiliary data. The model has good performance: a spatial resolution of 0.05° × 0.05° and strong predictive power, as indicated by a Coefficient of Determination (R2) of 0.9 and a Root Mean Square Error (RMSE) of 9.417 µg/m³ in sample-based 10-fold cross-validation (CV).In the final stage of our study, we applied the 3-D CNN model to generate a comprehensive daily maximum 8-hour average ozone dataset for Yunnan Province for the year 2021. This application has furnished us with a crucial high-resolution and highly accurate dataset for further in-depth studies on the issue of ozone pollution in Yunnan Province.