Improved seamless mapping of surface O3 concentrations using an integrated deep learning framework

Satellite-derived ozone (O3) data often contain spatial gaps due to factors such as cloud cover. To achieve seamless O3 mapping, researchers typically either reconstructed the missing satellite input data before the O3 inversion or reconstructed the missing O3 data after inversion. Unlike previous step-by-step approaches, this study proposed a deep learning-based “inversion-reconstruction” integrated framework to estimate seamless surface O3. By inputting gapped satellite data and other auxiliary information, the framework directly yielded gap-free O3 data. The O3 inversion and reconstruction results were jointly optimized in the framework, ensuring high consistency in the seamless mapping of O3 concentrations. Holdout, spatial, and temporal validations demonstrated the effectiveness of our method for mapping seamless O3 across China in 2019, with R² values of 0.809, 0.760, and 0.733, respectively. Daily seamless mapping revealed the spatiotemporal patterns of O3, pollution episodes, and their potential transport routes. The satellite-inverted gapped O3 data showed a 7.37 ± 4.18% difference from the gap-free merged O3 data on a national daily scale.