Estimation of ground-level O concentration in the Yangtze River Delta region based on a high-performance spatiotemporal model MixNet.

In recent years, the escalating ozone (O) concentration has significantly damaged human health. The machine learning models are widely used to estimate ground-level O concentrations, but the spatial and temporal features in the data are less considered. To address the issue, this study proposed a novel framework named MixNet to estimate daily O concentration from 2020 to 2021 over the Yangtze River Delta. The MixNet utilized image convolution to extract the potential spatial information related to O fully. The temporal features were extracted by a Long Short-Term Memory (LSTM). A U-Net, a new jump connection method with an attention mechanism and residual blocks, facilitated a more comprehensive extraction of spatial features in the data. The extracted temporal and spatial features were fused to estimate ground-level O. Meanwhile, a novel training method was proposed to enhance the accuracy of MixNet. The daily mean O maps have high validation results in comparison with ground-level O measurement, with R (RMSE) of 0.903 (14.511 μg/m) for sample-based validation, 0.831 (19.036 μg/m) for site-based validation, and 0.712 (25.108 μg/m) for time-based validation. The season-average maps indicate that O concentration is summer > autumn > spring > winter. The highest value was 137.41 μg/m in the summer of 2021 over the Yangtze River Delta urban agglomeration, and the lowest value was 52.73 μg/m in winter 2020. The MixNet showed better performance compared with other models, and thus the "point-plane image thinking" will contribute to future studies in developing better methods to estimate atmospheric pollutants.