Estimating Daily Concentrations of Near-Surface CO, NO2, and O3 Simultaneously Over China Based on Spatiotemporal Multi-Task Transformer Model

Accurate and efficient estimation of near-surface air pollutant concentrations holds significant practical importance. Current models for estimating near-surface concentrations (NSC) primarily rely on shallow methods and focus on estimating a single pollutant. However, these models face challenges in capturing the complex spatiotemporal patterns of NSC and demonstrate inefficiency. To overcome these limitations, we propose a spatiotemporal multi-task Transformer model (stmtTransformer) to simultaneously estimate the NSC of carbon monoxide (CO), nitrogen dioxide (NO2), and ozone (O3). Estimation experiments conducted in China from 2021 to 2022 demonstrate that stmtTransformer achieves optimal performance by effectively capturing the spatiotemporal variations of NSC. Based on sample-based validation, the R2 values are 0.643 (CO), 0.781 (NO2), and 0.902 (O3), and the RMSE values are 0.194 mg/m3 (CO), 5.613 mu g/m3 (NO2), and 13.330 mu g/m3 (O3), respectively. In terms of efficiency, stmtTransformer significantly improved the training efficiency by 185.21 % and the estimation efficiency by 129.44 % compared to the single-task model. Finally, when plotting the daily and seasonal maps of NSC for 2022, it is evident that the estimates exhibit a consistent spatial distribution.

Automated summary

Accurate and efficient estimation of near-surface air pollutant concentrations is of practical importance. Current models primarily rely on shallow methods and focus on estimating a single pollutant, facing challenges in capturing complex spatiotemporal patterns and demonstrating inefficiency. To overcome these limitations, a spatiotemporal multi-task Transformer model is proposed to simultaneously estimate the near-surface concentrations of carbon monoxide, nitrogen dioxide, and ozone. Experiments conducted in China demonstrate that the model achieves optimal performance and significantly improves efficiency compared to single-task models.