| Summary: | Owing to a series of air pollution prevention and control policies, China’s PM<sub>2.5</sub> pollution has greatly improved; however, the long-term spatial contiguous products that facilitate the analysis of the distribution and variation of PM<sub>2.5</sub> pollution are insufficient. Due to the limitations of missing values in aerosol optical depth (AOD) products, the reconstruction of full-coverage PM<sub>2.5</sub> concentration remains challenging. In this study, we present a two-stage daily adaptive modeling framework, based on machine learning, to solve this problem. We built the annual models in the first stage, then daily models were constructed in the second stage based on the output of the annual models, which incorporated the parameter and feature adaptive tuning strategy. Within this study, PM<sub>2.5</sub> concentrations were adaptively modeled and reconstructed daily based on the multi-angle implementation of atmospheric correction (MAIAC) AOD products and other ancillary data, such as meteorological factors, population, and elevation. Our model validation showed excellent performance with an overall R<sup>2</sup> = 0.91 and RMSE = 9.91 μg/m<sup>3</sup> for the daily models, along with the site-based cross-validation R<sup>2</sup>s and RMSEs of 0.86–0.87 and 12–12.33 μg/m<sup>3</sup>; these results indicated the reliability and feasibility of the proposed approach. The daily full-coverage PM<sub>2.5</sub> concentrations at 1 km resolution across China during the Three-Year Blue-Sky Action Plan were reconstructed in this study. We analyzed the distribution and variations of reconstructed PM<sub>2.5</sub> at three different time scales. Overall, national PM<sub>2.5</sub> pollution has significantly improved with the annual average concentration dropping from 33.67–28.03 μg/m<sup>3</sup>, which demonstrated that air pollution control policies are effective and beneficial. However, some areas still have severe PM<sub>2.5</sub> pollution problems that cannot be ignored. In conclusion, the approach proposed in this study can accurately present daily full-coverage PM<sub>2.5</sub> concentrations and the research outcomes could provide a reference for subsequent air pollution prevention and control decision-making.
|
|---|