Estimating daily full-coverage and high-accuracy 5-km ambient particulate matters across China: considering their precursors and chemical compositions

Abstract. The ambient concentrations of particulate matters (PM_2.5 and PM₁₀) are significant indicators for monitoring the air quality relevant to living conditions. Most of the existing approaches for the estimation of PM_2.5 and PM₁₀ employed the remote sensing Aerosol Optical Depth (AOD) products as the main variate. Nevertheless, the coverage of missing data is generally large in AOD products, which can cause inconvenience to the researchers. To efficiently address this issue, our study explores a novel approach using the datasets of the precursors and chemical compositions for PM_2.5 and PM₁₀ instead of AOD products. Specifically, the daily full-coverage ambient concentrations of PM_2.5 and PM₁₀ are estimated at 5-km (0.05°) spatial girds across China based on Sentinel-5P and GEOS-FP. In this paper, the Light Gradient Boosting Machine is exploited to train the estimation models, which will fully fuse the multi-source data. For comparison, the Deep Blue AOD product from VIIRS is adopted in a similar framework as a baseline (AOD-based). The validation results show that the ambient concentrations are well estimated through the proposed approach, with the sample-based Cross-Validation R²s and RMSEs of 0.93 (0.9) and 8.982 (17.604) μg/m³ for PM_2.5 (PM₁₀), respectively. Meanwhile, the proposed approach achieves better performance than the AOD-based in different cases (e.g., overall and seasonal). Compared to the related previous works over China, the estimation accuracy of our method is also satisfactory. Furthermore, all the variates of the precursors and chemical compositions for PM_2.5 and PM₁₀ positively contribute to the estimation in the proposed approach, as expected. With regard to the mapping, the estimated results through the proposed approach present consecutive spatial distribution and can exactly express the seasonal variations of PM_2.5 and PM₁₀. It is concluded that the full-coverage estimated results in our study are conducive to the researches on PM_2.5 and PM₁₀ over the regions where the AOD values are missing.