Constructing a spatiotemporally coherent long-term PM2.5 concentration dataset over China during 1980-2019 using a machine learning approach

The lack of long-term observations and satellite retrievals of health-damaging fine particulate matter in China has demanded the estimates of historical PM2.5 (particulate matter less than 2.5 mu m in diameter) concentrations. This study constructs a gridded near-surface PM2.5 concentration dataset across China covering 1980-2019 using the space-time randomforest model with atmospheric visibility observations and other auxiliary data. Themodeled daily PM2.5 concentrations are in excellent agreementwith groundmeasurements, with a coefficient of determination of 0.95 and mean relative error of 12%. Besides the atmospheric visibility which explains 30% of total importance of variables in the model, emissions and meteorological conditions are also key factors affecting PM2.5 predictions. From 1980 to 2014, the model-predicted PM2.5 concentrations increased constantly with the maximum growth rate of 5-10 mu g/m(3)/decade over eastern China. Due to the clean air actions, PM2.5 concentrations have decreased effectively at a rate over 50 mu g/m(3)/decade in the North China Plain and 20-50 mu g/m(3)/decade over many regions of China during 2014-2019. The newly generated dataset of 1-degree gridded PM2.5 concentrations for the past 40 years across China provides a useful means for investigating interannual and decadal environmental and climate impacts related to aerosols. (C) 2020 Elsevier B.V. All rights reserved.

Automated summary

This study constructed a dataset of near-surface PM2.5 concentrations in China from 1980 to 2019 using a space-time randomforest model, with simulated concentrations showing excellent agreement with ground measurements. Atmospheric visibility, emissions, and meteorological conditions were identified as key factors affecting PM2.5 predictions. Clean air actions have effectively reduced PM2.5 concentrations in certain regions of China.