Estimating historical SO2 level across the whole China during 1973–2014 using random forest model. (May 2020)

• Record Type: Journal Article
• Authors: Li, Rui; Cui, Lulu; Liang, Jianhong; Zhao, Yilong; Zhang, Ziyu; Fu, Hongbo
• Is Part Of: Chemosphere. Volume 247(2020)
• Publication Date: 2020-05
• Subjects: SO2 level -- Emission inventory -- Random forest model -- China; Pollution -- Periodicals; Pollution -- Physiological effect -- Periodicals; Environmental sciences -- Periodicals; Atmospheric chemistry -- Periodicals; 551.511
• DOI: 10.1016/j.chemosphere.2020.125839
• Languages: English
• ISSNs: 0045-6535

Abstract

Abstract: Ambient SO2 pollution poses a great threat on air quality, human health, and ecosystem safety. The ground-level SO2 monitoring sites over China have been established during the past years, while the long-term SO2 data was still missing before 2014, which cannot reveal the evolution trend of SO2 pollution and assess its response to the anthropogenic activity. In this work, we developed a high-quality random forest (RF) model to simulate the long-term SO2 concentration across the entire China from 1973 to 2014, based on substantial explanatory variables (e.g., meteorological factors, SO2 emission intensity, land use types). The 10-fold cross-validation R 2 value and root mean square error (RMSE) over China reached 0.64 and 17.06 μg/m 3, respectively, both of which were significantly higher than those of other models such as back propagation neural network (BPNN) and generalized regression neutral network (GRNN). Among all of the predictors, T displayed the highest relative importance value, followed by WS, Prec, SO2 emission intensity, RH, DOY, elevation, and the lower one for land use types and P . The estimated mean SO2 concentration during 1973–2014 displayed the remarkably spatial variation with the higher value in North China Plain (NCP) and Middle part of Inner Mongolia. This historical SO2 level estimation suggested that air pollution was not a new environmental issue that could be dated back to 1973. Overall, the annually mean SO2 level for each grid increased from 29.46 ± 9.79 to 31.44 ± 8.77 μg/m 3 from 1973 to 2014. The annually mean SO2 concentration in NCP showed rapid increase from 34.32 ± 3.05 to 36.97 ± 3.18 μg/m 3 during 1973–2002, whereas they decreased significantly after 2003 (from 37.46 ± 3.20 to 36.13 ± 3.48 μg/m 3 during 2003–2014). The gradual decrease since 2003 was benefitted from the adjustment of the energy consumption structure and the adoption of emission control technologies. However, the SO2 levels in some western regions showed the violent increases since 2003 due to the proposal of "development of the western region". The estimated daily SO2 concentration across the entire China could provide the essential data for epidemiological research and air pollution prevention. Highlights: 1 The random forest model achieved the better performance in the long-term SO2 estimation based on SO2 emission and meteorological factors. 2 The annually mean SO2 level in China increased by 6.72% during 1973–2014. 3 The change of the energy consumption structure led to the SO2 decrease in North China Plain since 2003.