TY - JOUR
T1 - A synchronized estimation of hourly surface concentrations of six criteria air pollutants with GEMS data
AU - Yang, Qianqian
AU - Kim, Jhoon
AU - Cho, Yeseul
AU - Lee, Won Jin
AU - Lee, Dong Won
AU - Yuan, Qiangqiang
AU - Wang, Fan
AU - Zhou, Chenhong
AU - Zhang, Xiaorui
AU - Xiao, Xiang
AU - Guo, Meiyu
AU - Guo, Yike
AU - Carmichael, Gregory R.
AU - Gao, Meng
N1 - This work is supported by the Research Grants Council of the Hong Kong Special Administrative Region, China (project no. HKBU12202021 and HKBU22201820) and the National Natural Science Foundation of China (No. 42005084). The authors are grateful to the GEMS science team for providing GEMS aerosol products and to China National Environmental Monitoring Center for providing ground-level air pollution data.
Publisher Copyright:
© 2023, The Author(s).
PY - 2023/7/18
Y1 - 2023/7/18
N2 - Machine learning is widely used to infer ground-level concentrations of air pollutants from satellite observations. However, a single pollutant is commonly targeted in previous explorations, which would lead to duplication of efforts and ignoration of interactions considering the interactive nature of air pollutants and their common influencing factors. We aim to build a unified model to offer a synchronized estimation of ground-level air pollution levels. We constructed a multi-output random forest (MORF) model and achieved simultaneous estimation of hourly concentrations of PM2.5, PM10, O3, NO2, CO, and SO2 in China, benefiting from the world’s first geostationary air-quality monitoring instrument Geostationary Environment Monitoring Spectrometer. MORF yielded a high accuracy with cross-validated R2 reaching 0.94. Meanwhile, model efficiency was significantly improved compared to single-output models. Based on retrieved results, the spatial distributions, seasonality, and diurnal variations of six air pollutants were analyzed and two typical pollution events were tracked.
AB - Machine learning is widely used to infer ground-level concentrations of air pollutants from satellite observations. However, a single pollutant is commonly targeted in previous explorations, which would lead to duplication of efforts and ignoration of interactions considering the interactive nature of air pollutants and their common influencing factors. We aim to build a unified model to offer a synchronized estimation of ground-level air pollution levels. We constructed a multi-output random forest (MORF) model and achieved simultaneous estimation of hourly concentrations of PM2.5, PM10, O3, NO2, CO, and SO2 in China, benefiting from the world’s first geostationary air-quality monitoring instrument Geostationary Environment Monitoring Spectrometer. MORF yielded a high accuracy with cross-validated R2 reaching 0.94. Meanwhile, model efficiency was significantly improved compared to single-output models. Based on retrieved results, the spatial distributions, seasonality, and diurnal variations of six air pollutants were analyzed and two typical pollution events were tracked.
UR - http://www.scopus.com/inward/record.url?scp=85165249022&partnerID=8YFLogxK
U2 - 10.1038/s41612-023-00407-1
DO - 10.1038/s41612-023-00407-1
M3 - Journal article
AN - SCOPUS:85165249022
SN - 2397-3722
VL - 6
JO - npj Climate and Atmospheric Science
JF - npj Climate and Atmospheric Science
IS - 1
M1 - 94
ER -
