TY - JOUR
T1 - An ensemble mixed spatial model in estimating long-term and diurnal variations of PM2.5 in Taiwan
AU - Wong, Pei Yi
AU - Su, Huey Jen
AU - Lung, Shih Chun Candice
AU - Wu, Chih Da
N1 - Funding Information:
This study was grant supported by the Ministry of Science and Technology, Taiwan (MOST 111-2121-M-006-004-; MOST 110-2628-M-006 -001 -MY3), the National Health Research Institutes (NHRI-109A1-EMCO-02202212), and the funding support from Academia Sinica, Taiwan, under “Trans-disciplinary PM2.5 Exposure Research in Urban Areas for Health-oriented Preventive Strategies (II)” Project No.: AS-SS-110-02. The authors are grateful to the National Aeronautics and Space Administration (NASA) and to the U.S. Geological Survey (USGS) for remote sensing data supports. The authors would like to thank the members (including Yu-Ting Zheng, Fang-Tzu Hsu, and Jennieveive Babaan) at Geomatics and Environmental Health Laboratory for validating the model development analysis. Conceptualization, P.Y.W. C.D.W. and S.C.C.L.; methodology, P.Y.W. and C.D.W.; formal analysis, P.Y.W.; writing—original draft preparation, P.Y.W. and C.D.W.; writing—review and editing, P.Y.W. H.J.S. and C.D.W.; resources, H.J.S. and C.D.W.; supervision, H.J.S. and S.C.C.L.; funding acquisition, C.D.W. and S.C.C.L.
Funding Information:
This study was grant supported by the Ministry of Science and Technology, Taiwan ( MOST 111-2121-M-006-004- ; MOST 110-2628-M-006 -001 -MY3 ), the National Health Research Institutes ( NHRI-109A1-EMCO-02202212 ), and the funding support from Academia Sinica , Taiwan, under “Trans-disciplinary PM 2.5 Exposure Research in Urban Areas for Health-oriented Preventive Strategies (II)” Project No.: AS-SS-110-02 . The authors are grateful to the National Aeronautics and Space Administration (NASA) and to the U.S. Geological Survey (USGS) for remote sensing data supports. The authors would like to thank the members (including Yu-Ting Zheng, Fang-Tzu Hsu, and Jennieveive Babaan) at Geomatics and Environmental Health Laboratory for validating the model development analysis.
Publisher Copyright:
© 2023 Elsevier B.V.
PY - 2023/3/25
Y1 - 2023/3/25
N2 - Meteorology, human activities, and other emission sources drive diurnal cyclic patterns of air pollution. Previous studies mainly focused on the variation of PM2.5 concentrations during daytime rather than nighttime. In addition, assessing the spatial variations of PM2.5 in large areas is a critical issue for environmental epidemiological studies to clarify the health effects from PM2.5 exposures. In terms of air pollution spatial modelling, using only a single model might lose information in capturing spatial and temporal correlation between predictors and pollutant levels. Hence, this study aimed to propose an ensemble mixed spatial model that incorporated Kriging interpolation, land-use regression (LUR), machine learning, and stacking ensemble approach to estimate long-term PM2.5 variations for nearly three decades in daytime and nighttime. Three steps of model development were applied: 1) linear based LUR and Hybrid Kriging-LUR were used to determine influential predictors; 2) machine learning algorithms were used to enhance model prediction accuracy; 3) predictions from the selected machine learning models were fitted and evaluated again to build the final ensemble mixed spatial model. The results showed that prediction performance increased from 0.514 to 0.895 for daily, 0.478 to 0.879 for daytime, and 0.523 to 0.878 for nighttime when applying the proposed ensemble mixed spatial model compared with LUR. Results of overfitting test and extrapolation ability test confirmed the robustness and reliability of the developed models. The distance to the nearest thermal power plant, density of soil and pebbles fields, and funeral facilities might affect the variation of PM2.5 levels between daytime and nighttime. The PM2.5 level was higher in daytime compared with nighttime with little difference, revealing the importance of estimating nighttime PM2.5 variations. Our findings also clarified the emission sources in daytime and nighttime, which serve as valuable information for air pollution control strategies establishment.
AB - Meteorology, human activities, and other emission sources drive diurnal cyclic patterns of air pollution. Previous studies mainly focused on the variation of PM2.5 concentrations during daytime rather than nighttime. In addition, assessing the spatial variations of PM2.5 in large areas is a critical issue for environmental epidemiological studies to clarify the health effects from PM2.5 exposures. In terms of air pollution spatial modelling, using only a single model might lose information in capturing spatial and temporal correlation between predictors and pollutant levels. Hence, this study aimed to propose an ensemble mixed spatial model that incorporated Kriging interpolation, land-use regression (LUR), machine learning, and stacking ensemble approach to estimate long-term PM2.5 variations for nearly three decades in daytime and nighttime. Three steps of model development were applied: 1) linear based LUR and Hybrid Kriging-LUR were used to determine influential predictors; 2) machine learning algorithms were used to enhance model prediction accuracy; 3) predictions from the selected machine learning models were fitted and evaluated again to build the final ensemble mixed spatial model. The results showed that prediction performance increased from 0.514 to 0.895 for daily, 0.478 to 0.879 for daytime, and 0.523 to 0.878 for nighttime when applying the proposed ensemble mixed spatial model compared with LUR. Results of overfitting test and extrapolation ability test confirmed the robustness and reliability of the developed models. The distance to the nearest thermal power plant, density of soil and pebbles fields, and funeral facilities might affect the variation of PM2.5 levels between daytime and nighttime. The PM2.5 level was higher in daytime compared with nighttime with little difference, revealing the importance of estimating nighttime PM2.5 variations. Our findings also clarified the emission sources in daytime and nighttime, which serve as valuable information for air pollution control strategies establishment.
UR - http://www.scopus.com/inward/record.url?scp=85145658663&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85145658663&partnerID=8YFLogxK
U2 - 10.1016/j.scitotenv.2022.161336
DO - 10.1016/j.scitotenv.2022.161336
M3 - Article
C2 - 36603626
AN - SCOPUS:85145658663
SN - 0048-9697
VL - 866
JO - Science of the Total Environment
JF - Science of the Total Environment
M1 - 161336
ER -