Contributions of various driving factors to air pollution events: Interpretability analysis from Machine learning perspective
The air quality in China has been improved substantially, however fine particulate matter (PM2.5) still remain at a high level in many areas. PM2.5 pollution is a complex process that is attributed to gaseous precursors, chemical, and meteorological factors. Quantifying the contribution of each vari...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-03-01
|
| Series: | Environment International |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S0160412023001344 |
| _version_ | 1827987860214513664 |
|---|---|
| author | Tianshuai Li Qingzhu Zhang Yanbo Peng Xu Guan Lei Li Jiangshan Mu Xinfeng Wang Xianwei Yin Qiao Wang |
| author_facet | Tianshuai Li Qingzhu Zhang Yanbo Peng Xu Guan Lei Li Jiangshan Mu Xinfeng Wang Xianwei Yin Qiao Wang |
| author_sort | Tianshuai Li |
| collection | DOAJ |
| description | The air quality in China has been improved substantially, however fine particulate matter (PM2.5) still remain at a high level in many areas. PM2.5 pollution is a complex process that is attributed to gaseous precursors, chemical, and meteorological factors. Quantifying the contribution of each variable to air pollution can facilitate the formulation of effective policies to precisely eliminate air pollution. In this study, we first used decision plot to map out the decision process of the Random Forest (RF) model for a single hourly data set and constructed a framework for analyzing the causes of air pollution using multiple interpretable methods. Permutation importance was used to qualitatively analyze the effect of each variable on PM2.5 concentrations. The sensitivity of secondary inorganic aerosols (SIA): SO42-, NO3- and NH4+ to PM2.5 was verified by Partial dependence plot (PDP). Shapley Additive Explanation (Shapley) was used to quantify the contribution of drivers behind the ten air pollution events. The RF model can accurately predict PM2.5 concentrations, with determination coefficient (R2) of 0.94, root mean square error (RMSE) and mean absolute error (MAE) of 9.4 μg/m3 and 5.7 μg/m3, respectively. This study revealed that the order of sensitivity of SIA to PM2.5 was NH4+＞NO3-＞SO42-. Fossil fuel and biomass combustion may be contributing factors to air pollution events in Zibo in 2021 autumn–winter. NH4+ contributed 19.9–65.4 μg/m3 among ten air pollution events (APs). K, NO3-, EC and OC were the other main drivers, contributing 8.7 ± 2.7 μg/m3, 6.8 ± 7.5 μg/m3, 3.6 ± 5.8 μg/m3 and 2.5 ± 2.0 μg/m3, respectively. Lower temperature and higher humidity were vital factors that promoted the formation of NO3-. Our study may provide a methodological framework for precise air pollution management. |
| first_indexed | 2024-04-09T23:54:41Z |
| format | Article |
| id | doaj.art-73c03124437e4efd9d3d7e727bcc6142 |
| institution | Directory Open Access Journal |
| issn | 0160-4120 |
| language | English |
| last_indexed | 2024-04-09T23:54:41Z |
| publishDate | 2023-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Environment International |
| spelling | doaj.art-73c03124437e4efd9d3d7e727bcc61422023-03-17T04:32:27ZengElsevierEnvironment International0160-41202023-03-01173107861Contributions of various driving factors to air pollution events: Interpretability analysis from Machine learning perspectiveTianshuai Li0Qingzhu Zhang1Yanbo Peng2Xu Guan3Lei Li4Jiangshan Mu5Xinfeng Wang6Xianwei Yin7Qiao Wang8Big Data Research Center for Ecology and Environment, Environment Research Institute, Shandong University, Qingdao 266003, PR ChinaBig Data Research Center for Ecology and Environment, Environment Research Institute, Shandong University, Qingdao 266003, PR ChinaBig Data Research Center for Ecology and Environment, Environment Research Institute, Shandong University, Qingdao 266003, PR China; Shandong Academy for Environmental Planning, Jinan 250101, PR China; Corresponding author at: Big Data Research Center for Ecology and Environment, Environment Research Institute, Shandong University, Qingdao 266003, PR China.Shandong Academy for Environmental Planning, Jinan 250101, PR ChinaBig Data Research Center for Ecology and Environment, Environment Research Institute, Shandong University, Qingdao 266003, PR ChinaBig Data Research Center for Ecology and Environment, Environment Research Institute, Shandong University, Qingdao 266003, PR ChinaBig Data Research Center for Ecology and Environment, Environment Research Institute, Shandong University, Qingdao 266003, PR ChinaZibo Ecological Environment Monitoring Center of Shandong Province, Zibo 255040, PR ChinaBig Data Research Center for Ecology and Environment, Environment Research Institute, Shandong University, Qingdao 266003, PR ChinaThe air quality in China has been improved substantially, however fine particulate matter (PM2.5) still remain at a high level in many areas. PM2.5 pollution is a complex process that is attributed to gaseous precursors, chemical, and meteorological factors. Quantifying the contribution of each variable to air pollution can facilitate the formulation of effective policies to precisely eliminate air pollution. In this study, we first used decision plot to map out the decision process of the Random Forest (RF) model for a single hourly data set and constructed a framework for analyzing the causes of air pollution using multiple interpretable methods. Permutation importance was used to qualitatively analyze the effect of each variable on PM2.5 concentrations. The sensitivity of secondary inorganic aerosols (SIA): SO42-, NO3- and NH4+ to PM2.5 was verified by Partial dependence plot (PDP). Shapley Additive Explanation (Shapley) was used to quantify the contribution of drivers behind the ten air pollution events. The RF model can accurately predict PM2.5 concentrations, with determination coefficient (R2) of 0.94, root mean square error (RMSE) and mean absolute error (MAE) of 9.4 μg/m3 and 5.7 μg/m3, respectively. This study revealed that the order of sensitivity of SIA to PM2.5 was NH4+＞NO3-＞SO42-. Fossil fuel and biomass combustion may be contributing factors to air pollution events in Zibo in 2021 autumn–winter. NH4+ contributed 19.9–65.4 μg/m3 among ten air pollution events (APs). K, NO3-, EC and OC were the other main drivers, contributing 8.7 ± 2.7 μg/m3, 6.8 ± 7.5 μg/m3, 3.6 ± 5.8 μg/m3 and 2.5 ± 2.0 μg/m3, respectively. Lower temperature and higher humidity were vital factors that promoted the formation of NO3-. Our study may provide a methodological framework for precise air pollution management.http://www.sciencedirect.com/science/article/pii/S0160412023001344PM2.5Air pollutionMachine learningPermutation importancePDPSHAP |
| spellingShingle | Tianshuai Li Qingzhu Zhang Yanbo Peng Xu Guan Lei Li Jiangshan Mu Xinfeng Wang Xianwei Yin Qiao Wang Contributions of various driving factors to air pollution events: Interpretability analysis from Machine learning perspective Environment International PM2.5 Air pollution Machine learning Permutation importance PDP SHAP |
| title | Contributions of various driving factors to air pollution events: Interpretability analysis from Machine learning perspective |
| title_full | Contributions of various driving factors to air pollution events: Interpretability analysis from Machine learning perspective |
| title_fullStr | Contributions of various driving factors to air pollution events: Interpretability analysis from Machine learning perspective |
| title_full_unstemmed | Contributions of various driving factors to air pollution events: Interpretability analysis from Machine learning perspective |
| title_short | Contributions of various driving factors to air pollution events: Interpretability analysis from Machine learning perspective |
| title_sort | contributions of various driving factors to air pollution events interpretability analysis from machine learning perspective |
| topic | PM2.5 Air pollution Machine learning Permutation importance PDP SHAP |
| url | http://www.sciencedirect.com/science/article/pii/S0160412023001344 |
| work_keys_str_mv | AT tianshuaili contributionsofvariousdrivingfactorstoairpollutioneventsinterpretabilityanalysisfrommachinelearningperspective AT qingzhuzhang contributionsofvariousdrivingfactorstoairpollutioneventsinterpretabilityanalysisfrommachinelearningperspective AT yanbopeng contributionsofvariousdrivingfactorstoairpollutioneventsinterpretabilityanalysisfrommachinelearningperspective AT xuguan contributionsofvariousdrivingfactorstoairpollutioneventsinterpretabilityanalysisfrommachinelearningperspective AT leili contributionsofvariousdrivingfactorstoairpollutioneventsinterpretabilityanalysisfrommachinelearningperspective AT jiangshanmu contributionsofvariousdrivingfactorstoairpollutioneventsinterpretabilityanalysisfrommachinelearningperspective AT xinfengwang contributionsofvariousdrivingfactorstoairpollutioneventsinterpretabilityanalysisfrommachinelearningperspective AT xianweiyin contributionsofvariousdrivingfactorstoairpollutioneventsinterpretabilityanalysisfrommachinelearningperspective AT qiaowang contributionsofvariousdrivingfactorstoairpollutioneventsinterpretabilityanalysisfrommachinelearningperspective |