High-Spatiotemporal-Resolution Estimation of Ground-Level Ozone in China Based on Machine Learning
High concentrations of ground-level ozone (O<sub>3</sub>) pose a significant threat to human health. Obtaining high-spatiotemporal-resolution information about ground-level O<sub>3</sub> is of paramount importance for O<sub>3</sub> pollution control. However, the...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-12-01
|
| Series: | Atmosphere |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4433/15/1/34 |
| _version_ | 1797344718834630656 |
|---|---|
| author | Jiahuan Chen Heng Dong Zili Zhang Bingqian Quan Lan Luo |
| author_facet | Jiahuan Chen Heng Dong Zili Zhang Bingqian Quan Lan Luo |
| author_sort | Jiahuan Chen |
| collection | DOAJ |
| description | High concentrations of ground-level ozone (O<sub>3</sub>) pose a significant threat to human health. Obtaining high-spatiotemporal-resolution information about ground-level O<sub>3</sub> is of paramount importance for O<sub>3</sub> pollution control. However, the current monitoring methods have a lot of limitations. Ground-based monitoring falls short in providing extensive coverage, and remote sensing based on satellites is constrained by specific spectral bands, lacking sensitivity to ground-level O<sub>3</sub>. To address this issue, we combined brightness temperature data from the Himawari-8 satellite with meteorological data and ground-based station data to train four machine learning models to obtain high-spatiotemporal-resolution information about ground-level O<sub>3</sub>, including Categorical Boosting (CatBoost), eXtreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LGBM), and Random Forest (RF). Among these, the CatBoost model exhibited superior performance, achieving a ten-fold cross-validation R<sup>2</sup> of 0.8534, an RMSE of 17.735 μg/m<sup>3</sup>, and an MAE of 12.6594 μg/m<sup>3</sup>. Furthermore, all the selected feature variables in our study positively influenced the model. Subsequently, we employed the CatBoost model to estimate averaged hourly ground-level O<sub>3</sub> concentrations at a 2 km resolution. The estimation results indicate a close relationship between ground-level O<sub>3</sub> concentrations and human activities and solar radiation. |
| first_indexed | 2024-03-08T11:06:48Z |
| format | Article |
| id | doaj.art-e94431e85eaf42d59bb52df5e8e34723 |
| institution | Directory Open Access Journal |
| issn | 2073-4433 |
| language | English |
| last_indexed | 2024-03-08T11:06:48Z |
| publishDate | 2023-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Atmosphere |
| spelling | doaj.art-e94431e85eaf42d59bb52df5e8e347232024-01-26T15:01:39ZengMDPI AGAtmosphere2073-44332023-12-011513410.3390/atmos15010034High-Spatiotemporal-Resolution Estimation of Ground-Level Ozone in China Based on Machine LearningJiahuan Chen0Heng Dong1Zili Zhang2Bingqian Quan3Lan Luo4School of Resources and Environment Engineering, Wuhan University of Technology, Wuhan 430070, ChinaSchool of Resources and Environment Engineering, Wuhan University of Technology, Wuhan 430070, ChinaEcological Environment Monitoring Center of Zhejiang, Hangzhou 310012, ChinaEcological Environment Monitoring Center of Zhejiang, Hangzhou 310012, ChinaZhejiang Key Laboratory of Ecological and Environmental Big Data (2022P10005), Zhejiang Ecological and Environmental Monitoring Center, Hangzhou 310012, ChinaHigh concentrations of ground-level ozone (O<sub>3</sub>) pose a significant threat to human health. Obtaining high-spatiotemporal-resolution information about ground-level O<sub>3</sub> is of paramount importance for O<sub>3</sub> pollution control. However, the current monitoring methods have a lot of limitations. Ground-based monitoring falls short in providing extensive coverage, and remote sensing based on satellites is constrained by specific spectral bands, lacking sensitivity to ground-level O<sub>3</sub>. To address this issue, we combined brightness temperature data from the Himawari-8 satellite with meteorological data and ground-based station data to train four machine learning models to obtain high-spatiotemporal-resolution information about ground-level O<sub>3</sub>, including Categorical Boosting (CatBoost), eXtreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LGBM), and Random Forest (RF). Among these, the CatBoost model exhibited superior performance, achieving a ten-fold cross-validation R<sup>2</sup> of 0.8534, an RMSE of 17.735 μg/m<sup>3</sup>, and an MAE of 12.6594 μg/m<sup>3</sup>. Furthermore, all the selected feature variables in our study positively influenced the model. Subsequently, we employed the CatBoost model to estimate averaged hourly ground-level O<sub>3</sub> concentrations at a 2 km resolution. The estimation results indicate a close relationship between ground-level O<sub>3</sub> concentrations and human activities and solar radiation.https://www.mdpi.com/2073-4433/15/1/34ground-level ozonehigh-spatiotemporal-resolutionmachine learning |
| spellingShingle | Jiahuan Chen Heng Dong Zili Zhang Bingqian Quan Lan Luo High-Spatiotemporal-Resolution Estimation of Ground-Level Ozone in China Based on Machine Learning Atmosphere ground-level ozone high-spatiotemporal-resolution machine learning |
| title | High-Spatiotemporal-Resolution Estimation of Ground-Level Ozone in China Based on Machine Learning |
| title_full | High-Spatiotemporal-Resolution Estimation of Ground-Level Ozone in China Based on Machine Learning |
| title_fullStr | High-Spatiotemporal-Resolution Estimation of Ground-Level Ozone in China Based on Machine Learning |
| title_full_unstemmed | High-Spatiotemporal-Resolution Estimation of Ground-Level Ozone in China Based on Machine Learning |
| title_short | High-Spatiotemporal-Resolution Estimation of Ground-Level Ozone in China Based on Machine Learning |
| title_sort | high spatiotemporal resolution estimation of ground level ozone in china based on machine learning |
| topic | ground-level ozone high-spatiotemporal-resolution machine learning |
| url | https://www.mdpi.com/2073-4433/15/1/34 |
| work_keys_str_mv | AT jiahuanchen highspatiotemporalresolutionestimationofgroundlevelozoneinchinabasedonmachinelearning AT hengdong highspatiotemporalresolutionestimationofgroundlevelozoneinchinabasedonmachinelearning AT zilizhang highspatiotemporalresolutionestimationofgroundlevelozoneinchinabasedonmachinelearning AT bingqianquan highspatiotemporalresolutionestimationofgroundlevelozoneinchinabasedonmachinelearning AT lanluo highspatiotemporalresolutionestimationofgroundlevelozoneinchinabasedonmachinelearning |