Long-Term (2015–2024) Daily PM<sub>2.5</sub> Estimation in China by Using XGBoost Combining Empirical Orthogonal Function Decomposition

Fine particulate matter (PM<sub>2.5</sub>) has garnered significant scientific and public health concern owing to its capacity for deep penetration into the human respiratory system, presenting significant health risks. Despite the implementation of strict environmental policies in China...

Full description

Saved in:
Bibliographic Details
Main Authors: Jiacheng Jiang, Jiaxin Dong, Yu Ding, Wenjia Ni, Jie Yang, Siwei Li
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/17/9/1632
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850030310990807040
author Jiacheng Jiang
Jiaxin Dong
Yu Ding
Wenjia Ni
Jie Yang
Siwei Li
author_facet Jiacheng Jiang
Jiaxin Dong
Yu Ding
Wenjia Ni
Jie Yang
Siwei Li
author_sort Jiacheng Jiang
collection DOAJ
description Fine particulate matter (PM<sub>2.5</sub>) has garnered significant scientific and public health concern owing to its capacity for deep penetration into the human respiratory system, presenting significant health risks. Despite the implementation of strict environmental policies in China over the past decade to reduce PM<sub>2.5</sub> levels, long-term public health concerns remain a serious issue. Our study aims to provide a high-quality, seamless daily PM<sub>2.5</sub> dataset for China covering the years 2015 to 2024. A two-step PM<sub>2.5</sub> estimation model is established based on a machine learning algorithm and a spatio-temporal decomposition method. First, we utilize the machine learning algorithm XGBoost (EXtreme Gradient Boosting) to address gaps in the daily MAIAC (Multi-Angle Implementation of Atmospheric Correction) AOD (Aerosol Optical Depth), with R<sup>2</sup>/RMSE (coefficient of determination/Root Mean Square Error) of 0.67/0.2678 compared to AERONET (Aerosol Robotic Network) AOD. Then, a novel approach by integrating XGBoost with EOF (Empirical Orthogonal Function) decomposition is introduced for PM<sub>2.5</sub> estimation. The integration of EOF allows for the incorporation of entire meteorological field information into the PM<sub>2.5</sub> estimation model, significantly enhancing its accuracy: spatial CV (cross-validation)-R<sup>2</sup> improved from 0.8340 to 0.8935, and spatial CV-RMSE reduced from 13.8177 to 11.0668. Leveraging the newly produced dataset, we analyze the spatio-temporal variations of PM<sub>2.5</sub> across China with EOF decomposition, particularly noting that PM<sub>2.5</sub> levels in the eastern anthropogenic intensive regions continuously declined from 2015 to 2020, and fluctuated steadily during 2020–2024. This research underscores the critical need for sustained and effective air quality management strategies in China.
format Article
id doaj-art-e410c92d24da4b449e6b88e35cc37861
institution DOAJ
issn 2072-4292
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-e410c92d24da4b449e6b88e35cc378612025-08-20T02:59:15ZengMDPI AGRemote Sensing2072-42922025-05-01179163210.3390/rs17091632Long-Term (2015–2024) Daily PM<sub>2.5</sub> Estimation in China by Using XGBoost Combining Empirical Orthogonal Function DecompositionJiacheng Jiang0Jiaxin Dong1Yu Ding2Wenjia Ni3Jie Yang4Siwei Li5Hubei Key Laboratory of Quantitative Remote Sensing of Land and Atmosphere, School of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, ChinaHubei Key Laboratory of Quantitative Remote Sensing of Land and Atmosphere, School of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, ChinaHubei Key Laboratory of Quantitative Remote Sensing of Land and Atmosphere, School of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, ChinaHubei Key Laboratory of Quantitative Remote Sensing of Land and Atmosphere, School of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, ChinaPerception and Effectiveness Assessment for Carbon-Neutrality Efforts, Engineering Research Center of Ministry of Education, Institute for Carbon Neutrality, Wuhan University, Wuhan 430072, ChinaHubei Key Laboratory of Quantitative Remote Sensing of Land and Atmosphere, School of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, ChinaFine particulate matter (PM<sub>2.5</sub>) has garnered significant scientific and public health concern owing to its capacity for deep penetration into the human respiratory system, presenting significant health risks. Despite the implementation of strict environmental policies in China over the past decade to reduce PM<sub>2.5</sub> levels, long-term public health concerns remain a serious issue. Our study aims to provide a high-quality, seamless daily PM<sub>2.5</sub> dataset for China covering the years 2015 to 2024. A two-step PM<sub>2.5</sub> estimation model is established based on a machine learning algorithm and a spatio-temporal decomposition method. First, we utilize the machine learning algorithm XGBoost (EXtreme Gradient Boosting) to address gaps in the daily MAIAC (Multi-Angle Implementation of Atmospheric Correction) AOD (Aerosol Optical Depth), with R<sup>2</sup>/RMSE (coefficient of determination/Root Mean Square Error) of 0.67/0.2678 compared to AERONET (Aerosol Robotic Network) AOD. Then, a novel approach by integrating XGBoost with EOF (Empirical Orthogonal Function) decomposition is introduced for PM<sub>2.5</sub> estimation. The integration of EOF allows for the incorporation of entire meteorological field information into the PM<sub>2.5</sub> estimation model, significantly enhancing its accuracy: spatial CV (cross-validation)-R<sup>2</sup> improved from 0.8340 to 0.8935, and spatial CV-RMSE reduced from 13.8177 to 11.0668. Leveraging the newly produced dataset, we analyze the spatio-temporal variations of PM<sub>2.5</sub> across China with EOF decomposition, particularly noting that PM<sub>2.5</sub> levels in the eastern anthropogenic intensive regions continuously declined from 2015 to 2020, and fluctuated steadily during 2020–2024. This research underscores the critical need for sustained and effective air quality management strategies in China.https://www.mdpi.com/2072-4292/17/9/1632XGBoostEOF decompositionPM<sub>2.5</sub>ChinaAOD
spellingShingle Jiacheng Jiang
Jiaxin Dong
Yu Ding
Wenjia Ni
Jie Yang
Siwei Li
Long-Term (2015–2024) Daily PM<sub>2.5</sub> Estimation in China by Using XGBoost Combining Empirical Orthogonal Function Decomposition
Remote Sensing
XGBoost
EOF decomposition
PM<sub>2.5</sub>
China
AOD
title Long-Term (2015–2024) Daily PM<sub>2.5</sub> Estimation in China by Using XGBoost Combining Empirical Orthogonal Function Decomposition
title_full Long-Term (2015–2024) Daily PM<sub>2.5</sub> Estimation in China by Using XGBoost Combining Empirical Orthogonal Function Decomposition
title_fullStr Long-Term (2015–2024) Daily PM<sub>2.5</sub> Estimation in China by Using XGBoost Combining Empirical Orthogonal Function Decomposition
title_full_unstemmed Long-Term (2015–2024) Daily PM<sub>2.5</sub> Estimation in China by Using XGBoost Combining Empirical Orthogonal Function Decomposition
title_short Long-Term (2015–2024) Daily PM<sub>2.5</sub> Estimation in China by Using XGBoost Combining Empirical Orthogonal Function Decomposition
title_sort long term 2015 2024 daily pm sub 2 5 sub estimation in china by using xgboost combining empirical orthogonal function decomposition
topic XGBoost
EOF decomposition
PM<sub>2.5</sub>
China
AOD
url https://www.mdpi.com/2072-4292/17/9/1632
work_keys_str_mv AT jiachengjiang longterm20152024dailypmsub25subestimationinchinabyusingxgboostcombiningempiricalorthogonalfunctiondecomposition
AT jiaxindong longterm20152024dailypmsub25subestimationinchinabyusingxgboostcombiningempiricalorthogonalfunctiondecomposition
AT yuding longterm20152024dailypmsub25subestimationinchinabyusingxgboostcombiningempiricalorthogonalfunctiondecomposition
AT wenjiani longterm20152024dailypmsub25subestimationinchinabyusingxgboostcombiningempiricalorthogonalfunctiondecomposition
AT jieyang longterm20152024dailypmsub25subestimationinchinabyusingxgboostcombiningempiricalorthogonalfunctiondecomposition
AT siweili longterm20152024dailypmsub25subestimationinchinabyusingxgboostcombiningempiricalorthogonalfunctiondecomposition