Source apportionment of PM10 particles in the urban atmosphere using PMF and LPO-XGBoost

Atmospheric particulate matter (PM), as a leading part of air pollution, affects health in many ways. Thus, identifying and quantifying the contribution of atmospheric particulate matter sources of PM is vital for developing effective air quality management strategies. Positive Matrix Factorization (PMF) is one of the most common methods for source apportionment. However, PMF has some limitations, particularly its assumption that each source contributes linearly. In reality, some sources may exhibit nonlinear behaviors, which can compromise the accuracy of source apportionment. This study introduces a Lung Performance Optimization-based XGBoost (LPO-XGBoost) model, which leverages adaptive optimization principles inspired by lung function to enhance classic PM source apportionment. We demonstrate the potential for efficient, real-time application of the LPO-XGBoost model across 21 monitoring sites in 6 European countries. Trained and validated on extensive environmental datasets, the model is capable of predicting major pollution sources, including road traffic, biomass burning, crustal, industrial, nitrate-rich particles, sulfate-rich particles, heavy fuel oil, and sea salt. It outperforms other machine learning models with an overall predictive coefficient of determination (r<sup>2</sup> = 0.88). Notably, the model performs exceptionally well in predicting sources such as sea salt (r<sup>2</sup> = 0.97) and biomass burning (r<sup>2</sup> = 0.89), but shows lower accuracy for the sulfate-rich particles source (r<sup>2</sup> = 0.75). Comparative analyses with models including Random Forest (RF), Support Vector Machine (SVM), and their LPO-enhanced variants confirm that LPO-XGBoost provides the most reliable performance in estimating pollution source contributions, offering scalability and robustness ideal for high-time-resolution observational data. This model has significant potential to support targeted air quality management strategies. Future research should focus on expanding key species measurements at monitoring sites, ensuring consistent temporal coverage, and optimizing the model for improved mixed-source predictions to strengthen its applicability in comprehensive urban air quality assessments.

Authors

Institutions

Publication Details

Journal
Environmental Research
Published
2025-04-22
DOI
https://doi.org/10.1016/j.envres.2025.121659
Citations
9
Primary Topic
Air Quality Monitoring and Forecasting
Type
article
Field-Weighted Citation Impact
7.19

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Source apportionment of PM10 particles in the urban atmosphere using PMF and LPO-XGBoost

Olivier Favez, Sébastien Conil, Christoph Hueglin, Manousos Ioannis Manousakas et al.
9 citations
Environmental Research
Air Quality Monitoring and Forecasting
7.19
article

Source apportionment of PM10 particles in the urban atmosphere using PMF and LPO-XGBoost

Olivier Favez, Sébastien Conil, Christoph Hueglin, Manousos Ioannis Manousakas, Maria I. Gini, Giulia Calzolai, Andrés Alástuey, Roy M. Harrison, Sophie Darfeuil, Sonia Castillo, Gaëlle Uzu, Xavier Querol, Jean‐Luc Jaffrezo, Cristina Reche, Stuart K. Grange, Nicolas Marchand, Konstantinos Eleftheriadis, Jesús de la Rosa, Vy Dinh Ngoc Thuy, Philip Karl Hopke, Evangelia Diapouli, Célia Alves, Xiansheng Liu, Pamela Dominutti, Ying Liu, Tao Wang, Marta Monge, Bowen Jin, Xun Zhang
article en
9 citations

Abstract

Atmospheric particulate matter (PM), as a leading part of air pollution, affects health in many ways. Thus, identifying and quantifying the contribution of atmospheric particulate matter sources of PM is vital for developing effective air quality management strategies. Positive Matrix Factorization (PMF) is one of the most common methods for source apportionment. However, PMF has some limitations, particularly its assumption that each source contributes linearly. In reality, some sources may exhibit nonlinear behaviors, which can compromise the accuracy of source apportionment. This study introduces a Lung Performance Optimization-based XGBoost (LPO-XGBoost) model, which leverages adaptive optimization principles inspired by lung function to enhance classic PM source apportionment. We demonstrate the potential for efficient, real-time application of the LPO-XGBoost model across 21 monitoring sites in 6 European countries. Trained and validated on extensive environmental datasets, the model is capable of predicting major pollution sources, including road traffic, biomass burning, crustal, industrial, nitrate-rich particles, sulfate-rich particles, heavy fuel oil, and sea salt. It outperforms other machine learning models with an overall predictive coefficient of determination (r<sup>2</sup> = 0.88). Notably, the model performs exceptionally well in predicting sources such as sea salt (r<sup>2</sup> = 0.97) and biomass burning (r<sup>2</sup> = 0.89), but shows lower accuracy for the sulfate-rich particles source (r<sup>2</sup> = 0.75). Comparative analyses with models including Random Forest (RF), Support Vector Machine (SVM), and their LPO-enhanced variants confirm that LPO-XGBoost provides the most reliable performance in estimating pollution source contributions, offering scalability and robustness ideal for high-time-resolution observational data. This model has significant potential to support targeted air quality management strategies. Future research should focus on expanding key species measurements at monitoring sites, ensuring consistent temporal coverage, and optimizing the model for improved mixed-source predictions to strengthen its applicability in comprehensive urban air quality assessments.

Environmental ResearchVol. 278
Institut polytechnique de Grenoble (FR), Centre National de la Recherche Scientifique (FR), Guangdong University of Technology (CN), Clarkson University (US), Universidad de Granada (ES), Beijing Technology and Business University (CN), National Centre of Scientific Research "Demokritos" (GR), Fudan University (CN), Institut National de Recherche pour l'Agriculture, l'Alimentation et l'Environnement (FR), Instituto Andaluz de Ciencias de la Tierra (ES), Beijing Academy of Artificial Intelligence (CN), Institut des Géosciences de l'Environnement (FR), Institut national de l'environnement industriel et des risques (FR), Istituto Nazionale di Fisica Nucleare, Sezione di Firenze (IT), Agence Nationale pour la Gestion des Déchets Radioactifs (FR), Institute of Environmental Assessment and Water Research (ES), Beijing Municipal Education Commission (CN), Institut de Recherche pour le Développement (FR), University of Florence (IT), University of Aveiro (PT), Swiss Federal Laboratories for Materials Science and Technology (CH), Universidad de Huelva (ES), University of Birmingham (GB), Université Grenoble Alpes (FR)
Universidade de Aveiro, European Commission, National Natural Science Foundation of China, Ministry of Environment, Science Fund for Creative Research Groups, Ministerio de Ciencia e Innovación, Universidad de Granada, State Key Laboratory of Resources and Environmental Information System, Région Normandie, Centro de Estudos Ambientais e Marinhos, Universidade de Aveiro, Fundação para a Ciência e a Tecnologia, Bundesamt für Umwelt
Sustainable cities and communities
Openalex Percentile: Top 3%
Air Quality Monitoring and Forecasting
7.19
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.