Development of a predictive model for PM2.5 over the greater Athens metropolitan area, Greece, at a 1 km by 1 km grid using satellite measurements and machine learning methods
PLOS One · 06.07.2026 · 21 (7) · e0352975
DOI: 10.1371/journal.pone.0352975 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS One, online erschienen am 06.07.2026. Verfasst von Seyong Chang und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Environmental Health, Harvard T. H. Chan School of Public Health, Boston, Massachusetts, United States of America. Hauptthemen laut MEDLINE: Particulate Matter, Machine Learning, Environmental Monitoring, Air Pollutants. Die Arbeit nennt 96 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Harvard-Cyprus Endowment Program for the Environment and Public Health.
Abstract (Originalsprache)
Background
Ambient PM 2.5 exposure is strongly associated with adverse health effects, including all-cause mortality. However, the lack of monitoring networks globally necessitates a better understanding of the spatiotemporal distribution of near-surface PM 2.5 pollution. While ground-level pollutants are traditionally measured at fixed stations, integrating land use and atmospheric reanalysis data captures the broad geospatial trends and specific aerosol compositions necessary for high-resolution exposure assessment. This study aims to demonstrate the accuracy and reliability of ensemble modeling for estimating near-surface PM concentrations at a high spatiotemporal resolution in Athens, Greece.
Methods
Daily PM 2.5 concentrations were estimated using a stacked ensemble machine learning model to incorporate distributed random forest, gradient boosting, and feedforward neural network algorithms to minimize predictive error compared to individual models. The input set for all base learners consisted of daily observations from air quality monitors between 2007–2019 combined with satellite-derived estimates, providing a total of 61 variables describing regional aerosol, weather, and land use characteristics.
Results
We observed strong predictive performance in our ensemble model, with a mean R 2 of 0.85 and an average error of 4.18 μg/m 3 . The annual average concentration of PM 2.5 (22.30 μg/m 3 ) exceeded current WHO and EU guidelines, with considerable spatiotemporal variation across greater Athens. The highest annual mean PM 2.5 concentrations were in 2007 (33.18 μg/m 3 ) and on average year-to-year, PM 2.5 concentrations were highest during the mid-winter months, in agreement with the expected seasonal maximum for the region, likely driven by increased residential heating alongside winter meteorological conditions, such as temperature inversions and a shallow, stable planetary boundary layer. Significance This is among the first studies to estimate PM 2.5 exposures in the greater Athens region at a high spatiotemporal resolution using diverse satellite and land use data. This framework enables the investigation of cumulative exposures, particularly in regions with limited ground-level monitoring.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Department of Environmental Health, Harvard T. H. Chan School of Public Health, Boston, Massachusetts, United States of America.
- Department of Hygiene, Epidemiology, and Medical Statistics, School of Medicine, National and Kapodistrian University of Athens, Athens, Greece.
- Institute of Nuclear & Radiological Sciences and Technology, Energy & Safety, National Centre for Scientific Research "Demokritos", Athens, Greece.
- Department of Basic and Clinical Sciences, Medical School, University of Nicosia, UNIC Athens, Athens, Greece.
ORCID-Profile: Seyong Chang
Förderung
- Harvard-Cyprus Endowment Program for the Environment and Public Health
Zitieren
APA
Chang, S., Samoli, E., Diapouli, E., Dimakopoulou, K., Alahmad, B., & Koutrakis, P. (2026). Development of a predictive model for PM2.5 over the greater Athens metropolitan area, Greece, at a 1 km by 1 km grid using satellite measurements and machine learning methods. PLOS One, 21(7), e0352975. https://doi.org/10.1371/journal.pone.0352975
Vancouver
Chang S, Samoli E, Diapouli E, Dimakopoulou K, Alahmad B, Koutrakis P. Development of a predictive model for PM2.5 over the greater Athens metropolitan area, Greece, at a 1 km by 1 km grid using satellite measurements and machine learning methods. PLOS One. 2026;21(7):e0352975. doi:10.1371/journal.pone.0352975
BibTeX
@article{Chang2026,
author = {Chang, Seyong and Samoli, Evangelia and Diapouli, Evangelia and Dimakopoulou, Konstantina and Alahmad, Barrak and Koutrakis, Petros},
title = {Development of a predictive model for PM2.5 over the greater Athens metropolitan area, Greece, at a 1 km by 1 km grid using satellite measurements and machine learning methods},
journal = {PLOS One},
year = {2026},
volume = {21},
number = {7},
pages = {e0352975},
doi = {10.1371/journal.pone.0352975}
}
EUVIMED