EUVIMEDEUVIMED entdecken
Crossref

Machine-learning-based estimation of ground-level NO2 concentrations across Southeastern Europe using multi-source satellite, reanalysis, and emission data

Efem Bilgiç, Tolga Elbir

Environmental Pollution · 2026

DOI: 10.1016/j.envpol.2026.128461 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Environmental Pollution, 2026. Verfasst von Efem Bilgiç und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Environmental Engineering, Faculty of Engineering, Dokuz Eylul University, Izmir, Türkiye. Electronic address: efem.bilgic@deu.edu.tr. Hauptthemen laut MEDLINE: Air Pollutants, Boosting Machine Learning Algorithms, Environmental Monitoring, Nitrogen Dioxide. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 53 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Reliable estimation of ground-level nitrogen dioxide (NO2) remains challenging due to limited monitoring coverage and complex interactions among emissions, meteorology, and land-use factors. This study develops a machine-learning framework to estimate surface NO2 concentrations across Southeastern Europe, covering thirteen countries with diverse emission sources and sparse monitoring networks. Daily NO2 observations from 317 monitoring stations were integrated with multiple satellite and ancillary datasets, including tropospheric NO2 from Sentinel-5P TROPOMI, vegetation indices from MODIS, and nighttime light data from VIIRS. Additional inputs included meteorological variables from ERA5, anthropogenic emissions from EDGAR v8.1, land-use data from Copernicus Corine Land Cover, elevation from the Copernicus Digital Elevation Model, and population density from the Global Human Settlement Layer. All datasets were processed and harmonized using Google Earth Engine. Three Gradient Boosting Decision Tree models (XGBoost, LGBM, and CatBoost) were trained and optimized with Optuna. Model performance was assessed using 10-fold cross-validation and independent test sets with R2, MAE, and RMSE metrics. All models performed consistently, with LGBM achieving the best results (R2 = 0.86; MAE = 5.34 μg/m3; RMSE = 8.32 μg/m3). SHapley Additive exPlanations (SHAP) identified key predictors and improved interpretability. Seasonal analysis showed higher accuracy in summer and autumn than in winter and spring. These results demonstrate that integrating multi-source data with advanced machine learning enables reliable surface NO2 estimation in regions with limited monitoring, supporting air quality assessment, exposure analysis, and evidence-based policymaking in Southeastern Europe.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Air PollutantsBoosting Machine Learning AlgorithmsEnvironmental MonitoringNitrogen DioxideAir PollutionEuropeSatellite Imagery

Stichwörter der Autor:innen

Air quality modelingMachine learningNitrogen dioxideSatellite data integrationSoutheastern Europe

Substanzen

Air PollutantsNitrogen Dioxide

Autor:innen und Institutionen

  • Department of Environmental Engineering, Faculty of Engineering, Dokuz Eylul University, Izmir, Türkiye. Electronic address: efem.bilgic@deu.edu.tr.
  • Department of Environmental Engineering, Faculty of Engineering, Dokuz Eylul University, Izmir, Türkiye; Environmental Research and Application Center (CEVMER), Dokuz Eylul University, Izmir, Türkiye.

Zitieren

APA

Bilgiç, E., & Elbir, T. (2026). Machine-learning-based estimation of ground-level NO2 concentrations across Southeastern Europe using multi-source satellite, reanalysis, and emission data. Environmental Pollution. https://doi.org/10.1016/j.envpol.2026.128461

Vancouver

Bilgiç E, Elbir T. Machine-learning-based estimation of ground-level NO2 concentrations across Southeastern Europe using multi-source satellite, reanalysis, and emission data. Environmental Pollution. 2026. doi:10.1016/j.envpol.2026.128461

BibTeX
@article{Bilgi2026,
  author = {Bilgiç, Efem and Elbir, Tolga},
  title = {Machine-learning-based estimation of ground-level NO2 concentrations across Southeastern Europe using multi-source satellite, reanalysis, and emission data},
  journal = {Environmental Pollution},
  year = {2026},
  doi = {10.1016/j.envpol.2026.128461}
}
Kostenlos starten →