Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Mar 16;13(3):217.
doi: 10.3390/toxics13030217.

Assessing the Impact of Aviation Emissions on Air Quality at a Regional Greek Airport Using Machine Learning

Christos Stefanis  1 , Ioannis Manisalidis  1   2 , Elisavet Stavropoulou  1 , Agathangelos Stavropoulos  3 , Christina Tsigalou  1 , Chrysoula Chrysa Voidarou  4 , Theodoros C Constantinidis  1 , Eugenia Bezirtzoglou  1
Affiliations

Assessing the Impact of Aviation Emissions on Air Quality at a Regional Greek Airport Using Machine Learning

Christos Stefanis et al. Toxics. .

Abstract

Aviation emissions significantly impact air quality, contributing to environmental degradation and public health risks. This study aims to assess the impact of aviation-related emissions on air quality at Alexandroupolis Regional Airport, Greece, and evaluate the role of meteorological factors in pollution dispersion. Using machine learning models, we analyzed emissions data, including CO2, NOx, CO, HC, SOx, PM2.5, fuel consumption, and meteorological parameters from 2019-2020. Results indicate that NOx and CO2 emissions showed the highest correlation with air traffic volume and fuel consumption (R = 0.63 and 0.67, respectively). Bayesian Linear Regression and Linear Regression emerged as the most accurate models, achieving an R2 value of 0.96 and 0.97, respectively, for predicting PM2.5 concentrations. Meteorological factors had a moderate influence, with precipitation negatively correlated with PM2.5 (-0.03), while temperature and wind speed showed limited effects on emissions. A significant decline in aviation emissions was observed in 2020, with CO2 emissions decreasing by 28.1%, NOx by 26.5%, and PM2.5 by 35.4% compared to 2019, reflecting the impact of COVID-19 travel restrictions. Carbon dioxide had the most extensive percentage distribution, accounting for 75.5% of total emissions, followed by fuels, which accounted for 24%, and the remaining pollutants, such as NOx, CO, HC, SOx, and PM2.5, had more minor impacts. These findings highlight the need for optimized air quality management at regional airports, integrating machine learning for predictive monitoring and supporting policy interventions to mitigate aviation-related pollution.

Keywords: air pollution; airport; environment; gas emission; health; machine learning models; public health; transport.

PubMed Disclaimer

Conflict of interest statement

Author Ioannis Manisalidis was employed by the company Delphis S.A. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures

Figure 1
Figure 1
Study area—Alexandroupolis airport, Greece.
Figure 2
Figure 2
Machine learning as a subfield of Artificial Intelligence.
Figure 3
Figure 3
Research workflow of the proposed methodology.
Figure 4
Figure 4
(a) Fleet composition in Alexandropoulis airport—2019. (b) Fleet composition in Alexandropoulis airport—2020. Abbreviations: A320—Airbus A320, A319—Airbus A319, AT43—ATR 42-300/320, AT45—ATR 42-500, AT72—ATR 72-200/500, DH8D—De Havilland Canada Dash 8 Q400.
Figure 5
Figure 5
Percentage contribution of pollutants in 2019 and 2020.
Figure 6
Figure 6
Comparison of monthly emissions for various pollutants in 2019 and 2020: CO; CO2; HC; NOx; PM2.5; SOx; fuel.
Figure 7
Figure 7
Heatmap of Pearson correlations between meteorological and pollutant variables. ** Correlation is significant at the 0.01 level (two-tailed), * moderate correlation, *** strong correlation.
Figure 8
Figure 8
Evaluation of the regression machine learning algorithms and the performance heat map.
Figure 9
Figure 9
Feature importance analysis for PM2.5 prediction.
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Manisalidis I., Stavropoulou E., Stavropoulos A., Bezirtzoglou E. Environmental and Health Impacts of Air Pollution: A Review. Front. Public Health. 2020;8:14. doi: 10.3389/fpubh.2020.00014. - DOI - PMC - PubMed
    1. Liu J., Attari M.U., Akhtar M., Amin M., Yu Z., Janjua L. Investigating the Impact of Transport Services and Renewable Energy on Macro-Economic and Environmental Indicators. Front. Environ. Sci. 2022;10:916176. doi: 10.3389/fenvs.2022.916176. - DOI
    1. Di Q., Dai L., Wang Y., Zanobetti A., Choirat C., Schwartz J.D., Dominici F. Association of Short-term Exposure to Air Pollution with Mortality in Older Adults. JAMA. 2017;318:2446–2456. doi: 10.1001/jama.2017.17923. - DOI - PMC - PubMed
    1. European Environment Agency Air Pollution. [(accessed on 21 September 2022)]. Available online: https://www.eea.europa.eu/themes/air.
    1. Cromar K.R., Gladson L.A., Hicks E.A., Marsh B., Ewart G. Excess Morbidity and Mortality Associated with Air Pollution above American Thoracic Society Recommended Standards, 2017–2019. Ann. Am. Thorac. Soc. 2022;19:603–613. doi: 10.1513/AnnalsATS.202107-860OC. - DOI - PubMed

LinkOut - more resources