English
 
Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT

Released

Conference Paper

An unsupervised machine learning-based methodology to study the seasonal dispersal pathways of short-lived pollutants from Major Population Centers

Authors

Poulidis,  Alexandros
IUGG 2023, General Assemblies, 1 General, International Union of Geodesy and Geophysics (IUGG), External Organizations;

Daskalakis,  Nikos
IUGG 2023, General Assemblies, 1 General, International Union of Geodesy and Geophysics (IUGG), External Organizations;

Kanakidou,  Maria
IUGG 2023, General Assemblies, 1 General, International Union of Geodesy and Geophysics (IUGG), External Organizations;

Vrekoussis,  Mihalis
IUGG 2023, General Assemblies, 1 General, International Union of Geodesy and Geophysics (IUGG), External Organizations;

External Ressource
No external resources are shared
Fulltext (public)
There are no public fulltexts stored in GFZpublic
Supplementary Material (public)
There is no public supplementary material available
Citation

Poulidis, A., Daskalakis, N., Kanakidou, M., Vrekoussis, M. (2023): An unsupervised machine learning-based methodology to study the seasonal dispersal pathways of short-lived pollutants from Major Population Centers, XXVIII General Assembly of the International Union of Geodesy and Geophysics (IUGG) (Berlin 2023).
https://doi.org/10.57757/IUGG23-3643


Cite as: https://gfzpublic.gfz-potsdam.de/pubman/item/item_5020899
Abstract
The rapid increase in human population since 1900 has occurred along with a rapid increase in urbanization - the process of migration from rural into urban areas. The fossil fuel combustion-based emissions affecting areas with high population densities induce a significant health risk for local populations. Protection of human health requires better knowledge of local and regional impacts of urban pollution dispersal. This study addresses the issue of short-lived pollutant transport (e.g. NOx) by constructing a methodology to study the seasonality of common dispersal pathways from Major Population Centers (MPCs); initially applied to 5 MPCs in south America during 2018. For each city, ERA5 reanalysis data were used to drive the FLEXPART emission transport model to simulate the dispersal of near-surface emissions. Simulations are performed for a total of 8 hours from release in the early morning and in the evening, to capture the effects of commuting. A total of 100,000 trajectories per release time per city were analyzed to create representative average trajectories. The k-means clustering algorithm was then applied to categorize the emissions per MPC. Clustering for each MPC led to a robust grouping of trajectories that were seen to reflect climatological and topographic phenomena during the simulation period and exhibit strong seasonality. This finding supports potential applicability of the proposed methodology for a global analysis of MPC emissions.