PuSH - Publication Server of Helmholtz Zentrum München: Multi-pollutant air quality assessment around urban schools using machine learning.

Information

Clues to quality of journals

Open Access Policy of Helmholtz Association 2016

CC Licencences

Publication Metrics

Navigation

Home

Deutsch

EVA: Electronic Publishing

New EVA Application

Research

Advanced Search

Browse by ...

... HMGU-Authors/Consortia

... Organizational Structure

... Journal

... Publication Type

... Research Data

... Arbeitsgruppen

... Publication Year

Publication overview

Statistics

Statistics (last 5 years)

OA Publications

Publish

Submit Publication

Import publication from...

...EVA

Report missing publication

Highlights

Support & Contact

Contact persons

Help

Data protection

Helmholtz Open Science

JANE Journal Estimator

SHERPA/RoMEO

DOAJ

Export:

Text

Endnote (RIS) BIB

BibTeX

Davtalab, M.* ; Davulienė, L.* ; Uogintė, I.* ; Kecorius, S. ; Lovrić, M.* ; Byčenkiene, S.*

Multi-pollutant air quality assessment around urban schools using machine learning.

Urban Climate 62:102567 (2025)

DOI

Abstract
Metrics
Extra information

Air pollution poses a significant threat to global public health, particularly in sensitive areas like schools. Reliable air quality assessment is critical to inform effective policymaking and protect vulnerable populations. However, accurately assessing air quality in cities with limited monitoring networks remains a significant challenge. This study bridges this gap by integrating satellite-derived columnar data for nitrogen dioxide (NO2), sulfur dioxide (SO2), and carbon monoxide (CO) with machine learning (ML) techniques—random forest (RF) and gradient boosting machine (GBM)—to develop spatial multi-pollutant classification maps for air quality assessment around schools in Vilnius, Lithuania, where ground-based monitoring stations are limited. The results showed that between the two ML models, RF demonstrated better performance than GBM, achieving an accuracy of 0.900, precision of 0.895, recall of 0.897, and an F1 score of 0.896. To enhance prediction accuracy, the models incorporate meteorological variables (e.g., temperature, wind speed, humidity) alongside urban characteristics (e.g., building density, road density, and proximity to road networks). Comparison with surface-level NO2, SO2, and CO concentrations reveals the model's capacity to capture pollution patterns, particularly in the city center and densely built urban areas. The findings indicate that 37 % of schools are situated within 100–250 m of major roads, where NO2 + CO and NO2 + CO + SO2 are the dominant columnar pollution classes. The results of the spatial analysis reveal that these schools are mostly in the city center primarily due to higher building and road densities, as well as lower levels of greenness.

Impact Factor

Scopus SNIP

Altmetric

0.000

1.540