- Tytuł:
- Fusing multiple open-source remote sensing data to estimate $\text{PM}_\text{2.5}$ and $\text{PM}_\text{10}$ monthly concentrations in Croatia
- Autorzy:
-
Mamić, Luka
Kaplan, Gordana
Gašparović, Mateo - Powiązania:
- https://bibliotekanauki.pl/articles/2191359.pdf
- Data publikacji:
- 2022
- Wydawca:
- Stowarzyszenie SILGIS
- Tematy:
-
air quality
TROPOMI
machine learning
PM2.5
PM10
remote sensing
jakość powietrza
uczenie maszynowe
teledetekcja - Opis:
- The objective of this study is to create a methodology for accurately estimating atmospheric concentrations of PM2.5 and PM10 using Sentinel-5P and other open-source remote sensing data from the Google Earth Engine (GEE) platform on a monthly basis for June, July and August which are considered as months of non-heating season in Croatia, and December, January and February, which, on the other hand, are considered as months of the heating season. Furthermore, machine learning algorithms were employed in this study to build models that can accurately identify air quality. The proposed method uses open-source remote sensing data accessible on the GEE platform, with in-situ data from Croatian National Network for Continuous Air Quality Monitoring as ground truth data. A common thing for all developed monthly models is that the predicted values slightly underestimate the actual ones and appear slightly lower. However, all models have shown the general ability to estimate PM2.5 and PM10 levels, even in areas without high pollution. All developed models show moderate to high correlation between in-situ and estimated PM2.5 and PM10 values, with overall better results for PM2.5 than for PM10 concentrations. Regarding PM2.5 models, the model with the highest correlation (r = 0.78) is for January. The PM10 model with the highest correlation (r = 0.79) is for December. All things considered, developed models can effectively detect all PM2.5 and PM10 hotspots.
- Źródło:
-
GIS Odyssey Journal; 2022, 2, 2; 59--77
2720-2682 - Pojawia się w:
- GIS Odyssey Journal
- Dostawca treści:
- Biblioteka Nauki
Artykuł