Addressing the impact of land use land cover changes on land surface temperature using machine learning algorithms

Autor: Sajid Ullah, Xiuchen Qiao, Mohsin Abbas
Jazyk: angličtina
Rok vydání: 2024
Předmět:
Zdroj: Scientific Reports, Vol 14, Iss 1, Pp 1-15 (2024)
Druh dokumentu: article
ISSN: 2045-2322
DOI: 10.1038/s41598-024-68492-7
Popis: Abstract Over the past two and a half decades, rapid urbanization has led to significant land use and land cover (LULC) changes in Kabul province, Afghanistan. To assess the impact of LULC changes on land surface temperature (LST), Kabul province was divided into four LULC classes applying the Support Vector Machine (SVM) algorithm using the Landsat satellite images from 1998 to 2022. The LST was assessed using Landsat data from the thermal band. The Cellular Automata-Logistic Regression (CA-LR) model was applied to predict the future patterns of LULC and LST for 2034 and 2046. Results showed significant changes in LULC classes, as the built-up areas increased about 9.37%, while the bare soil and vegetation cover decreased 7.20% and 2.35%, respectively, from 1998 to 2022. The analysis of annual LST revealed that built-up areas showed the highest mean LST, followed by bare soil and vegetation. The future simulation results indicate an expected increase in built-up areas to 17.08% and 23.10% by 2034 and 2046, respectively, compared to 11.23% in 2022. Similarly, the simulation results for LST indicated that the area experiencing the highest LST class (≥ 32 °C) is expected to increase to 27.01% and 43.05% by 2034 and 2046, respectively, compared to 11.21% in 2022. The results indicate that LST increases considerably as built-up areas increase and vegetation cover decreases, revealing a direct link between urbanization and rising temperatures.
Databáze: Directory of Open Access Journals
Nepřihlášeným uživatelům se plný text nezobrazuje