Integrating Landsat 8 and Sentinel 2 in increasing the spatial resolution of land surface temperature: A case study in Ho Chi Minh city

One of the most popular methods for calculating land surface temperature (LST) is based on the thermal band of satellite images, but the resolution is relatively low and medium. Therefore, the technique of sharpening or downscaling the resolution to calculate LST values is getting more popular. The...

Full description

Saved in:
Bibliographic Details
Main Authors: Ha Tuan Cuong, Ha Tran Phuong
Format: Article
Language:English
Published: EDP Sciences 2025-01-01
Series:E3S Web of Conferences
Online Access:https://www.e3s-conferences.org/articles/e3sconf/pdf/2025/26/e3sconf_eier2025_01002.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:One of the most popular methods for calculating land surface temperature (LST) is based on the thermal band of satellite images, but the resolution is relatively low and medium. Therefore, the technique of sharpening or downscaling the resolution to calculate LST values is getting more popular. The method of ameliorating LST’s resolution is approached by building a linear regression equation between land surface indices from high-resolution satellite images and LST values retrieved from low to medium-resolution images. This study experimented in Ho Chi Minh city from 2015 to 2022 by integrating Landsat 8 images and Sentinel 2 images. The results showed that when comparing LST at 10m resolution (built from the regression equation) and LST at 30m resolution (retrieved from the thermal band of Landsat 8 image), land objects such as built-up land, vegetation cover, water surface, etc. had more particular results. In addition, the correlation coefficient of the regression equation was at a relative level (r > 0.5), proving that this method can be applied to further research on LST changes on land surface objects in urban areas.
ISSN:2267-1242