What question did this study set out to answer?

To explore the relationships between urbanization, land surface temperature, and vegetation density in Prague.

April 30, 2026Advances in Space Research2 citationsOpen Access

Assessing the changes of vegetation density, urbanization and surface urban heat islands in Prague using Landsat-8 spectral remote sensing indices

FYFurkan YilganCzech University of Life Sciences Prague TDTuğba DoğanInonu University MÜMustafa ÜstünerSamsun University

Key Points

To explore the relationships between urbanization, land surface temperature, and vegetation density in Prague.
Analyzed Landsat-8 satellite data from 2015 to 2022 using spectral remote sensing indices.
Calculated NDVI, VCI, NDBI, and LST to assess environmental changes.
Evaluated correlations between Landsat LST and MODIS datasets, along with ground station data.
Found a negative correlation between NDVI and LST, and a positive correlation between NDBI and LST.
Landsat LST showed a 28.5% increase in NDBI from 2015 to 2022 and a 13% decrease in NDVI.
The study reveals the persistence of the surface urban heat island effect despite varying LST trends.

Abstract

Rapid urbanization and industrialization, coupled with uncontrolled demographic shifts, have led to significant challenges, including increased pressure on energy resources and unsustainable encroachment on natural landscapes, as rural to urban migration continues to pose serious implications. Thus, urban intensification, characterized by an increased impervious surface, has far-reaching consequences on local climate patterns, accentuating surface temperature and thereby influencing broader climatic dynamics. High temperatures in metropolitan cities can trigger health problems in humans. This experimental research aims to assess the relationships among vegetation density, urbanization, drought, land surface temperature, and surface urban heat island (SUHI) in Prague, Czech Republic, from 2015 to 2022. For these reasons, the normalized difference vegetation index (NDVI), vegetation condition index (VCI), the normalized difference built-up index (NDBI), land surface temperature (LST), and SUHI were calculated in Prague, Czech Republic. The dataset was chosen as the Landsat-8 operational land imager (OLI) and thermal infrared sensor (TIRS) satellite data by Google Earth Engine (GEE) (a cloud-based platform) for summer, the hottest season of the year, from 2015 to 2022. The accuracy of Landsat LST data was assessed using the Moderate Resolution Imaging Spectroradiometer (MODIS) daily 1 km spatial resolution LST dataset as well as data from meteorological ground stations. The correlation between Landsat LST and MODIS was r ≥ 0.38 for all years, and root mean square errors (RMSE) were ≤ 6 °C. In addition, a 0.97 correlation was found with 1.21°C RMSE between Landsat LST and ground data. The study reveals a negative correlation between NDVI and LST, while a positive correlation exists between NDBI and LST. The change in average NDBI from 2015 to 2022 shows an increase of 28.5%, while the average NDVI decreased by 13%. The study shows that, despite the different trends in LST variation, SUHI remains persistent. The cooling effect of vegetation and the heat-retaining nature of built-up areas contribute to these changes. The study also highlights the rising urban heat island effect, particularly in outer city areas, due to demographic shifts. Despite its limitations, this research provides valuable insights for sustainable urban planning. Additionally, this study will contribute to landscape and environmental planning.

KI fragen

Bookmark

View Full Paper