Spatial Analysis of Land Surface Thermal Hot and Cool Spots and Their Relationship with Land Cover in Behbahan and Its Surroundings, Khuzestan Province

Document Type : Original Article

Authors

1 Ph.D. Student in Climatology, Faculty of Planning and Environmental Sciences, University of Tabriz, Tabriz, Iran

2 Phd Student in Geology, Faculty of Earth Sciences, Shahid Bahonar University of Kerman, Kerman, Iran

3 Department of Remote sensing and geographic information system, Dez.C., Islamic Azad University, Dezful, Iran

4 Department of Geography and Urban Planning, Mahs.C., Islamic Azad University, Mahshahr, Iran

5 Ph.D. Student in Climatology, Faculty of Humanities, University of Zanjan, Zanjan, Iran

10.22034/rsgi.2026.74589.1179

Abstract

Land surface temperature (LST) can vary markedly among built-up surfaces, vegetation, and bare land in hot, semi-arid settings. This study examined land cover and daytime surface thermal patterns in the Behbahan urban–peri-urban window using five summer Landsat acquisitions from 1987, 1999, 2007, 2016, and 2024. LST was obtained from the Landsat Collection 2 Level-2 surface-temperature product. Built-up surfaces, vegetation, and bare land were mapped with a supervised Random Trees classifier using surface-reflectance bands and the NDVI, NDBI, and MNDWI indices. Local Moran’s I and Getis–Ord Gi* were used to identify thermal clusters. Quantitative comparisons were restricted to the 50.79 km² area with valid observations at all five dates. Within this area, estimated built-up surfaces increased from 10.04 km² in 1987 to 17.16 km² in 2024, vegetation increased from 14.55 to 23.90 km², and bare land decreased from 26.19 to 9.73 km². Most of the vegetation increase occurred outside the fixed city boundary; vegetation within that boundary decreased from 2.04 to 1.36 km². Bare land generally coincided with higher daytime LST, although some vegetation-classified pixels also occurred within thermal hot clusters. Significant local thermal clusters were identified, and the LST–NDBI association was positive at all five dates. Classification accuracy was assessed with spatially separated samples drawn from the authors’ labelled polygons. Because independent ground reference data were unavailable for either land cover or LST, the results describe observed spatial associations rather than causal effects or a continuous temperature trend.

Keywords

Main Subjects

Volume 6, Issue 21
January 2026
  • Receive Date: 05 September 2026
  • Revise Date: 08 October 2026
  • Accept Date: 10 November 2026