Introduction: Climate change in cities has been in close focus in the past few decades. Heat stress in urban areas has also had adverse effects on human health and is expected to worsen in the future due to the global warming. In recent years, the mapping of urban biophysical and thermal conditions as well as their relation to land use and land cover (LULC) and air pollution has attracted increasing interest. In most cases, UHIs are a result of the use of fossil fuels which affect air-pollution in urban areas. It is a well-known fact that an increase in the population, particularly in developing countries, intensifies the pressure on natural resources. Rapid population growth, in conjunction with urbanization, expansion, and encroachment into limited agricultural and green areas lead to the destruction of vegetation coverage. It is obvious that such destruction, in combination with population growth, causes environmental impacts such as intensified land surface temperature (LST), UHIs and air-pollution. LST is considered as one of the important parameters in urban climate, which directly controls the UHI effects. LST is believed to be closely associated with LULC, resulting in heat islands. In general, thermal remote sensing is regarded as an efficient technology which provides a synoptic and uniform means of studying UHI effects on a regional scale. In the absence of a dense network of land-based meteorological stations, the spatiotemporal distribution of LSTs from thermal remote sensing imagery can be used as data to support UHI management and, potentially, countermeasures. thermal satellite-measured LST has been utilized in various studies on heat balance, climate modelling, and global change. This is because LST is determined by the effective radiating temperature of the Earth’ s surface to assess UHIs. There some methods devised in this regard, such as SplitWindow Algorithm (SWA) and Single Channel Method (SChM). In this line, the present research seeks to investigate the relationship between LST and LULC in Isfahan City, using Landsat thermal remote sensing satellite images. For this purpose, Landsat 8 (OLI) and Landsat 5(TM) satellite images were utilized, and the obtained images were received and preprocessed from 2000 to 2018. Methodology: The study area is Isfahan City located at the longitudes of 51° 50′ E and 51° 78′ E and the latitudes of 32° 50′ N and 32° 80′ N. The data used in this study were Landsat 8-sensor (TIRS-OLI) and Landsat 5-sensor (TM) satellite images. Land use classification was also done by the Ecognition and Envi 5. 3 software programs. Then, Hot Spot analysis was done to determine the hot and cold clusters in Isfahan thermal islands. Finally, the ArcGIS 10. 5 software was used to plot the corresponding maps. Results and Discussion: Land surface temperature was determined by the split window algorithm for images obtained via TIRS and the single-channel TM sensor of regional weather stations. The statistical analysis of the temperature data was done via Roots of Mean Square Errors (RMSE), and the values found for 2000 and 2018 were 1. 64 and 0. 93 respectively. The split window method proved to have a better performance than the single-channel method. The other achievement of this study is to identify the relationship of land surface temperature and land use. According to the results, the temperature of the desert areas was 42. 76° C and 46. 06° C in 2000 and 2018 respectively. It can be claimed that the high temperature of these area is due to the lack of vegetation. The temperature of the lands with water consumption was also 21. 6° C and 32. 09° C in 2000 and 2018 respectively. This phenomenon can be attributed to the development of urban areas and the recent droughts which have occurred in the Zayandeh Rood River. Consequently, Hot Spot analysis was done to determine the hot and cold clusters in Isfahan thermal islands. The analysis of the spatial correlation with global Moron indexes suggested that the surface temperature of the earth is distributed in a cluster form. Hot Spot analysis, indeed, provided evidence for the spatial concentration and clustering of the thermal islands in Isfahan over the time. Conclusion: The results of this study showed that the object-oriented classification method provides better results and has a better kappa coefficient and higher overall accuracy. There is also a close relationship between land use and surface temperature. The results indicated a negative correlation between LST on one hand and vegetation canopy and moisture, as already known from many other studies. LST was found to be very sensitive to vegetation and humidity. The study also rejected the hypothesis of irrelevance of spatial temperature of Isfahan (H0) and proved that the surface temperature data of Isfahan have a spatial structure and are distributed in clusters.