Quantitative analysis of spatial distribution of land surface temperature (LST) in relation Ecohydrological, terrain and socio- economic factors based on Landsat data in mountainous area

Abstract

Land surface temperature (LST) is considered as one of the most significantly effective factors on the regional climate and ecology, playing an important role in connecting surface energy and water exchange. In mountainous regions, LST reveals lots of inconsistencies due to the effect of such factors as topography, vegetation, solar radiation, etc. We sought to investigate the the temporal and spatial variation LST in different years and its relationship with effective factors in 5 dimensions using Multiple statistical methods, the sepidan region in northwest Iran. The multi-factorial land use, topographic (elevation, slope, aspect), biophysical indices (normalized difference vegetation index (NDVI), normalized difference moisture index (NDMI), normalized difference built up index (NDBI), and modified of normalized difference water index (MNDWI)), socio-economic (fossil fuel CO2 emissions (FFCOE) and road density(RD)), and climate (temperature and solar radiation) was studied in the current research. To this end, Images of July 1998 and 2017 were extracted from Thematic Mapper (TM5) and Operational Land Imager/Thermal infrared sensors (OLI/TIRS8). Moreover, ordinary least squares regression (OLS), Best subset regression, and Hierarchical Partitioning Analysis (HP) were used to investigate the relationship between LST and relevant effective factors. The results indicated that the temperature range varied from 10 to 53 C in the time period mentioned. The highest amount of LST was observed in barren land use and the lowest one was found in garden lands. An negative correlation was found between LST and elevation. On the other hand, the highest value of the Laps rate of surface temperature was observed in the southern aspects and the lowest one was observed in the western aspects. Furthermore, the highest and lowest values of lase rate were found in slopes less than 10, and in 50 to 60-degree slopes, respectively. The results of the OLS correlation indicated a negative correlation between LST and NDVI, NDMI, and MNDWI, and a positive correlation of LST with climatic and socio-economic indicators. LST’s highest and lowest correlations were found to be with vegetation (R2 = 0.95) and road density (R2 = 0.1). Finally, while in 1998 temperature and vegetation were identified as the most influential factors on LST, it was the elevation that was found to be the most effective factor on LST in 2017 with the effective rate of 82.72%. This study offers a valuable viewpoint on the temporal and spatial variations of LST, their complexity, and the environmental factors that affect them. The viewpoint could, therefore, be used for prospective studies on the analysis of the ecosystem’s reaction to climate changes.