Examining human activities in response to land surface temperature in Sekota watershed, northern Ethiopia

Abstract

The alteration of land use/land cover change (LULCC) is an environmental issue that impacts affects ecosystems by increasing the land surface temperature (LST). This study aimed to investigate the influence of human activities on LST in the Sekota watershed northern Ethiopia. This study used Landsat images and a supervised support vector machine (SVM) classification algorithm to map LU/LC and estimate LST. The findings revealed that farmland exhibited the most substantial expansion, with a net gain of 16,970.84 ha, while shrubland experienced the most significant decline, with a net loss of 20,768.57 ha. Moreover, forest cover by 329.73 ha, bare land by 2048.97 ha, and settlements by 131.07 ha increased from 2000 to 2022. The mean LST increased from 32.31 °C in 2000 to 36.01 °C in 2014, followed by a gradual decrease to 34.18 °C in 2022. The overall accuracy and kappa coefficients of the LULC maps were 87.6 % (0.8421), 91.5 % (0.8901), and 92 % (0.8973) in 2000, 2014, and 2022, respectively. This study also investigated the correlation between the normalized difference vegetation index (NDVI) and LST. The results demonstrated a negative relationship, with correlation coefficient R² values of 0.70, 0.65, and 0.75 for 2000, 2014, and 2022, respectively. This indicates that non-vegetated e areas had higher LST levels than forested areas. As a result, it is recommended that government agencies and local communities focus on preserving vegetation cover and adopting practices such as planting perennial fruit crops and implementing agroforestry systems in the study area.