Land use transformation contributes to land surface temperature (LST) change, which has been considered as one of the most critical environmental challenges. This study aim to analyze the impacts of land use dynamics on normalized difference vegetation index (NDVI) and LST between 1984 and 2024 in selected districts of Jimma Zone, southwestern Ethiopia. Landsat 5 TM (1984), Landsat 5 TM + (1997), Landsat 7 ETM + (2010), and Landsat 8 OLI-TIRS (2024) with 30 m spatial resolution were acquired from the USGS website. All spatial data were georeferenced to the UTM projection (Zone 37), and the WGS 84 datum. In this study, geospatial data were layer stacked and mosaicked using the geospatial analyst tool. The classified land use class was validated using GPS field survey points and Google Earth imagery. To realize the significant effects between NDVI and LST, correlation analysis was conducted for each year. One-way ANOVA was performed to test differences between land use classes. Mean separation analysis was performed using LSD at p= 0.05, using R statistical software to determine land use practices that are most effective. Results showed that the dense forest declined by 12.92%, indicating large-scale deforestation likely driven by agricultural expansion. In contrast, agricultural land increased by 7.09%, while open forest decreased by 4.66%. More expansion was observed in settlements, increasing by + 11.27%, underscoring intense unscientific infrastructure development that consumes surrounding landscapes. Even water bodies were reduced by 0.80%, a critical indicator of stress from drought and irrigation demands. The consistent and substantial decline in the maximum NDVI value, which has been declined from 0.51 in 1984 to a much lower 0.41 in 2024.Dense forest experienced a significant warming of 8.2 °C, diminishing its natural cooling capacity. Agricultural land had its minimum temperature rise from (18.6 °C-25.3 °C) by 6.7 °C. Open forest showed a consistent warming trend. The relationship among NDVI and LST is negative; this means that for each single unit of increase in NDVI, the LST drops. The most extreme LSTs were consistently found in settlements. The statistical test of surface temperature among LULC classes shows that settlements and agricultural land are a high level of significant warmer at the range of p = from 1.09 × 10^ (-5) to 2.24 × 10^ (-2).
