Geospatial analysis of LULC change, future prediction, and its impact on LST in Bahir Dar City, Ethiopia

Understanding the dynamics of land use/land cover (LULC) and land surface temperature (LST) is important for sustainable urban planning and environmental management. This study analyzed historical changes in LULC and LST and projected their future patterns in Bahir Dar City, Ethiopia. Landsat imagery from 1992, 2007, and 2022 was used to assess LULC and LST dynamics. The Normalized Difference Vegetation Index (NDVI), Normalized Difference Built-up Index (NDBI), and Normalized Difference Water Index (NDWI) were used to examine relationships between surface characteristics and LST. LULC classification was performed using supervised object-based image analysis, and classification accuracy was assessed using a confusion matrix. Future LULC and LST patterns for 2035 were projected using Markov Chain, Cellular Automata, and Artificial Neural Network approaches implemented in the MOLUSCE plugin in QGIS. The results showed substantial landscape transformation between 1992 and 2022. Built-up land expanded from 350.4 to 5027.9 ha, while agricultural land declined considerably. The expansion of built-up surfaces was generally associated with higher LST, whereas vegetation showed a cooling influence. The 2035 projections indicate continued urban expansion and a potential increase in surface thermal conditions if current land-use trends persist. These findings highlight the importance of integrating green infrastructure and sustainable land-use planning into urban development. The study provides geospatial evidence to support environmentally sustainable and climate-resilient urban planning in Bahir Dar City and other rapidly growing cities.