3. Dynamics and Predictability of Land Use/Land Cover Change using Artificial Neural Network-Based Cellular Automata (ANN-CA): The Case of the Upper Awash River Basin, Ethiopia

Changes in land use and land cover (LULC) modify the hydrological and hydraulic processes of watersheds and can increase surface runoff and streamflow, which results in flooding. Increasing flood events have been observed globally and in Ethiopia, particularly in the Awash River Basin. It has been a chronic problem resulting in loss of human lives and economic damage. The upper Awash River Basin is an area that is highly susceptible to seasonal flooding. Taking a wider view of the interactions between land and water environments within a river basin and the broader socioeconomic and environmental implications of floods, the integrated flood management (IFM) approach provides a sound, conceptual basis to bring about a convergence between land use planning and flood management. Therefore, to implement sustainable IFM plans and mitigate flooding consequences, it is vital to have information and understand the dynamics of the current and future land use and land cover changes in the study area, thus helping to model, quantify and predict the impact of LULC change on flooding. This study assessed the LULC change dynamics of the upper Awash River Basin over the past 20 years (1998–2018) and predicted future land use and land cover change scenarios for 2038 and 2058 using an artificial neural network-based cellular automata (ANN-CA) model. The land use-land cover changes in the upper Awash River Basin were modeled and validated using historical data. Spatial variables such as elevation, population density, distance to rivers and distance to the road network were considered in predicting the land use and land cover for years 2038 and 2058. The historical analysis showed that there was an increase in forest, wetland, and settlement/urban areas by 0.82%, 0.002% and 1.77%, respectively, while there was a decrease in agricultural land, shrub land, grassland, bare land, and water bodies by 2.17%, 0.34%, 0.11%, 0.02%, and 0.15%, respectively. On the other hand, analysis of the predicted LULC change for the coming 20 years (i.e., for 2038) revealed that there will be an increase in settlement/urban areas by 0.24%. However, it also resulted in a decrease in the LULC types of agriculture, forest, shrub land, grassland, wetland, bare land, and water bodies by 0.19%, 0.02%, 0.052%, 0.002%, 0.0008%, 0.0002%, and 0.003%, respectively.