Spatio-Temporal Urban Growth Analysis and Modeling Using the CA-ANN Model: A Case Of Shashamene City, Ethiopia

Abstract

Urbanization is a global trend with significant environmental effects, urging a clear policy at all levels. Analyzing the dynamics of urban growth on both spatial and temporal scales has been recognized as a critical step in understanding the consequences of urbanization and developing long-term planning strategies. In this study, Shashemene City, one of the rapidly growing cities in Ethiopia, is considered to analyze it’s spatiotemporal growth from 1991 to 2021.Through supervised image classification using the support vector machine algorithm, LULC maps of the study area in 1991, 2006, 2013, and 2021 were prepared, and the spatiotemporal changes of LULC were investigated. Selected spatial metrics (number of patches, patch density, largest path index, class area, perimeter area fractional dimension) were used to examine the dynamics of landscape structure in the study area over time. The urban area of the city experienced a fragmented urban growth process. In order to simulate the spatiotemporal change transition potential and future LULC simulation in the Shashemene city, the five driving criteria, such as DEM, slope, aspect, distance from the road, and distance from the river, were integrated with CA-ANN technique within the MOLUSCE plugin of QGIS. The CA-ANN was then used to forecast changes in LULC for the years 2031 and 2041. The simulation result shows that almost 6900 ha of the total land was in the built up category of land use during 2021 and it would be further increased to 11191 ha in 2041. The Agricultural and vegetation area would be decreased for the next twenty years. This study would support planners and stakeholders to plan a new direction before negative outcomes become irreversible. The formulation of urban planning policies should also aim to create a balance between the sustainable use of limited resources and socio-economic needs.