Modeling urban growth in Kigali city Rwanda

Abstract

In Kigali city, rapid urbanization and urban growth are recognized facts. The planning policy of urban areas in Kigali city has been disapproved of being ineffective. A likely reason of this inappropriate or lack of urban planning approaches was mainly approached to a limited understanding of the spatial- temporal patterns of the city and its dynamism. Moreover, the lack of appropriate techniques of urban growth modeling that can help in making valuable urban planning decisions is still critical.This study aims to determine the main driving forces behind the urban dynamics of Kigali city by coupling GIS and Logistic Regression (LR) model. This information is expected to support in determining the possible future directions of the city for the next 26 years, in turn, will form the basis of policy options/ decisions of the city. Consistent set of land cover maps of Kigali city during the period 1987, 1999, 2009 and 2014 were generated. Pixel- based approach by Maximum Likelihood method was applied to classify Landsat images. As the purpose of the study was to look at built-up land cover only, from the land cover maps, built-up area were extracted and visualized. A remarkable trends and sprawl pattern was observed in urban areas in Kigali city in that, the rapid trend was more accelerated between 1987 and 2009 while it continued at a moderate pace up to 2014. Urban areas expanded mainly towards the Northern, Eastern and Southern directions. However, the Western part of the city was posed due to the area topography and environmental policy on protecting forests. The results also highlighted the implication of human related activities with regard to contributing to a decrease in natural reserves (forests and vegetation) for the period between 1987 and 2014. The main drivers of Kigali city growth were identified according to both Logistic Regression model and local planning experts’ views. The results indicated that new urban developments in Kigali city will tend to be close to the existing urban areas, hence a compact pattern. In addition, new urban development will have a tendency to occur further from the CBD and wetlands but on low slope sites. The model predicted that if the current physical urban expansion rates continues, urban development will expand towards Northern and Southern direction of the city rather than Western and Eastern parts. The new developments have a tendency to replace forest cover and wetlands in the Western part of the city and this constitutes a serious environmental threat to the city. The research suggests further study by using other important factors revealed by local urban planning experts like neighborhood of water and electricity, but not used because of data unavailability. This study should be repeated by coupling LRM with other micro-level approaches CA model or CommunityViz.