Development of a cellular automata model using open source technologies for monitoring urbanisation in the global south: The case of Maputo, Mozambique

Cities throughout the world are expanding, and globally we are witnessing a significant amount of urbanisation. Currently, 54% of the world's population lives in cities, and an increase of 66% is expected by 2050. Based on the World Urbanisation Prospects' report, most of this growth is projected to concentrate in Asia and Africa. The resulting uncontrolled urbanisation can lead to many issues with profound impacts on health, poverty, and social inequality. Monitoring our global landscapes in the most rapidly growing regions particularly in the global south becomes thus an important endeavour. The estimation of future urban patterns become thus part of important topics that foresight the necessary support systems that efficiently cope with the carrying capacity of these urban regions. In this sense, this paper aims to i) monitor historical land cover changes in one of such major cities Maputo, Mozambique, and, ii) project future patterns of urban fabric in its metropolitan region. To achieve this, satellite imagery from Landsat between 2000 and 2015 was collected and classified using a maximum likelihood algorithm. Spatiotemporal monitoring of urban settlements showed that within the proposed temporal scope, massive urban development took place while depleting a substantial amount of other land use types. Based on this spatiotemporal cognisance, a customized cellular automaton (CA) model was built that embeds the trends of changes while predicting future patterns. Our predictions exposed a significant amount of urbanisation will evolve over the next years if no significant controlling mechanism to cope with rapid urbanisation is integrated. Our findings, as well as conclusions, aim to aid local urban planners and policy makers in the course of urban planning in the region while taking advantage of novel open source methodologies that offer cost reduction and enable efficient monitoring for large urban extents.