Modeling urban change using cellular automata: the case study of Johannesburg, South Africa

Abstract

Urbanization is one of the most evident human-induced global changes. Despite its economic importance, urban growth has a considerable impact on the surrounding environment. The most hazardous impacts caused by the informal and sometimes poorly planned developments are: the destruction of green spaces, increase in traffic, air pollution, congestion with crowding and lack of significant contribution to national income. Remote sensing provides an excellent source of data, from which updated land use/land cover information and changes can be extracted, analyzed, and simulated efficiently. Recent advances in computer models, GIS and remote sensing tools enable researchers to model and predict urban growth effectively. Cellular automata models have better performance in simulating urban development than conventional mathematical models. Johannesburg is the economic powerhouse of South Africa and it is the most populous metropolitan area. The city has experienced a significant growth in informal settlements. This growth has led to the loss of vast expanses of land, thus reducing the land available for other land uses, and contributing to a series of environmental problems. This paper quantified, mapped, and analyzed, the urban growth of Johannesburg from 1995 to 2010 using Landsat TM & ETM+ data. Cellular automata techniques were implemented for modeling the urban growth of the city of Johannesburg up to 2030. The model predicted future urban changes within and at the periphery of the city. The forecasted urban land cover change would prove useful for future urban planning and management of space in Johannesburg.