Urban growth in Latin American cities : exploring urban dynamics through agent-based simulation
The high rates of urban growth in Latin America during the 1960s and 1970s produced rapid urbanisation and housing problems. Planning policies as well as the research community have approached urban growth as a static problem rather than as a spatial form that emerges from the urban development process and that is part of a constant dynamic process. This thesis focuses on a specific kind of urban growth that happens in Latin American cities, called 'peripherisation'. This is characterised by the formation of low-income residential areas in the peripheral ring of the city and a perpetuation of a dynamic core-periphery spatial pattern. The dynamics of growth and change in Latin American cities are explored using agent-based simulation. The objective is to increase the understanding of urban spatial phenomena in Latin American cities, which is essential to providing a basis for future planning actions and policies. The thesis consists of two parts. The first part presents an overview of urban growth and dynamics in Latin American cities, drawing on previous work on urbanisation in Latin American cities, spontaneous settlements and inner city dynamic processes. The second part focuses on the development of a simulation model based on the theoretical framework established in the first part. A brief review of the literature of automata models is presented, with particular reference to agent-based simulation for land-use dynamics. The Peripherisation Model is introduced, its computer implementation described, and sensitivity analysis tests reported. Simulation exercises were used to revisit assumptions about urbanisation issues in Latin American cities and investigate important aspects of growth and change in these cities. These exercises allowed the problem of urban growth in Latin American cities to be unfolded through their dynamics, relating these dynamics to urban morphology, and thus presenting a new and important perspective on the phenomenon.