Omnipresent sprawl? A review of urban simulation models with respect to urban shrinkage

Simulation models on urban land-use change help in understanding urban systems and assist in urban planning. One of the challenges of simulating urban regions in Europe as well as in North America or Japan is urban shrinkage, where deindustrialisation, massive population losses, and ageing cause unforeseen (or unexpected) commercial and housing vacancies in cities. In order to set up a conceptual framework for model improvement to assist such challenges, we review recent urban land-use-change simulation models, using four different modelling approaches: system dynamics, linked transport urban models, cellular automata, and agent-based modelling. The focus of the review is to assess the causalities and feedback mechanisms that were implemented in these models. The results show that simulation models are very heterogeneous in implemented mechanisms leading to urban land-use dynamics. No single model fulfils all of the criteria required to model urban shrinkage in a spatially explicit way. However, system-dynamic models that are documented in the literature can serve as a good starting point for spatially nonexplicit simulation, and one example was found for linked transport urban models which encompasses aspects of urban shrinkage. The potential of cellular automata is unclear as spatially explicit data on vacancies to feed this class of models is usually not available. Agent-based models appear to be the most promising approach for spatially explicit modelling of urban shrinkage.