Assessing the impact of temporal dynamics on land-use change modeling

Time is a fundamental dimension in dynamic land-use change modeling. An appropriate treatment of time is essential to realistically simulate landscape dynamics. Current land-use models provide little justification of their treatment of time. As a result, little is known about how the time dimension impacts the spatio-temporal patterns produced by land-use simulation models. This paper reports a first exploration on this issue. It examines the impact of the degree of temporal dynamics on the behavior of an urban growth model which is based on a modified Markov random field and probabilistic cellular automata. Experimental results from this case study suggest that the degree of temporal dynamics does have an important impact on the urban morphology produced by the model. Too much or too little dynamics could both lead to unrealistic patterns. However, the impact seems to vary for processes with different levels of change intensity. In the case of a process with moderate changes, the impact of temporal dynamics is also moderate. For a process with high change rate, the degree of temporal dynamics affects the model output significantly. The implication of these findings is discussed in the context of information accessibility and operational land-use modeling.