N. Moreno (Canada, Venezuela) and D.J. Marceau (Canada)
Land-use change, cellular automata, vector-based cellular automata, raster-based cellular automata, scale sensitivity
Cellular automata (CA) models have been increasingly used to simulate land-use changes due to their computational simplicity and their explicit representation of space and time. Typically, these models use the raster model, as defined in Geographic Information Systems, to represent geographic space. However, recent studies have demonstrated that raster-based CA are sensitive to spatial scale, i.e. cell size and neighborhood configuration. To overcome this limitation, a novel Vector-based Geographic Cellular Automata (VecGCA) model has been developed in which space is represented as a collection of geographic objects corresponding to meaningful entities of irregular shape and size composing a landscape. This paper presents a land-use change model using this new approach, tested on two study areas of different spatial complexity, in southern Quebec and in the Calgary region, respectively. The results obtained are compared to the patterns produced by a conventional raster-based CA and with land-use maps in each study area. They reveal that VecGCA generates an adequate evolution of the geometry of the objects composing the landscape and produces spatial patterns that are more similar to the land-use maps in each region.
Important Links:
Go Back
