Sensitivity of a standard Land Use Cover change cellular automata model to resample input Land Use Cover maps

Abstract

Input data is one of the main sources of uncertainty in Land Use Cover Change (LUCC) modelling. Research has focussed on the sensitivity of LUCC models to the spatial resolution of Land Use Cover (LUC) maps. However, little attention has been paid to the way that spatial resolution is changed. Both the spatial resolution and the resampling method change the modelled landscape composition and configuration. This may affect the way Cellular Automata LUCC models behave and, accordingly, the landscapes they simulate. This paper analyses the sensitivity of a standard LUCC model (Metronamica) to changes in the spatial resolution and resampling method of input LUC maps. Results prove how the model is more sensitive to changes in the spatial resolution than to variations of the resampling method. This last component has not much influence in the simulated landscape, although it alters the landscape composition.