MODELLING URBAN LAND-USE CHANGE USING ARTIFICIAL NEURAL NETWORKS. COMPARING TWO SOFTWARE APPLICATIONS

This paper discusses the use of a methodology based on artificial neural networks (ANNs) for modeling land use change in the region of Madrid. The aim of this paper is to explore the potential for urban land use change transition in Madrid and also to compare the results produced by two different applications that implement ANNs: Land Transformation Modeler (LTM), developed by Purdue University and the Land Change Modeler (LCM) software implemented in IDRISI Taiga (Clark University). The geodatabase used covered urban land uses for two time periods (2000 and 2006), employed to train and calibrate the model, and a third time period (2009), used to conduct a validation process. The results showed that even though both applications implement the same algorithm, they can produce different results. In addition, this work underlines the usefulness of ANN-based land use change simulation models to reproduce patterns of land change to generate land use suitability maps.