Why have land use change models for the Amazon failed to capture the amount of deforestation over the last decade?

Land cover change in the Neotropics represents one of the major drivers of global environmental change. Several models have been proposed to explore future trajectories of land use and cover change, particularly in the Amazon. Despite the remarkable development of these tools, model results are still surrounded by uncertainties. None of the model projections available in the literature plausibly captured the overall trajectory of land use and cover change that has been observed in the Amazon over the last decade. In this context, this study aims to review and analyze the general structure of the land use models that have most recently been used to explore land use change in the Amazon, seeking to investigate methodological factors that could explain the divergence between the observed and projected rates, paying special attention to the land demand calculations. Based on this review, the primary limitations inherent to this type of model and the extent to which these limitations can affect the consistency of the projections will also be analyzed. Finally, we discuss potential drivers that could have influenced the recent dynamic of the land use system in the Amazon and produced the unforeseen land cover change trajectory observed in this period. In a complementary way, the primary challenges of the new generation of land use models for the Amazon are synthesized. (c) 2014 Elsevier Ltd. All rights reserved.