摘要：The study uses information from different sources and on different scales in an integrated set of models in order to analyze possible land use change scenarios arising in response to CAP reform. Five main steps were followed: (1) analysis of past land use changes, (2) multivariate analysis of future land use changes using a neural network time series forecast model (Multi-Layer Perceptron Method), (3) modelization of land use change demand (Markovian Chains Method), (4) allocation of the demand to define transition localization, (5) definition of policy scenarios. The final stage is the comparison of CAP scenarios using a multicriteria decision making approach, in order to supply valuable information to policy makers regarding the possible local effects of key direction changes in CAP.
