A minimum cross entropy model to generate disaggregated agricultural data at the local level

This work presents an entropy approach to disaggregate agricultural data at a local level. It comprises two different steps. In a first one, an information prior at disaggregated level based on experts’ opinions, available cartography of land use and biophysical data is created. In a second step, it’s used a minimum cross entropy process in order to manage these information inputs and to guaranty a solution compatible with all the different restrictions. The model was applied to the region of Algarve in the year of 1999, in order to disaggregate the data at the pixel and county level. Results show that the model was able to provide some satisfactory results since the estimated values obtained for the different areas revealed a good approximation to the true values. These results were then analyzed and provided a new insight about the policies’ consequences in the territory.