Resumen de Agricultural data disaggregation at a local level:: An approach using entropy and supervised classifications
António Manuel de Sousa Xavier, Maria de Belém Costa Freitas, Rui Manuel de Sousa Fragoso, Maria do Socorro Rosário
In Portugal, studies show that the changes in the Common Agricultural Policy (CAP) have led to several consequencesin land uses and in the environment. This calls for disaggregated agricultural data with precise geographical references.In order to tackle such problems data disaggregation processes are needed and a series of studies are being carried outat international level. However, these approaches didn’t take the utmost advantage of remote sensing methodologiesor considered the combination of these methodologies with point information from the Land Use/Cover Areaframe Statistical survey (LUCAS survey). Techniques such as supervised classifications, in which land use maps areautomatically built, using as basis their spectral signatures, are important tools. Therefore, the objective of this articleis to provide an approach to disaggregate agricultural data at local level, taking the utmost value of the existent upto-date satellite imagery. It has two steps: in a first step, a supervised classification is developed using satellite imageryand the LUCAS survey; in a second step an entropy approach is applied to guaranty consistence among the differentsources of information and with the aggregate. The results were compared with other approaches and showed to besatisfactory, and may be improved further.
