ABSTRACT
Agricultural land use data are often collated on an annual basis by national and supranational agencies by sets of administrative polygons. These administrative regions are often inconvenient for the analysis of land use change in relation to other phenomena, such as biodiversity. This study introduces an approach that spatially disaggregates land use data to fi ner resolution output geographies. The approach fi rst calculates models of land use activity at the original resolution, based on land cover map proportions. A simulation process follows whereby individual parcels of land use are allocated to the fi ner resolution geography on the basis of the models’ scores. The method is applied to British agricultural census data and to European agricultural statistics. Results show very good fi ts for some land uses for both British and European data, despite the very simple models employed in the fi rst stage of the approach. The spatial patterns obtained are seen to be stable after as few as 100 iterations of the simulation stage. The approach can be generalized to most input and output geographies.
