Files
Abstract
In Germany, since several decades the RAUMIS modelling system is applied for policy impact
assessments to measure the impact of agriculture on the environment. A disaggregation at the
municipality level with more than 9.600 administrative units, instead of currently used 316
counties, would tremendously improve the environmental impact analysis. Two sets of data
are used for this purpose. The first are geo-referenced data, that are, however, incomplete
with respect its coverage of production activities in agriculture. The second set is the micro
census statistic itself, that has a full coverage, but data protection rules (DPR) prohibit its
straightforward use. The paper show how this bottleneck can be passed to obtain a reliable
modelling data set at municipality level with a complete coverage of the agricultural sector in
Germany. We successfully applied a Bayesian estimator, that uses prior information derived a
cluster analysis based on the micro census and GIS information. Our test statistics of the
estimation, calculated by the statistical office, comparing our estimates and the real protected
data, reveals that the proposed approach adequately estimates most activities and can be
used to fed the municipality layer in the RAUMIS modelling system for an extended policy
analysis.