[Regionalisation of Germany by data of agricultural structures].

In order to simplify the design of representative studies in animal populations the structural differences of animal husbandry (cattle, pigs and laying hens) in Germany were characterised. Several regions were defined and thus districts identified which are typical for the respective region and can be regarded as representatives for the whole region. Data on animal husbandry as well as human population per district originated from the Federal Statistical Office and were linked to the geometric data of the Federal Agency for Cartography and Geodesy. By this, data of "livestock units/square kilometre area" and "farms/square kilometre area" per district were calculated using methods of the spatial statistics Global Moran's Index, Anselin Local Moran's Index and Getis-Ord Gi*. With the help of these analyses six clusters could be identified which resulted in four large (Middle, Northwest, East, and South) and one smaller region (Northern Upper-Rhine) respecting the federal state borders. These regions differed significantly regarding animal and farm densities. The selection of typical districts was carried out with the help of the respective animal and farm data of the species pigs, dairy cattle and laying hens. The means of the selected districts (three to six per region) were within the 60%- and the 80%-percentile of at least two of the analysed variables. Concerning the region Northern Upper-Rhine no representative district was selected. This presented regionalisation including representative districts can be used for the design of scientific studies that are associated with animal husbandry in Germany.