Regionalised optimum control problems for agroecosystem management

This contribution presents a methodological framework and solutions for optimum control problems on regionalised agroecological models. The focus is laid on the estimation of optimum fertiliser input and crop rotation schemes as a dynamic control problem with different time scales. Beside the solution of the optimum control problems, which was presented in a former paper, this contribution focuses on the regionalisation of the optimum control problem. The task is solved by the identification of homogeneous units in the observed region by a geographic information system (GIS) using digital maps. These maps are set up by a vector-orientated database. The second innovative topic, which enables a regionalised solution of the optimum control problem, is the estimation of families of optimum solutions parameterised by spatial properties, e.g. soil parameters. Technical problems concern an efficient database access, which is solved within the GIS Arc/INFO, ArcView. Several solutions are presented for different regionalisation scenarios for a German investigation site. The proposed methodology supports the step of decision support in precision farming.