A domain-specific language for patchy landscape modelling : the Brittany agricultural mosaic as a case study

Recent developments in landscape ecology have emphasised the functional role of heterogeneity of mosaics such as a ?patchworks? of agricultural fields. Explicit (process) landscape models are important tools to study this functional role, although very few model are dedicated to patchy landscape dynamics. We construct such a model (L1) manipulating patches (polygons) instead of pixels by a combination of pure attributive or geometrical modifications. In addition, our aim was to build a perennial and dynamic software platform. Hence, the L1 platform contributes to the design of distinct scenarios, as well as distinct models of (forested, agricultural, suburban...) landscapes. It is designed around a kernel modelling, a generic patchy landscape that could be progressively driven to follow natural (climatic, ecological...) forces and human (sociological, economical...) decisions. We present, here, the first model of the L1 platform with an application on a Brittany agricultural landscape and four simplified scenarios. We focused on human decisions driving the patchy mosaic and demonstrated the importance of taking into account the different decision levels, from the main characteristics of the European Common Agricultural Policies (CAP) to the farmer individual land use allocation. Results are a set of rules that reproduces Brittany landscape evolutions with increasing complexity driving-decisions, and therefore, with increasing realism degree. They illustrate part of the L1 platform flexibility