W12: Dynamic and fuzziness in spatial and land development processes

organised by H. Wiggering, K.-O. Wenkel, R. Wieland


Title: New directions and challenges in spatial dynamic modelling of ecosystem functions in heterogeneous landscapes as basis for a better sustainable landscape management

Authors: Ralf Wieland, Wilfried Mirschel, Karl-Otto Wenkel

Abstract: The aim of this workshop is to discuss scientific challenges and new possibilities for a better modelling of consequences of land use changes in heterogeneous landscapes on ecosystem functions in space and time. The landscape or regional scale means an area of about 100 km2 up to some 1000 km2. Main problems on this scale are high complexity, structural diversity, ecological heterogeneity and uncertainty in data, in understanding of the process dynamic and by uncertainty in models.

Title: Integrated Modeling of Land Use Change

Authors: Andrei Kirilenko

Abstract: Urbanization is an important driving force of land use change. Change in land cover significantly alters numerous processes at different spatial and temporal levels and leads to subsequent changes in the environment. To reflect the changes at different hierarchical levels, a decision support system should include a number of components acting at various scales, and integrated into a unified model framework. The degree of flexibility of the final decision support system with regard to incorporating new features and components is defined by the selected method of model integration. A more traditional way of deep model integration suggests that a new component is customized each time the system is changed, essentially leading to a new model being developed. As opposed to the deep integration, the functional integration leaves the original models at each functional level intact. Instead of customizing the models, it makes an emphasis on coordinating the computations a! nd redirecting the data flows inside the system. The paradigm of functional integration offers a modeler a greater degree of flexibility. We applied the principles of functional integration to development of a model of land use change in the Upper Wabash river basin. The system components include two models of land use change acting at different spatial resolutions, a hydrological model, and a wildlife model. An integrated GIS tool provides an access to spatially distributed data used by the model and data visualization, as well as assists in scenario selection. The Upper Wabash river model is being developed for analysis of various strategies of land use change at regional and local scales.

Title: Fuzzy Habitat Models for Birds of Agricultural Landscapes with SAMT

Authors: Alfred Schultz, Ralf Wieland

Abstract: The habitat suitability for birds may serve as a meaningful indicator for biotic potentials of agricultural landscapes. The retrospective and prospective indication of habitat suitabilities for an appropriate range of birds will help to analyze the effects of land use and to derive land use systems compatible with multifunctional targets. The Spatial Analysis and Modelling Tool SAMT supports the development of habitat suitability models using grid data and fuzzy algorithms. Modelling ecological relationships with the fuzzy approach comes close to the actual behaviour of birds, where patterns such as amplification or compensation of influences are frequently experienced. The paper introduces a general approach for the development of habitat models based on SAMT. Main steps of the approach, especially the modelling of the fuzzy relationships, are exemplified with the Corn Bunting (Emberiza calandra; Miliaria calandra) and the Red-backed Shrike (Lanius Collurio). In ! the model development process the input variables are subdivided into a permanent and a dynamic section. Due to their different biological and ecological effects the input variables may require different spatial resolutions. Sometimes it seems to be necessary to supply very specific and highly detailed spatial inputs in order to meet key ecological requirements and to avoid trivial model results. For practical reasons it is on the other hand necessary to keep the number and the size of the spatial inputs at a manageable size. Appropriate strategies to decompose the bird-environment-system, to substitute needed data or to derive them from analogous data, and to pre-process spatial inputs may help to solve the conflict between professional requirements and practical options.