ИСТИНА

Physical geography and landscape ecology demonstrate obviously increasing interest in short-term and long-term effects of natural and anthropogenic processes. Research efforts concentrate on the problem of landscape sensitivity which is crucial for evaluating duration of benefits derived from ecosystem services. Landscapes are treated as dissipative structures generated due to self-organization which manifests itself in emergence of spatial patterns and temporal rhythms. To reveal patterns we need to determine coherent scale for each interaction between landscape components. We treat landscape stability is a multidimensional and multiscale phenomenon. On the one hand, it depends on interior properties of a landscape unit. On the other hand, landscape exists in a framework imposed by higher-order system which provides constants. Multiscale analysis of landscape structure becomes essential part of investigation in order to forecast of chain reactions between components that are able to enforce or weaken ecosystem services. We propose the multiscale statistical procedure to determine spatial parameters of higher-order systems imposing constraints for properties of soil and plant cover. Morphometric features of landforms combination in the surroundings of a unit were taken as a surrogate for the spatial pattern of a higher-order geosystem which controls matter and energy flows and generates emergent properties. Multiple second-order regression models were composed to describe relationships between lower-order and higher-order geosystems. Residual analysis was applied as a tool to reveal areas where the relations between properties follow uniform pattern. We came to the conclusion that reliability of the unit in providing ecosystem services depends on the concordance between the landscape unit individual properties and the framework conditions imposed by the higher order geosystem.