One of the basic characteristics of landscapes is the diver sity or heterogeneity of the landscape pattern (mosaic).
Hundreds of landscape metrics have been proposed by various researchers to analyze the landscape pattern. Most of these are covered by the computer program FRAGSTATS. The most typical use of the FRAGSTATS based landscape metrics is for the predic tion of species diversity. Also, several researchers have used FRAGSTATS based landscape metrics as indicators of various landscape changes (management activities and nat ural disturbances) such as the change in the spatial structure of landscapes, forest planning and management, landscape destruction and rehabilitation, and landscape disturbances by fire and road construction. This demon strates that temporal (time series based) indicators are inseparably related to spatial indicators. In order to control how landscape metrics respond to changing grain size, extent, the number of zones, the direction of analysis, etc., landscape simulators are applied. Gardner et al. introduced the concept of neutral models into landscape ecology. The aim of a neutral model is to have an expected pattern in the absence of specific landscape processes. In order to have a random pattern, the first application of this concept stemmed from the percolation theory, but different types of regular artificial landscapes are also used.
Landscape coherence has been considered one of the criteria for the development of sustainable rural land scapes. Proceeding from Bockemühl's concept of landscape identity and perception, which was developed in biodynamic farms, van Mansvelt classifies the ecologi cal coherences of rural landscape in three groups: vertical (on site), horizontal (landscape level), and cyclical (tem poral) coherences. The first type can be referred to as coherence between biodiversity and the local abiotic environmental conditions. For instance, soil bound agri cultural production would be an example of vertically coherent biodiversity management. The horizontal type of ecocoherence is ''that between coherence within a habitat (biotope or mini ecosystem) and that of habitats in a landscape (macro ecosystem)'' (van Mansvelt, 1997). This coherence refers to the functional (ecophysiological) interdependency of species within the ecosystems, but also to the relationships of habitats within the larger system. According to Kuiper, horizontal coherence is characterized by the connectivity between similar eco systems in a landscape. Cyclical (temporal) coherences are characterized not only by the full life cycles ofspecies and systems, but also by the self production ofspecies and biotopes, and season compliant management (e.g., sow ing, mowing, coppicing, etc.).
From the methodological point of view, van Mansvelt's concept of landscape coherence is rather hol istic and is used in the context of landscape perception and visual characteristics, with no studies that quantify this category in landscape validation. The most common estimates of different ecological coherences are their appearance or absence or relative scores. Another attempt to estimate coherence refers to the connectivity between landscape components. However, as in the case of various analogous indices that have been developed to describe landscape connectivity, this approach does not consider the quantification of coherence.
Wascher (2000) defines landscape coherence as the ''adequacy of land use according to biophysical conditions.''
Mander and Murka developed a dynamic landscape coherent concept which links issues of landscape diversity and landscape change. This concept refers to the correspondence between changes in actual (cultural or man made) landscape diversity caused by land ameliora tion or transformation of landscape pattern (e.g., due to changing socioeconomic conditions) and potential (bio physically determined) landscape diversity. According to this concept, the homogenization of landscape diversity caused by amelioration or other anthropogenic distur bances and determined on the basis of ecotone length per area unit can be lowest in the most sensitive (less resistant) landscapes. These are landscapes with both very simple and very complicated potential (biophysical) diversity, determined by heterogeneity of soil cover (Figure 1 ).
Was this article helpful?