201 
through it. It is the wide range of options available in this process of discretisation that makes 
environmental data modelling so complex. 
It is shown, that spatial data analysis is concerned with detecting and modelling spatial pattern, 
but pattern at one scale may be simply random variations at a different scale. The example of 
structural diversity (Chapter 4) illustrates, that the extent, to which the analyst zooms is not 
something that can be determined in any absolute sense. It depends on the phenomenon under 
study, the objectives of the analysis and the scale at which data have been collected. Some data 
are derived from satellites or scanners and come in the form of a regular grid or lattice while 
other data are a patchwork of areal units or irregularly distributed set of sites. Landscape 
ecology has developed indices to quantify landscape structure (landscape metrics). The primary 
interest is in the amount and distribution of the smallest achieved spatial units (patches) and 
particular patch types (classes). Class indices for example, separately quantify the amount and 
distribution of each patch type in the landscape and thus can be considered indices of 
fragmentation. If a clear relationship between such indices and ecological variables exist, they 
can be useful for a nature conservation approach. But specifically if landscapes and ecosystems 
under investigation are small in extent and relatively homogeneous, the use for conservation 
purposes should be considered carefully. It is important to understand, how measures of 
landscape structure are influenced by the designation of a landscape element. We nowadays 
have the tools within the GIS-and RS-environment to analyse data e g. directly derived from 
satellite data. We are able to quantify, compare, model, intra- and extrapolate data. But we 
need clear theory-driven rules to successfully derive ecologically meaningful results. 
4. Example of a GIS-applieation for sustainable environmental management 
A case study in central Europe is briefly described 
with a focus on the problems of measurements of the 
diversity of the ecosystem. The study area is the 
alluvial floodplain of the river Salzach on the left 
(German) shore from north of the city of Salzburg 
(Austria) to north of Burghausen, where the Salzach 
flows into the Inn. It is generally situated in the 
southeasternmost part of Germany and has a core 
area of about 18.6 km 2 (for more detailed description 
see Blaschke and Vogel 1993). This special focus 
herein is only a small part of a comprehensive study of 
the remaining riparian forests, of their composition, 
dynamics, and interactions of landscape components 
in a more and more vanishing and highly valued forest 
ecosystem. 
Constructing structural diversity with GIS 
GIS-technology and remote sensing integrates different spatial, temporal and thematic 
properties of ecological questions. Alternatively and in addition to traditional conservation 
evaluation methods mainly based on vegetation, it is tried to estimate structural diversity and to 
correlate it with the distribution of some indicator species in order to find a cost- and time-