Creation of RDF-Functions

Figure 7.2: Example RDF configuration map.

The basic idea stemmed from an inspection of variograms used in kriging (section 3.3). Variograms and more often semi-variograms are used to describe spatial autocorrelation with respect to distance for estimating interpolation parameters (Bucher, 1998; Cressie, 1993). A habitat parameter description in relation to distance is exactly what is needed and used here.

I will begin with a simple observation point in a woodland area as illustrated in figure 7.2. The amount of woodland around this point can be estimated by calculating its areal extent within a certain radius around the observation.

This is done for different radii (hence the name radial distance function) until a suitable resolution and extent is achieved. Further increase in resolution would finally result in a measure of circle length at a certain radius. In a second step a graph is created using these calculated values. In figure 7.3 the RDF-area plot is shown. The horizontal axis represents the distance from the observation point, whereas the vertical axis shows the area (m²) of woodland. This is the simplest form of a RDF plot which can be guessed approximately by visual inspection of figure 7.2. In the close vicinity of the observation (center) only little woodland is available. At distances around 1000m the absolute amount of woodland is highest. Below and above that value only a small area of wood is present with two secondary peaks around 275 and 1750m, whereas at distances around 400-600m and 1500m local minima are present.

Figure 7.3: Example RDF area plot.

This procedure can be applied to a variety of measurements of the surrounding habitat in different ways. The scope of the following section is to provide an overview of the variety of aspects that can be used in RDFs and the different types of RDF-plots.