ABSTRACT

This chapter presents and discusses two issues related to spatial data processing for modeling the natural environment. The fi rst issue is the mismatch between the spatial scale of processes and the neighborhood size used in spatial analysis for characterizing those processes. The second is scale incompatibility and its impact on the characterization of spatial joint distributions used in the modeling of the natural environment. It is shown in this chapter that information characterized through spatial analysis is very sensitive to the neighborhood size used. It is recommended that spatial analysis should employ a neighborhood size comparable to the spatial scale of the process. This can be achieved by altering the default neighborhood size in spatial analysis algorithms. It is also shown that scale incompatibility of geographic data can lead to signifi cant mischaracterization of spatial joint distributions of geographic factors. It is suggested that scale transformation of spatial data is needed to mitigate the impact of scale incompatibility on the characterization of spatial joint distributions.