Fractal dimensions of Florida wetlands

Application of fractal concepts to geographic problems is reviewed. The basics of fractals are presented. Scale, self-similarity and their effects on natural phenomena are discussed. Eight sites (six wetland and two non-wetland) are studied. Using Landsat Multi-Spectral Scanner data, the fractal dimensions of the surface reflectance of the sites are calculated by the contour method. Values obtained are related to the type of wetland and the structure and processes that characterize that type of wetland. Forested wetlands were found to have the lowest fractal dimension. Mixed forested and non-forested had the highest. Marshy non-forested sites were higher than expected. Conclusions drawn are that physical characteristics of different wetlands result in measurable differences in fractal dimension. Fractals will be valuable in computer aided wetland identification in the future.