Integrating remote sensing and wetland ecology: A case study on South African wetlands

Remote sensing is a valuable tool for wetland ecology and conservation. With this study, we aimed at providing relevant information on wetland characteristics, obtained by standard techniques and relatively cheap optical imagery. The number, surface area, distance, and dynamics of temporary and permanent wetlands were determined for the Western Cape, South Africa. These characteristics are important for the metacommunity structure of amphibians and invertebrates. Isolated open water wetlands were classified by supervised maximum likelihood classification on seven Landsat images (1987 - 2002). Imagery acquired in summer contained fewer wetlands than those acquired in winter. The number of winter wetlands showed an increasing trend over time, which was not significantly correlated with yearly rainfall. Most classified wetlands were smaller than 1.5 ha. The distance to the nearest-wetland was longer in winter. In comparison to temporary wetlands, fewer, but on average larger permanent wetlands were classified. The relatively high number of wetlands is essential for local and migrating wading birds. The many small observed wetlands could also serve as stepping-stones, important for species conservation. We conclude that through relatively cheap imagery and standard Geographical Information System (GIS) techniques, basic ecological data can be generated. However, the resolution of Landsat imagery is too low to detect small wetlands. High accuracy images (such as IKONOS) would give more detailed results, but the high cost and the lack of long term data are at present restricting factors for their use by ecologists.