Wetlands are considered to perform a number of ecosystem services, including the improvement of water quality. However, few empirical studies have been carried out on the capacity of wetlands to perform a water treatment function, or on the demand for this function, with the result that most estimates of the value of this service are based on a weak foundation. Furthermore, the measurement of this function is confounded by the complications in trying to assess water quality entering and leaving wetlands because of surface to groundwater interactions.
The aim of this study is to investigate the water treatement capacity of wetlands by means of a landscape-level analysis in which the water quality of rivers is related to the land use and wetland area in their catchment areas. The study is based on measurement of N, P and sediment load for a total of 100 catchment areas for which fine-scale land cover data are available will be sampled in the Western Cape. Sampling will be conducted in August, to capture the main period of wetland influence. The following hypotheses will be tested: (1) Water quality is a function of land use, point source pollution, proportional area of wetlands in the catchment, area of farm dams, and underlying geology. (2) Where wetlands predominate, their impact on water quality and sediment trapping will be measurable. (3) Where natural vegetation predominates, the influence of wetlands will be dampened. (4) Where farm dams predominate, the influence of wetlands may be negligible. The next phase involves the valuation of the treatment capacity of wetlands using the replacement cost method. Data will be collected from South African water treatment plants in order to obtain mean treatment costs, and demand for the service will be rated on the basis of existing data relating to water use downstream. This will be used to modify the replacement costs in order to provide a more realistic estimate of actual value. Data will be analysed using generalized linear modelling to obtain the coefficients pertaining to wetland area. If the model is significant we can theoretically derive the average uptake per ha of wetland and assign a monetary value to this. If not it might be argued that wetlands play a minor role in these catchment areas.
