The use of remote sensing data for assessing water quality in wetlands within the Limpopo River Basin
Wetlands are unique ecosystems that are acknowledged among the world’s most productive and valuable ecosystems. They are recognized as being essential to sustainable development and human welfare due to their unique environmental and socio-economic value. These highly productive ecosystems provide functions such as recycling of nutrients, watershed protection and flood control as well as grazing resources. Wetlands provide the basis for human livelihoods in Africa through ecosystem services. However, these ecosystems are affected by internal and external factors within and outside their catchments, hence the importance of monitoring those changes around these wetlands. The aim of this study was to identify the major land use and land cover changes (LULC) from two selected wetlands (i.e. Makuleke and Nyslvei) and their impacts on water quality within the Limpopo Transboundary River Basin, South Africa. To achieve this aim, firstly the study assessed the impacts of LULC changes on these two wetlands between 2014 and 2018. Multi-date Landsat series data were used to map and estimate the rate of LULC changes in Makuleke and Nylsvlei wetland ecosystems during the study period. The results obtained showed that the spatial extent of Makuleke declined by 2% between 2014 and 2018, whereas the Nylsvlei wetland decreased by 3%. Some of the noticeable changes were that the coverage of natural vegetation tends to increase during the wet seasons. Secondly, Chlorophyll-a was predicted and mapped for Makuleke and Nysvlei between September 2018 and June 2019. Moderate resolution Landsat 8 images and in-situ field measurements were used to estimate and map chlorophyll-a concentrations from these two wetlands. Landsat-derived chlorophyll-a concentrations were validated using field-derived chlorophyll-a measurements. The results showed a variation of chl-a concentration in these two wetlands, with Makuleke wetlands concentrations ranging from 0 to 1.15 μg/L whereas for Nylsvlei wetland the ranges varied between 0 and 1.42μg/L. The finding of this study can be used in enforcing of wetland legislation and LULC management practices and highlights the relevance of remotely sensed data in assessing and routine monitoring wetland water quality.