Abstract
Waterbodies often make up a key component of wetland systems and therefore should be given due consideration in the management of wetlands. The need to monitor wetland waterbodies is becoming ever-pressing in the face of climate change and population growth that is putting increasing pressure on water resources. Monitoring waterbodies as part of land cover mapping in wetland ecosystems using remote sensing is well documented. Furthermore, much of the emphasis has been on large wetland and waterbodies that are fairly easy to map on publicly available remotely-sensed data such as moderate resolution Landsat imagery. There is a need to expand the utility of remote sensing to monitoring wetland waterbodies that have a great deal of variation in terms of spatial extent. This study therefore investigated the performances of remotely-sensed data in identifying and quantifying the spatial extents of urban wetland waterbodies found in the south of the City of Johannesburg, South Africa. The first goal of this study was to assess the seasonal and multi-year dynamics of wetland waterbodies using Landsat imagery. Seasonal assessment (summer, autumn, winter, spring) revealed that summer and spring had the most number of waterbodies in the study area. Analysis of the multi-year data (1995, 2005, 2008, 2016) taken in the same season revealed fairly unchanged waterbody amounts. The second goal of the study focussed on two specific objectives; 1) compare SPOT versus Landsat images, 2) compare Normalised Difference Vegetation Index (NDVI) versus Normalised Difference Water Index (NDWI) to delineate wetland waterbodies. The SPOT imagery identified more waterbodies than the Landsat and was also better in estimating waterbody extents, indicating the importance of spatial resolution as well as the sufficiency of visible and near-infrared spectral bands available in SPOT imagery. The comparison between NDVI and NDWI favoured the latter, which enabled identification of more waterbodies comparable to the reference data. in general, the findings of this study demonstrated the suitability of publicly available remotely-sensed data in monitoring wetland waterbodies with wide spatial variations common in urban environments. It is envisaged that continuous efforts in similar researches and the increasing availability of multisource remote sensing data will improve the monitoring efforts at high temporal resolutions.
M.Sc. (Geography)