Browsing by Author "Jovanovic, Nebo"

Now showing 1 - 4 of 4
  • Thumbnail Image
    Item
    Assessing the Effects of Land Use on Surface Water Quality in the Lower uMfolozi Floodplain System, South Africa
    (International Journal of Environmental Research and Public Health, 2021-01-11) Chirima, George; Dlamini, Mandla; Jovanovic, Nebo; Adam, Elhadi
    This study investigated the impacts of cultivation on water and soil quality in the lower uMfolozi floodplain system in KwaZulu-Natal province, South Africa. We did this by assessing seasonal variations in purposefully selected water and soil properties in these two land-use systems. The observed values were statistically analysed by performing Student’s paired t-tests to determine seasonal trends in these variables. Results revealed significant seasonal differences in chloride and sodium concentrations and electrical conductivity (EC) and the sodium adsorption ratio (SAR) with cultivated sites exhibiting higher values. Most of the analyzed chemical parameters were within acceptable limits specified by the South African agricultural-water-quality (SAWQ) water quality guidelines for irrigation except for sodium adsorption ratio (SAR), chloride, sodium and EC. EC, pH and nitrate content which were higher than the specified SAWQ limits in cultivated sites. Quantities of glyphosate, ametryn and imidacloprid could not be measured because they were below detectable limits. The study concludes that most water quality parameters met SAWQ’s standards. These results argue for concerted efforts to systematically monitor water and soil quality characteristics in this environment to enhance sustainability by providing timely information for management purposes.
  • Thumbnail Image
    Item
    A comparison of ensemble and deep learning algorithms to model groundwater levels in a data-scarce aquifer of Southern Africa
    (MDPI, 2022) Gaffoor, Zaheed; Pietersen, Kevin; Jovanovic, Nebo
    Machine learning and deep learning have demonstrated usefulness in modelling various groundwater phenomena. However, these techniques require large amounts of data to develop reliable models. In the Southern African Development Community, groundwater datasets are generally poorly developed. Hence, the question arises as to whether machine learning can be a reliable tool to support groundwater management in the data-scarce environments of Southern Africa. This study tests two machine learning algorithms, a gradient-boosted decision tree (GBDT) and a long short-term memory neural network (LSTM-NN), to model groundwater level (GWL) changes in the Shire Valley Alluvial Aquifer.
  • Thumbnail Image
    Item
    Use of multi-source remotely sensed data in monitoring the spatial distribution of pools and pool dynamics along non-perennial rivers in semi-arid environments, South Africa
    (Taylor and Francis Group, 2022) Maswanganye, Sagwati Eugene; Dube, Timothy; Jovanovic, Nebo
    This study explored the use of multi-source remotely sensed data in monitoring the spatial distribution of pools and pool dynamics in two distinct semi-arid sites in South Africa. The factors that control the pool dynamics were also examined. Three water extraction indices were used, these included Normalised Difference Water Index (NDWI), Modified NDWI and Normalised Difference Vegetation Index. In addition, random forest classifier and Sentinel-1 SAR data were used in mapping pools and pools dynamics for both sites.
  • Thumbnail Image
    Item
    Using the water balance approach to understand pool dynamics along non-perennial rivers in the semi-arid areas of South Africa
    (Elsevier, 2022) Maswanganye, Sagwati E.; Dube, Timothy; Jovanovic, Nebo
    The Touws River in the Klein Karoo region of South Africa Study focus: This study sought to improve the understanding of pool dynamics along non-perennial rivers (NPRs) by utilising the water balance approach to assess the water fluxes that influence pool dynamics in the Touws River. The water balance model made use of various in-situ and satellite-derived data. New hydrological insights: The analysis of the water losses from the pool showed that most of the water was lost through evaporation. The interaction between the pool and groundwater is dependent on the water levels, as the pool loses water to the subsurface up to a certain depth then it starts gaining. When the Wolverfontein 2 pool is full, it can retained water for approximately 258 days without having a surface water inflow.