This work examines the use of gravity data and its application to subsurface water reservoirs in the immediate vicinity of the South African Geodynamic Observatory, Sutherland (SAGOS), situated in a semi-arid region of the Karoo region of South Africa, and underlain by the Karoo sedimentary rocks intruded by dolerite dykes and sills. SAGOS houses the only supergravity metre (SG) in Africa, and this thesis sets out to test its use in monitoring groundwater dynamics using hydrological and gravity data. The main aim of this work is the application of the SG data, in conjunction with hydrological data, to better understand episodic recharge of subsurface reservoirs. The importance of water as a resource, globally and specifically the Karoo, is reviewed in conjunction with supply and demand of water. This is to contextualise the socio-economic, technical as well as policy issues related to water resource management. Applicable technologies for water resource management and efficient water use are highlighted and the application of gravity to hydrology is introduced, including satellite as well as ground based tools. In addition, arid zone hydrology as well as recharge and its mechanisms are analysed in order to better understand these processes when examined from gravity measurements. Issues related to understanding flow within the vadose zone as well as in secondary aquifers are examined, and gravity residuals and subsurface hydrology are highlighted. Thereafter, a conceptual groundwater flow modelof the study area is developed using multiple tools. First, the geology around SAGOS was mapped using SPOT 5 imagery and then ground truthed. Second, stable isotopes and water chemistry analysis was undertaken on water samples from selected boreholes. The results allude to preferential flow acting as the main mechanism for groundwater recharge. Follow-up pump-tests illustrate that fracture connectivity is greatest at close proximity to the dyke. Soil mapping, using aerial photography was also undertaken. Duplex soils, enriched with clay at depth, dominate the study area. Using in-situ infiltration tests, it is shown that the alluvium, which lines the river beds, has a higher hydraulic conductivity than the other soils, confirming that these streams act as preferential conduits for subsurface recharge. Precipitation events were correlated against gravity residuals at 4 wells, over different time periods. The results are examined using time series analyses. Gravity residuals from well SA BK07, over a period of 24 hours after the rainfall event, delineate instances of negative correlations, as well as strong positive correlations (of up to 0.9). On the whole however, correlations between gravity and groundwater at SA BK07 are variable and weak, and in conjunction with water level measurements and water chemistry, the data suggest that this well is located in a dynamic conduit (throughflow) and not in a permanent groundwater reservoir. By contrast, other wells show strong positive correlations between gravity residuals and water levels following episodic recharge events for a later time series. Correlations between the water levels and gravity residuals in wells SA BK04, SA BK05 and SA BK 01 are in excess of 0.7 for specific rainfall events. In summary, the results suggests that gravity is an excellent tool for measuring episodic groundwater recharge within the immediate vicinity of the SAGOS. This implies that gravity can aid in monitoring groundwater losses/gains in arid and semi-arid areas. Recommendations for future work are highlighted at the end; these include the possible use of hydrological modelling of reservoirs at various scales and then comparing these results to the SG as well as GOCE and GRACE satellites data, and then improving numerical modelling of the groundwater dynamics for sites like Sutherland and the surrounding arid Karoo region, where sparse water shortages, and potential pollution related to fracking for shale-gas, are likely to compete with established water needs for farming and human consumption. It is also suggested that the gravity modelling be examined to better understand site specific scenarios and thus aid in improving the processing of the gravity signal.

MENVSC / Department of Hydrology and Water Resources / ABSTRACT Siloam and Tshipise Springs are scalding geothermal springs geologically located within the Soutpansberg Group in the Limpopo Province of South Africa. These geothermal springs are associated with faults and impermeable dykes and are assumed to be of meteoric origin. The optimal use of a geothermal spring largely depends upon its physical and chemical properties as well as the geological controls at source and surrounding pathway to the surface. This study aimed at investigating trace element concentrations in these geothermal springs in order to quantify their impacts on neighbouring soil and vegetation. Impact on vegetation was assessed by incorporating seasonal variations of the trace element mobility from the geothermal springs to the vegetation (Mangifera indica at Siloam and Acacia robusta at Tshipise) via soil. The geothermal spring water, soil and vegetation samples at both sites were collected from May – July (winter) and September – November (summer), 2014. The soil samples were collected at 5 m intervals up to 20 m away from the geothermal spring in each of the sites. The bark and leaf parts of the vegetation were sampled. The control samples for water, soil and vegetation were obtained from Riverside residence at University of Venda, Thohoyandou, Limpopo Province, where there is non-geothermal source of water. The temperature, electrical conductivity (EC), pH and total dissolved solid (TDS) of the geothermal spring water and control samples were determined in situ and in the laboratory. The water samples were acidified for major cations and trace elements determination. There were also non-acidified water samples for major anion analyses. The soil and vegetation samples were digested using microwave and hot block methods, respectively. Concentrations of arsenic (As), antimony (Sb), barium (Ba), beryllium (Be), boron (B), cadmium (Cd), chromium (Cr), cobalt (Co), copper (Cu), mercury (Hg), lanthanum (La), lead (Pb), lithium (Li), manganese (Mn), molybdenum (Mo), nickel (Ni), selenium (Se), tin (Sn), strontium (Sr), tellurium (Te), thallium (Tl), titanium (Ti), tungsten (W), vanadium (V), and zinc (Zn) were determined by inductively coupled plasma – mass spectrometry (ICP-MS) (Agilent 7700 series). Concentrations of calcium (Ca), magnesium (Mg), sodium (Na) and potassium (K) were analysed using inductively coupled plasma – optical emission spectrometry (ICP-OES) (X – Series 2) whereas the concentrations of chloride (Cl-), fluoride (F-), nitrate (NO3-), phosphate (PO42-), bicarbonate (HCO3-) and sulphate (SO42-) were determined by ion chromatography (IC) (Dionex Model DX 500). Results from this study revealed that the geothermal springs were rich in trace elements compared to that from non-geothermal source of water. The mineral elements present were vii mainly due to rock-water interaction in the deep aquifer at both sites. The geothermal spring water is not fit for drinking because it is particularly high in fluoride (F) having 6.66 and 5.97 mg/L at Siloam ; 6.72 and 7.28 mg/L at Tshipise for winter and summer, respectively. Also, high Nickel (Ni) with 462 µg/L and 868 µg/L: Lead (Pb) with 652 µg/L and 211 µg/L at Siloam and Tshipise respectively, for summer season. In addition, it is not suitable for irrigation owing to high sodium absorption ratio (SAR) values which were above the standard guidelines (˂1) by South African Bureau of Standards (SABS) and World Health Organization (WHO) at both sites. In summer season, there were higher trace elements concentrations than in the winter season. The higher concentration values could be attributed to rainfall, which aids in the dissociation of rock particles, resulting in higher concentrations of these elements. Siloam spring water was more mineralised than Tshipise spring water, hence its neighbouring soils and vegetation possess more trace elements concentrations than the latter. Owing to their high mineral elements content, the geothermal spring water flows across the soil, making it vulnerable to sorption of the trace elements. The trace elements present in the surrounding soil of the geothermal spring were as a result of geothermal water and soil pedogenesis. The geothermal water contaminates the surrounding soil with substantial quantity of trace elements, which decreases with the distance from the geothermal spring, making far distanced soil less-contaminated. High levels of Cr, Co, Ni, Cu, Zn and Pb at Siloam soil can be attributed to more minerals present in the spring, therefore making absorption by Mangifera indica inevitable. Soils at Tshipise are moderately concentrated owing to moderate trace elements concentrations from the geothermal spring water. Generally, seasonal variations were observed in the parameters analysed in the geothermal spring water, surrounding soil and vegetation to ascertain the most favourable season in terms of the trace elements concentrations. There were higher concentrations of trace elements in the geothermal spring, particularly during the summer season, compared to the winter season; this leads to more contamination of the surrounding soils and vegetation. This study showed that geothermal spring has potential to enrich the neighbouring soils and vegetation with trace elements, which could result in contamination. It can be concluded that geothermal spring, despite its benefits to humans, also contaminates the surrounding surface soils with toxic trace elements. Soils are a platform for vegetation. Therefore, if the soil is contaminated by toxic elements, there are high possibilities that these trace elements are absorbed by the neighbouring vegetation, which is likely to affect human beings adversely.