INTRODUCTION AND SCOPE OF INVESTIGATION Irrigated land in South Africa currently amounts to approximately 1.3 million hectares. Agricultural water use is estimated to comprise the largest amount of water users in Southern Africa, with much of the region dependent on sufficient water of adequate quality for survival. The Vaalharts is the largest irrigation scheme in South Africa. Approximately 32 000 hectares of land is currently being irrigated. The salinity of the irrigated water has steadily increased over time (Herold and Bailey, 1996). Several research projects have been undertaken to determine the fate of added salts. The conclusion in these reports is that a very large proportion of the salts added to the subsurface due to irrigation are not returned to the surface water. A sink for these salts is therefore believed present. Research into the salt balance of the area and the effect of the salts on soils and crops have suggested that the majority of this salt is being leached through the soil and into the groundwater resources underlying the irrigated area. The underlying aquifer was believed to have a limited storage capacity. Once this capacity is exceeded, a flow reversal is expected. This process is likely to add a tremendous salt load (roughly estimated to be in the order of 100000t/annum (Herold and Bailey, 1996)) to an already stressed river system. The adverse effects of such an addition would be catastrophic to the irrigation scheme and the receiving aquatic environment. This thesis aims to determine the processes leading to the scenario outlined above.

The objective of this research is to characterise fractured aquifers contaminated with DNAPLs in South Africa. The site characterisation approach is conducted in two case studies by integrating a range of site assessment techniques. It started with the noninvasive methodologies that provide rapid and cost-effective ways for screening. After developing the initial conceptual model, the investigation proceeded to invasive techniques that provide more information on DNAPLs and fractures on a local scale. This research indicated that fracture âhuntingâ is extremely difficult with the current technology, making DNAPL characterization very challenging. It also indicated that no single technique provides an unequivocal indication of fracture position and orientation, but that best results were obtained when several invasive techniques were used in a toolbox approach. These tools were most appropriate when integrated with hydraulic testing methods. The site characterisation was done in a continuous and iterative process, and each phase of investigation was used to refine the conceptual model of the site. The non-invasive techniques applied included: ⢠General assessment of Test Site 1 by site inspection, interviewing employees, documentation studies and aerial photo analysis. ⢠Hydrocensus ⢠Direct observing for DNAPL contamination using UV light and Sudan IV shake tests. ⢠Soil gas surveys by using various models of PIDs to delineate the organic vapour plumes. ⢠Surface and airborne geophysics for outlining major structures that may play an important role for DNAPL migration pathways. This did not yield useful results at Test Site 1 due to noise interference from industrial activities and infrastructure. The non-invasive techniques were not applicable for direct evaluation of DNAPLs migrating in fractured aquifers; however, they were essential in understanding DNAPL-release zones and vapour plumes. They were also found to be relatively cost and time efficient. The invasive techniques applied include: ⢠Test pits (excavations on the top 3 m soil) ⢠Drilling using augur, percussion and coring methods. (An outside-in approach was followed during the drilling to avoid unnecessary DNAPL mobilisation). ⢠Several different borehole geophysical and geochemical logging methods. Results from this diverse range of activities were integrated to construct conceptual models of the preferential fracture pathways of the two test sites. Riemann (2002) found that early time drawdown data could be used to estimate the Tvalue of the fracture zone with the Cooper-Jacop 2 method. However, in this research, the method did not yield results that were consistent with other observations. The tracer experiment was extended to evaluate the effective matrix diffusion coefficient of the mudstone in the Campus Test Site. The values at depths of 27 and 36 m were estimated as 3.4 x 10-6 and 7.8 x 10-6 m2/day, respectively. This means that the rate of dissolved mass disappearance from the fracture to the matrix is relatively high compared to other geologic formations. An in situ method of effective porosity estimation using tracer tests was employed instead of the traditionally measurements on rock/soil samples from the aquifer. In this way, uncertainty in porosity estimation due to the inherent alteration during sampling was eliminated. Column experiments were conducted to evaluate the accuracy of PITT for residual DNAPL saturation estimation. The saturation was estimated with an accuracy of 95%. However, the experiment was conducted in unconsolidated, homogenous sand and the experiment may not be accurate as such when applied to heterogeneous aquifer systems usually encountered in South Africa. With the current technologies available, PITT is believed to have limited applicability in fractured rock aquifers. Finally, based on the findings, a methodology has been proposed for the characterisation strategy of DNAPL contaminated sites in fractured South African aquifers.

Not available

Water in South Africa is becoming a scarce and important resource and therefore has to be managed and protected in order to ensure sustainability, equity and efficiency. The SAGDT is designed to provide methods and tools to assist groundwater professionals and regulators in making informed decisions concerning groundwater use, management and protection, while taking into account that groundwater forms part of an integrated water resource. The SAGDT is spatially-based software, which includes: ⢠A GIS interface to allow a user to import shape files, various CAD formats and georeferenced images. The GIS interface also provides for spatial queries to assist in the decision-making process. The GIS interface contains default data sets in the form of shape files and grid files depicting various hydrogeological parameters across South Africa. ⢠A risk assessment interface introduces fuzzy logic based risk assessments to assist in decision making by systematically considering all possibilities. Included risk assessments relate to the sustainability of a groundwater resource, vulnerability of an aquifer, pollution of a groundwater resource (including seawater intrusion), human health risks associated with a polluted groundwater resource, impacts of changes in groundwater on aquatic ecosystems and waste site impact on an area. ⢠Third-party software such as a shape file editor, an interpolator, a georeference tool, a unit converter and a groundwater dictionary. ⢠A report generator, which automatically generates documentation concerning the results of the risk assessment performed and the input values for the risk assessment. ⢠A scenario wizard for the novice to obtain step by step instructions in setting up a scenario. All case studies presented in this thesis is available in the scenario wizard. ⢠The SAGDT allows problem solving at a regional scale or a local scale, depending on the problem at hand. This thesis discusses the origin, research, development and implementation of the SAGDT. Case studies are included to demonstrate the working of the SAGDT. They include: ⢠Vunerability (Fish River Lighthouse) ⢠Waste Site (Bloemfontein Suidstort) ⢠Sustainability (De Hoop) ⢠Mine (Van Tonder Opencast) The SAGDT relies heavily on the expertise of hydrogeologists, assumptions and approximations of real world conditions. Together with the heterogeneities present in groundwater systems it is impossible to guarantee the accuracy of the methodologies and this must be taken into consideration.

Groundwater has been identified as a viable alternative for future freshwater production in South Africa. The management thereof is steadily gaining more recognition from governmental institutions. A significant obstacle in the development of this resource is the conceptual understanding of surface water and groundwater interaction. The availability of reliable data for rainfall, flow volumes in rivers and water levels in boreholes have prompted an investigation into patching incomplete data sets. This study also focused on predicting the influence of rainfall and flow volumes in a river on the surrounding groundwater levels. Neural networks have been used to investigate both data patching and forward prediction of water levels in selected data sets. Uittreksel

Climate change is major threat to our world particularly poor countries. Climate change is driven by changes in the atmospheric concentrations of Greenhouse Gases and aerosols. These gases affect the absorption, scattering and emission of radiation within the atmosphere and the earthâs surface thus resulting in changes in the energy balance (IPCC, 2007). Much strain will be placed on water resources especially in areas where water infrastructure does not exist, or where water delivery is difficult due to aridity (Pietersen, 2005). This dissertation examines the causes of climate change and explores the resulting effects on the environment, social and economic sectors. This study focuses its attention on South Africa as a whole. South Africa is viewed as a waterâstressed country with an average annual rainfall of 500 mm and any climatic change could have adverse impacts on water resources of the country. The potential impacts of climate change on water resources and hydrology for Africa and Southern Africa have received considerable attention from hydrologists during the last decade. Very little research has been conducted on the future impact of climate change on groundwater resources in South Africa. Climate change can affect groundwater levels, recharge and groundwater contribution to baseflow. The first step in the approach involves the creation of a climate change groundwater vulnerability profile. In analogy with the DRASTIC methodology the DART methodology was developed. The parameters considered in the DART methodology are as follows: D â Depth to water level change A â Aquifer type (storativity) R â Recharge T â Transmissivity The DART methodology focus more on typical parameters used in sustainability studies, but also indirectly accommodate the issue of quality due to the fact that the water quality is likely to deteriorate with a drop in water level over time as the salt load will concentrate. Two scenarios are considered; current and future. The current scenario is representative of the current precipitation patterns and the future scenario is representative of the predicted scenario based on the selected GCM. Vulnerability indices are developed to assess the impacts of global change at spatial scale to enhance the understanding of impacts on people, and develop the appropriate policies for adaptation. For the purpose of this study, a vulnerability index was developed to assess the impacts of climate change on groundwater resources of South Africa on rural communities. At first glance, the results indicate there is not a significant difference between the current and future average indices, which indicates that climate change impacts on groundwater have very little impact on communities and therefore few adaptation requirements are necessary for community impacts due to groundwater impacts based on climate change.

Acid mine drainage is by far the most significant long term groundwater quality impact associated with both opencast and underground coal mining, in both a local and international context. The modern day geohydrologist has access to numerous tools, which can be used to determine important decant issues â issues ranging from when decanting will begin to occur, and the volumes of water that are expected to decant. The continuous development and improvement of numerical groundwater flow models is steadily leading to an increasing dependence on them. The main aim of the thesis was to determine whether there exists any correlation between modern day numerical groundwater flow models and analytical calculations, and the presentation of a toolbox of tools that may be used for decant related issues. The following conclusions were drawn after numerous numerical and analytical scenarios and statistical correlations were performed: ⢠Given the amount of uncertainty regarding aquifer heterogeneity, there do exist a good correlation between the numerical and analytical groundwater decant volume estimations, ⢠An increase in the effective porosity of the backfilled opencast pits cause an increase in the time-to-decant, as more water is required to fill the pits to their decant elevations, ⢠An increase in the effective aquifer recharge cause an increase in the decant volumes and a decrease in the time-to-decant, because more water is available to fill the pits to their decant elevations, ⢠The effective aquifer recharge is a very sensitive parameter (more so than specific yield, storage coefficient, and transmissivity), as significant decreases in the time-to-decant were simulated with an increase in the aquifer recharge, as were significant increases in decant volumes simulated with an increase in recharge, ⢠The volumes of groundwater decant are more sensitive to variations in the transmissivity of the surrounding aquifer/s compared to the transmissivity of the backfilled opencast pits, ⢠During the numerous flow model scenarios it was found that the groundwater contribution to pit water is far less compared to the recharge component. The above conclusions prove that there are still applications for analytical calculations in modern day geohydrology, despite the continuous development of numerical groundwater flow models. Based on experience in similar coal mining operations within the Mpumalanga coal fields, the results of both the analytical decant volume and time-to-decant estimations correspond well with actual figures. One must however understand and master the various equations and keep in mind that an aquifer is a highly heterogeneous system. The results of both numerical groundwater flow model simulations and analytical calculations are only as good as the understanding of the geohydrological environment and the data they are based on.

Groundwater contamination as a result of Light Non Aqueous Phase Liquid (LNAPL) releases into the subsurface is a widespread occurrence across South Africa which threatens current and future water resources within the country. Groundwater contaminant fate and transport modelling are common elements of hydrogeological investigations and remedial design methodologies in many developed countries where the models are use as management and decision making tools. In South Africa this is not the case, with contaminant flow and transport modelling rarely being employed as part of LNAPL contamination investigations. Over the last 3 years the Beaufort West study area has had extensive investigative work carried with regards to the determination and delineation of LNAPL related groundwater contaminant plumes which are present underneath a significant portion of the town. As a result an extensive data set has been generated with regards to aquifer geometry, fracture network distribution, aquifer parameters and contaminant plume concentrations and extent. The dataset should in theory provide an opportunity to construct a groundwater contaminant fate and transport model for the area as a remedial management tool. By means of collating previously existing data through a comprehensive desktop study, and supplementing this data with a toolkit of field investigations techniques such as diamond barrel core drilling, percussion drilling, electrical conductivity profiling, fluid electrical conductivity profiling, aquifer pump testing, and low flow inorganic and organic groundwater sampling, the conceptual model of the study area was updated and refined to a point where feasibility of constructing a groundwater contaminant fate and transport model could be assessed. Based upon the conceptual understanding of the study area as defined in the conceptual model developed in the study, a groundwater contaminant fate and transport model is not considered feasible for the study area with body of data currently available. This is attributed mainly to the high level of complexity of the observed natural environment and the challenges in acquiring acceptable quality field data such as aquifer parameters given the uncontrolled pumping environment which is present due to the high number of private groundwater users. Potentially an even greater detractor to the construction of a model, is that considering the conceptual understanding of the study area, there are very few questions of significance whose answers could be provided by a model, and this would indicate that a model would not be an effective remedial management or decision making tool in the current scenario.

The fertilizer production facility had a negative impact on the soil, groundwater and surface water environment due to the handling / storage and production activities at the site. Observations and numerical modelling found the fertilizer product loadings areas as the main source area of contaminants viz. Ca, Mg, NO3, Cl, SO4, EC and TDS. Uncontrolled run-off emanating from the site is a major contributing factor to contaminating the groundwater and surface water resources. A distinct difference could be observed between the geochemical signature of the potential contaminated seepage and that of the groundwater. This geochemical characterisation of the contaminant plume identified an interaction of the leachate and the soil with a high clay (montmorillonite) content, with various cation exchange and sorption processes occurring. Potassium is largely exchanged (for sodium), while phosphates are likely to sorbed on the clay particles. Nitrate is likely to be retarded to a limited extent, especially when redox conditions are conducive to the conservation of the nitrate specie. The elevated contaminant concentrations pose a health risk to potential users and livestock which may ingest the water, especially nitrate concentrations. Numerical modelling was used to validate and develop the site conceptual model. Iterative modelling improved the initial correlation R2 of modelled and observed nitrate concentrations, the correlation improved from 0.29 to 0.64. The model was validated by assuming that horizontal and bedding plan fractures are likely to play a role in contaminant transport (which was not modelled). Artificial recharge (seepage and leachate infiltration) was present at the plant area. Groundwater abstraction from farmers boreholes downstream had an influence on the development of the nitrate plume. Surface water contamination contributed to the current plume geometry and therefore partly responsible for the current plume extent. A secondary groundwater contaminant source was found in the south western part of the study area. Predictive modelling found abstraction of groundwater from site to be the most effective containment measure when compared to a cut-off trench. The groundwater contamination is likely to pose a low current and future risk to groundwater users, as no current groundwater users are found in proximity to the site and the contaminant plume. However a potential surface contaminant risk does occur to down stream surface water bodies during a flood event.

The Waterberg coalfields represent the last area in South Africa, which contain large quantities of coal resources. According to Dreyer (pers. comm. 2009) the Waterberg coalfields contain nearly 50% of the remaining coal resources of South Africa. Given the great demand for coal both local and abroad, primarily to be used as a fuel source for the power generation, the Waterberg coalfields have been targeted for large scale developments in order to exploit the coal. The primary method for exploiting coal is through mining. Mining, in any setting and any location has a diverse and often very serious impact on the environment. A scoping level study was under taken in order to determine the effect the mines will have on the groundwater resources and the pre-mining conditions of the aquifers and the quality of the groundwater. At present there is one operational colliery in the study area, the Grootegeluk mine. This mine has been in operation since the 1980âs and has had a well planned and operated monitoring system in place since the beginning of mining operations. This mine was used as a model to determine the impact new mines will have on the area. From the investigations it became apparent that the coalfield is situated in the Karoo Supergroup geology with the Mokolian Supergroup being represented in the study area by the Waterberg group quartzites. The coalfield is delineated by three major geological structures, the Daarby-, Eezaamheid- and the Zoetfontein faults. With the Daarby- and Eenzaamheid faults being impermeable according to Dreyer (pers. comm. 2009), The Daary fault serves to divide the study area into an area west of the fault with shallow coal and an area east of the fault with deeper coal. Only the shallow coal will be mined. According to Dreyer (pers. comm. 2009), all of the planned infrastructure for the new mines will be located on the Waterberg group rocks south of the Eenzaamheid fault or on the Karoo rocks east of the Daarby fault. To determine the impact the mines would have on the groundwater of the study area, aquifer parameter testing (pumping test and slug tests), water quality determinations (inductively coupled spectrometry), acid-base accounting and numerical modelling were conducted. The results of the aquifer testing indicated low yielding aquifers with the harmonic mean of the transmissivities indicating a low transmissivity of 0.4 m2/d. In addition the recharge for the study area was calculated by means of the Cl and E.A.R.T.H. methods, resulting a value of 1.5% for the area. The average water level for the area was found to be approximately 28 m. The water quality determinations for areas that had not been affected by mining, indicated waters that had high EC values, near neautral pH value and medium to high Cl and sulphate values. The areas that have been affected by activities such as power generation and mining, displayed higher EC, Cl, and sulphate values than the unaffected areas. To more accurately determine the impact the mines would have on the area, numerical modelling was done. Three scenarios were simulated using similar parameters to determine the expected inflow into the mines and whether the mines would ever decant. The results indicated that the worst possible scenario there was an influx varying between 755 m3/d and 1283 m3/d depending on the location of the pits. For the decant models, 50 years after mining had stopped there was a rise of 3 m in the pits themselves. With the pits being simulated being 110 m deep it is concluded that the mines in the area will never decant. The results of the project indicate that the addition of new mines to the area will have an effect on the groundwater quality and quantity and steps should be taken to minimise this as much as possible.