Conference Abstracts

Abstract

Tailings storage facilities are significant contributors of dissolved solids to underlying aquifers and adjacent watercourses. Salt balances indicate estimated seepage loads of the order of 1 500 tonnes of chloride per year. Actual seepage loads will be determined by the hydraulic conductivity of the tailings and mechanisms of flow within the tailings. Field observations and sample analytical results from several platinum tailings facilities are presented. These indicate the development of lenses of clay sized material within coarser silty material and suggest a tortuous seepage flow path, perhaps characterised by zones of preferential flow. The implications of seepage modelling and geochemical data on the salt loads mobilised from tailings are discussed. Results suggest that tailings facilities are effective at retaining salts and that release of accumulated salts after closure may take place at long time scales. {List only- not presented}

Abstract

The complexity of real world systems inspire scientists to continually advance methods used to represent these systems as knowledge and technology advances. This fundamental principle has been applied to groundwater transport, a real world problem where the current understanding often cannot describe what is observed in nature. There are two main approaches to improve the simulation of groundwater transport in heterogeneous systems, namely 1) improve the physical characterisation of the heterogeneous system, or 2) improve the formulation of the governing equations used to simulate the system. The latter approach has been pursued by incorporating fractal and fractional derivatives into the governing equation formulation, as well as combining fractional and fractal derivatives. A fractal advection-dispersion equation, with numerical integration and approximation methods for solution, is explored to simulate anomalous transport in fractured aquifer systems. The fractal advection-dispersion equation has been proven to simulate superdiffusion and subdiffusion by varying the fractal dimension, without explicit characterisation of fractures or preferential pathways. A fractional-fractal advection-dispersion equation has also been developed to provide an efficient non-local modelling tool. The fractional-fractal model provides a flexible tool to model anomalous diffusion, where the fractional order controls the breakthrough curve peak, and the fractal dimension controls the position of the peak and tailing effect. These two controls potentially provide the tools to improve the representation of anomalous breakthrough curves that cannot be described by the classical-equation model. In conclusion, the use of fractional calculus and fractal geometry to achieve the collective mission of resolving the difference between modelled and observed is explored for the better understanding and management of fractured systems.

Abstract

This study aims to contribute to the conceptual and methodological development of units of joint management in transboundary aquifers (TBAs) to prevent and mitigate cross-border groundwater impacts (GWIs) in quantity and/or quality. Joint management units are a relatively new but growing topic in the field of TBAs, and their conceptualisation and appropriate identification are still at an early stage. By reviewing the literature on the subject and elaborating on its terminology, main features, and current methodological progress, a comparison of the existing methodologies for identifying such units is analysed. On this basis, trends and recommendations for further research and application of such methodologies to the joint management of TBAs are presented. The literature on this issue is scarce and has been published mainly in the last five years. These publications lack consistency in the use of concepts and terminology. The above has led to miscommunication and semantic issues in the concept behind such units and in comprehending the particular challenges of identifying them. Still, some directions and methodologies for identifying or directly delineating these management units have been proposed in the literature. However, no analysis from these methodological attempts has been conducted; thus, there are no lessons to be learned about this progress. This research looks forward to closing these gaps and making headway toward dealing with cross-border GWIs in TBAs, thus helping countries meet international law responsibilities and maintaining stable relationships among them.

Abstract

Underground coal gasification (UCG) is considered a cleaner energy source as its known effect on the environment is minimal; it is cheaper and a lesser contributor to greenhouse gas emissions when compared to conventional coal mining. It has various potential impacts but the subsidence of the surface as well as the potential groundwater contamination is the biggest concerns. Subsidence caused by UCG processes will impact on the groundwater flow and levels due to potential artificial groundwater recharge. The geochemistry of the gasifier is strongly depended upon site specific conditions such as coal composition/type and groundwater chemistry. Independent of the coal rank, the most characteristic organic components of the condensates is phenols, naphthalene and benzene. In the selection of inorganic constituents, ammonia, sulphates and selected metals and metalloids such as mercury, arsenic, and selenium, are identified as the dominant environmental phases. The constituents of concern are generated during the pyrolysis and after gasification as dispersion and penetration of the pyrolysis take place, emission and dispersion of gas products, migration by leaching and penetration of groundwater. A laboratory-based predictive study was conducted using a high pressure thermimetric gasification analyser (HPTGA) to simulate UCG processes where syngas is produced. The HPTGA allows for simulation of the actual operational gasifier pressure on the coal seam and the use of the groundwater sample consumed during gasification. A gasification residue was produced by gasifying the coal sample at 800 °C temperature and by using air as the input gas. The gasification residue was leached using the high temperature experimental leaching procedure to identify the soluble phases of the gasified sample. The leachate analysis is used to determine the proportion of constituents present after gasification which will be removed by leaching as it is exposed to external forces and how it will affect the environment. The loading to groundwater for the whole gasifier is then determined by applying the leachate chemistry and rock-water ratio to the gasifier mine plan and volumes of coal consumed. 

Abstract

The electrical resistivity tomography (ERT) method has become one of the most commonly used geophysical techniques to investigate the shallow subsurface, and has found wide application in geohydrological studies. The standard protocols used for 2D ERT surveying assume that the survey lines are straight; however, due to the presence of infrastructure and other surface constraints it is not always possible to conduct surveys along straight lines. Previous studies have shown that curved and angled survey lines could impact on the recorded ERT data in the following ways: 1) the true geometric factors may differ from the assumed geometric factors and thus affect the calculated apparent resistivities, 2) the depths of investigation may be overestimated, and 3) the recorded apparent resistivities may be representative of the subsurface conditions at positions laterally displaced from the survey line. In addition, previous studies have shown that although the errors in the apparent resistivities may be small even for large angles and curvatures, these errors may rapidly increase in magnitude during inversion. In this paper we expand on the previous work by further examining the influence of angled survey lines on ERT data recorded with the Wenner (?) array. We do this by: 1) calculating the changes in the geometric factors and pseudo-depths for angled survey lines, 2) forward and inverse modelling of ERT datasets affected by angled survey lines, and 3) examining the impact of angled survey lines on real ERT datasets recorded across different geological structures.

Abstract

Millions of tons of coal ash are produced across the globe, during coal combustion for power generation. South Africa relies largely on coal for electricity generation. The current disposal methods of coal ash are not sustainable, due to landfill space limitations and operational costs. One way/means of disposing of coal ash that could provide environmental and financial benefits; is to backfill opencast mines with the ash. However, a limited number of studies have been conducted to assess the feasibility of this method in South Africa. Thus the aim of the experiment is to monitor bulk ash disposal under field conditions to improve the understanding of the geochemical and hydrogeological processes occurring during the actual deposition of coal ash in opencast coal mines. To achieve the aim (1) a gravity lysimeter will be built containing both mine spoils and coal ash representing field conditions; (2) the factors (CO2, water level and moisture content) affecting acid mine drainage will be monitored in the lysimeter and (3) the change in the quantity and quality of the discharge released from the lysimeter.

Abstract

There are various software packages used by hydrogeologists for a variety of purposes ranging from project management, database management, data interpretation, conceptual and numerical modelling and decision making. Software is either commercial (produced for sale) or open source (freely available to anyone and for any purpose).

The objective of this paper is to promote open source software that can be used by the hydrogeological community to reduce expenses, enhance productivity and maximise efficiency.

Free software was previously associated as being inferior in quality in the corporate world. Companies often use commercial software at a hefty price, but little do they know that open source is often equal to, or superior to their commercial counterparts. The source code of open source software can freely be modified and enhanced by anybody. Open source software is a prominent example of open collaboration as it is developed by users for the user community. Companies using open source software do not need to worry about licensing and do not require anti-piracy measures such as product activation or a serial number.

However, the decision of adopting open source software should not just be taken just on the basis of the low-cost involved. It should entail a detailed analysis and understanding of the requirements at stake, before switching to open source to achieve the full benefits it offers and to understand what the down side is. There are plenty of open source products that can be used by hydrogeologists. The packages considered in this article are those that are frequently used by the author and do not necessarily mean that they are the best available. Software gets updated or abandoned with time and what is considered powerful today may be obsolete in a few years.

Some of the well-known open source packages recommended for hydrogeologists include: OpenLibre for project management, Blender 3D or Sketchup for 3D conceptual modelling, QGIS for GIS mapping and database management, SAGA GIS for interpolation and ModelMuse for numerical modelling (comprising of Modflow for finite difference, Sutra for finite element and Phast for geochemical modelling). In addition, there are a number of free software packages developed by the USGS, various universities and consultants across the globe that can be used for aquifer test interpretation, borehole logging and time-series data analysis. A saving of more than R250,000 can be made per hydrogeologist by utilising such open source packages, while maintaining high quality work that is traditionally completed using commercial software.

Abstract

A groundwater assessment was conducted to identify and predict the contamination and transport properties of a groundwater system. The motivation for the study was the rising concern of a farm owner about the deteriorating water quality of the aquifer system. An investigation of the surface and groundwater quality indicated that two fertilizer dumpsites were the sources of pollution. Water analyses revealed elevated concentrations of Ca, Mg, K, F, NO3, SO4, Mn and NH4 within boreholes near the pollution sources. The NH4 and NO3 concentrations were exceptionally high: 11 941 mg/L and 12 689 mg/L, respectively. These high concentrations were the direct result of the dumping of fertilizer. The rise in these concentrations may also have been catalysed by the nitrogen cycle and the presence of the Nitrosomonas bacterium species. Due to the high solubility of NO3, and because soils are largely unable to retain anions, NO3 may enter groundwater with ease, and could migrate over large distances from the source. Elevated NO3 in groundwater is a concern for drinking water because it can interfere with blood-oxygen levels in infants and cause methemoglobinemia (blue-baby syndrome). A geophysical study was undertaken within the area of investigation to gain insight on the underlying geological structures. The survey indicated preferential flow paths within the aquifer system along which rapid transport of contaminant is likely to occur.
Key words: aquifer system, groundwater quality analyses, fertilizer, nitrogen cycle, Nitrosomonas species, geophysics.

Abstract

Hydraulic behaviour of an aquifer is defined in terms of the volumes of water present, both producible and not (specific yield and specific retention), and the productivity of the water (hydraulic conductivity). These parameters are typically evaluated using pumping tests, which provide zonal average properties, or more rarely on core samples, which provide discrete point measurements. Both methods can be costly and time-consuming, potentially limiting the amount of characterisation that can be conducted on a given project, and a significant measurement scale difference exists between the two. Borehole magnetic resonance has been applied in the oil and gas industry for the evaluation of bound and free fluid volumes, analogous to specific retention and specific yield, and permeability, analogous to hydraulic conductivity, for over twenty years. These quantities are evaluated continuously, allowing for cost-effective characterisation, and at a measurement scale that is intermediate between that of core and pumping tests, providing a convenient framework for the integration of all measurements. The role of borehole magnetic resonance measurements in hydrogeological characterisation is illustrated as part of a larger hydrogeological study of aquifer modeling. Borehole magnetic resonance has been used for aquifer and aquitard identification, and to provide continuous estimates of hydraulic properties. These results have been compared and reconciled with pumping test and core data, considering the scale differences between measurements. Finally, an integrated hydrogeological description of the target rock units has been developed.

Abstract

Big data analytics (BDA) is a modern and innovative platform of applications that include advanced analytical techniques such as data mining, statistical analysis, artificial intelligence, machine learning, and natural language processing. Regional data are generated through groundwater monitoring, remote sensing applications or global circulation models (GCM), however this is often too course for a local understanding. Groundwater managers rely on locally relevant information for effective operational decision making, however this is often missing. A Transboundary Aquifer (TBA) Analytic Framework was developed to match, integrate and model local hydrogeological data with regional earth-observation data using BDA. Drawing on the literature on BDA, a reference architecture for the TBA analytical framework was identified for application to various groundwater management scenarios in the Ramotswa Dolomitic Aquifer (Botswana - South Africa) and Shire Valley Alluvial Aquifer (Malawi - Mozambique). The TBA analytical framework allows for local clouds to store the local and regional structured and unstructured datasets and interconnecting these local clouds through a federated cloud infrastructure. In this regard, tools that are incorporated in the TBA analytical framework include data ingestion operators, data transformation operators, and feature extractors. Various machine learning algorithms and statistical techniques are incorporated in the TBA analytical framework to downscale the regional datasets. The downscaling involves selection of potential predictors and predictants variables based on data needs to address local groundwater management scenarios such as regulating groundwater abstraction to prevent groundwater depletion. Using the downscaled data the TBA analytical framework can be utilised to uncover patterns and statistical relationships in the datasets in order to model local groundwater processes such as cone of depression, groundwater levels forecasting, well protection zoning, amongst others.

Abstract

The 'maintainable aquifer yield' can be defined as a yield that can be maintained indefinitely without mining an aquifer. It is a yield that can be met by a combination of reduced discharge, induced recharge and reduced storage, and results in a new dynamic equilibrium of an aquifer system. It does not directly or solely depend on natural recharge rates. Whether long-term abstraction of the 'maintainable aquifer yield' can be considered sustainable groundwater use should be based on a socio-economic-environmental decision, by relevant stakeholders and authorities, over the conditions at this new dynamic equilibrium.
This description of aquifer yields is well established scientifically and referred to as the Capture Principle, and the link to groundwater use sustainability is also well established. However, implementation of the Capture Principle remains incomplete. Water balance type calculations persist, in which sustainability is linked directly to some portion of recharge, and aquifers with high use compared to recharge are considered stressed or over-allocated. Application of the water balance type approach to sustainability may lead to groundwater being underutilised.
Implementation of the capture principle is hindered because the approach is intertwined with adaptive management: not all information can be known upfront, the future dynamic equilibrium must be estimated, and management decisions updated as more information is available. This is awkward to regulate.
This paper presents a Decision Framework designed to support implementation of the capture principle in groundwater management. The Decision framework combines a collection of various measures. At its centre, it provides an accessible description of the theory underlying the capture principle, and describes the ideal approach for the development operating rules based on a capture principle groundwater assessment. Sustainability indicators are incorporated to guide a groundwater user through the necessary cycles of adaptive management in updating initial estimations of the future dynamic equilibrium. Furthermore, the capture principle approach to sustainable groundwater use requires a socio-economic-environmental decision to be taken by wide relevant stakeholders, and recommendations for a hydrogeologists' contribution to this decision are also provided. Applying the decision framework in several settings highlights that aquifer assessment often lags far behind infrastructure development, and that abstraction often proceeds without an estimation of future impacts, and without qualification of the source of abstracted water, confirming the need for enhanced implementation of the capture principle.

Abstract

Monitored Natural Attenuation (MNA) refers to the monitoring of naturally occurring physical, chemical and biological processes. Three lines of evidence are commonly used to evaluate if MNA is occurring, and this paper focusses on the second line of evidence: The geochemical indicators of naturally occurring degradation processes and the site-specific estimation of attenuation rates.

The MNA geochemical indicators include the microbial electron acceptors (e.g. dissolved oxygen, nitrate and sulphate) and the metabolic by-products (manganese (II), iron (II) and methane). In addition, redox and alkalinity are important groundwater indicators. So as to properly assess the geochemical trends a groundwater monitoring well network tailored to assessing and defining the contaminant plume is required.

The expressed assimilative capacity (EAC) is used to estimate the capacity of the aquifer to degrade benzene, toluene, ethylbenzene and xylene (BTEX compounds) using the concentrations of geochemical indicators. Using the EAC, the groundwater flow through a perpendicular cross-section of the source area, and the source mass, the life of the contaminant source can be made.

A practical example of the performance monitoring of MNA using geochemical parameters is described for a retail service station in KwaZulu-Natal, which has groundwater impacted by a petroleum hydrocarbon plume. This includes a description of the monitoring well network, the geochemical measurements, the calculation of the EAC, and the estimated life of the contaminant source.

Abstract

Mining site remnants are everlasting and impact the groundwater regime on a long term scale. An integrated approach to geoscience is necessary due to the complexity of nature and the unknown relationships that must be discovered to further the understanding of impacts on the natural environment. Furthermore, groundwater resources are negatively impacted by mining activities affecting the groundwater quality and quantity. Underground coal mining can be accompanied by roof failure events. This may change the matrix which subsequently alters the flow regime; leads to variations within the water chemistry, provided there is inter- aquifer connectivity; and alters the recharge rate. Dewatered mine voids are in direct contact with oxygen initiating oxidation reactions, depending on the geology of the specific site. A change in water chemistry was analyzed, and this coincides with a roof failure event as interpreted from water level measurements. Concentrations of Mg, Ca, and alkalinity indicate anomalous changes that are still in effect, five to six years after the majority of water levels had stabilized. The changes in the system coincides with and correlates to events of roof failure and different parameters. The latter changes are applied as extra tools when interpreting different site specific anthropogenic induced impacts on the system. Also within this study, constant rate pumping tests were conducted for the interest of the hydraulic properties, using three farming boreholes. The results put forward a range of 0.21 – 0.44L/s and 6.5 – 11.5m2 /d, for sustainable yield and transmissivity, respectively. Furthermore, it is recommended that a better understanding can be gained on system behaviors if chemistry correlations can be gathered through certain events causing specific systems to be in disequilibrium. It is also recommended that additional pumping tests will allow more insightful interpretation and delineation between the abovementioned chemical and water level changes. Finally, the combination of parameters during events can aid in deciding the most appropriate analytical models used for further analysis.

Abstract

In the management of water resources especially groundwater resources, implementing existing regulations is one of the much needed aspects ensuring water security through the regulated use. However, such regulations are not regulated to ensure that they served the intended purpose in their original formulation. In South Africa, a study was carried out to assess the relevance and efficient of adhering to procedural requirements during water use licence application (WULA) process. Lived-experiences and observation methods were used to collect data. The department of water and sanitation was used as a case study. Interpretative analysis approach was used to provide the meaning on the analysed information. The WARMS database was accessed where the number of days that WULA process was extracted. The regulation No. 40713 about WULA process was analysed. The five-year-data prior and post the promulgation of regulation No. 40713 were extracted from WARMS database and evaluated in terms of the duration each application took to be processed for WULA. Data on water use for abstractions from all the regions were obtained from WARMS database and assessed. Dates when applications were submitted and when such applications were finalised were analysis per month and per years for temporal analysis. The number of entitlements received during the particular period and the number of applications recommended to be declined and issued were assessed using exploratory data analysis methods. Graphical method was adapted to increase results visualisation on water use entitlements. Key results showed that the process of WULA was generally slow and reasons were provided for such outcome. However, the temporal analysis revealed an increasing trend in the post promulgation of regulation No. 40713 suggesting that regulations when re-regulated serve its intended purpose. Although such findings are not conclusive but they inform a basis for re-regulating enforcement regulations in Southern African countries with issues similar to South Africa on water entitlement.

Abstract

Groundwater is an important resource for multiple uses in South Africa. Hence, setting limits to its sustainable abstraction while assuring basic human needs is required. Due to prevalent data scarcity related to groundwater replenishment, which is the traditional basis for estimating groundwater availability, the present article presents a novel method for determining allocatable groundwater in quaternary catchments through information on streamflow. Using established methodologies for assessing baseflow, recession flow, and instream ecological flow requirement, the methodology develops a combined stepwise methodology to determine annual groundwater storage volume using linear reservoir theory, essentially linking low flows proportionally to upstream groundwater storages. The approach was trialled for twenty-one perennial and relatively undisturbed quaternary catchments with longterm and reliable streamflow records. Using the Desktop Reserve Model, maintenance low instream flow requirements necessary to meet present ecological state of the streams were determined, and baseflows in excess of these flows were converted into allocatable groundwater storages on an annual basis. Results show that groundwater development potential exists in nineteen of the catchments, with upper limits to allocatable groundwater volumes (including present uses) ranging from 0.02 to 2.60 Mm3/a over the catchments. With a secured availability of these volumes 75% of the years, variability between years is assumed to be manageable. A significant (R2 = 0.86) correlation between baseflow index and the drainage time scale for the catchments underscores the physical basis of the methodology and also enables the reduction of the procedure by one step, omitting recession flow analysis. The method serves as an important complementary tool for the assessment of the groundwater part of the Reserve and the Groundwater Resource Directed Measures in South Africa.

Abstract

The costs of acid mine drainage (AMD) monitoring result in the quest for alternative non-invasive method that can provide qualitative data on the progression of the pollution plume and ground geophysics was the ideal solution. However, the monitoring of AMD plume progression by ground geophysics (time-lapse electrical resistance) proves to be non-invasive but also time consuming. This gave way to a study that focuses on the modeling of different scenarios of the karstic aquifer. The models use the field parameters such as the electrical resistivity of the host rock and the target rock, depth to the target, noise level and electrode configuration in order to ensure that the model outcomes represent the field data as much as possible. This geoelectric modeling process uses Complex Resistivity Model (CRMod) and Complex Resistivity Tomography (CRTomo) to generate geoelectric subsurface images. Different resistivity values are applied to targets in order to assess the difference against the baseline model for each target scenario. The model resistivity difference is reduced to the smallest difference possible between the reference and new models in order to gauge the lowest percentage change in the model at which the background noises start to have impact on the results. The study shows that the behavior of targets (aquifer) could be clearly detected through resistivity difference tomography rather than inversion tomography. The electrode array plays a significant part in the detection of target areas and their differences in resistance because of its sensitivity. This therefore indicates that the electrode array should be chosen according to study requirements. Furthermore, the model geometry also plays a role and this can be seen with the modelling of different target sizes, alignments and shapes. Future studies that can provide a correlation between the field quantitative data from sampling and the model outcomes have the ability to add to the knowledge field of geophysical modelling therefore reducing costs associated with field based plume AMD monitoring300-500 words without references; reach your conclusions rather than only delivering promises.

Abstract

There is growing concern that South Africa's urban centres are becoming increasingly vulnerable to water scarcity due to stressed surface water resources, rapid urbanisation, climate change and increasing demand for water. Given South Africa's water scarcity, global trends for sustainable development, and awareness around the issues of environmental degradation and climate change, there is a need to consider alternative water management strategies. Water Sensitive Urban Design (WSUD) is an approach to sustainable urban water management that attempts to achieve the goal of a 'Water Sensitive City'. The concept of a Water Sensitive City seeks to ensure the sustainable management of water using a range of approaches such as the reuse of water (stormwater and wastewater), exploiting alternative available sources of supply, sustainable stormwater management and improving the resource value of urban water through aesthetic and recreational appeal. Therefore, WSUD attempts to assign a resource value to all forms of water in the urban context, viz. stormwater, wastewater, potable water and groundwater. However, groundwater is often the least considered because it is a hidden resource, often overlooked as a form a water supply (potable and non-potable) and it is often poorly protected. The management of urban groundwater and understanding the impacts of WSUD on groundwater in South African cities is challenging, due to complex geology, ambiguous groundwater regulations and management, data limitations, and lack of capacity. Thus, there is a need for an approach to assess the feasibility of management strategies such as WSUD, so that the potential opportunities and impacts can be quantified and used to inform the decision making process. An integrated modelling approach, incorporating both surface and subsurface hydrological processes, allows various urban water management strategies to be tested due to the complete representation of the hydrological cycle. This integration is important as WSUD is used to manage surface water, but WSUD known to utilise groundwater as a means of treatment and storage. This paper assesses the application, calibration and testing of the integrated model, MIKE SHE, and examines the complexities and value of establishing an integrated groundwater and surface water model for urban applications in South Africa. The paper serves to demonstrate the value of the application of MIKE SHE and integrated modelling for urban applications in a South African context and to test the models performance in Cape Town's unique conditions, accounting for a semi-arid climate, complex land use, variable topography and data limitations. Furthermore, this paper illustrates the value of integrated modelling as a management tool for assessing the implementation of WSUD strategies on the Cape Flats, helping identifying potential impacts of WSUD interventions on groundwater and the potential opportunities for groundwater to contribute towards ensuring to Cape Town's water security into the future.

Abstract

The University of the Free State investigated the possible dewatering of boreholes situated on the farm properties in the vicinity of an underground coal mine. The investigation consisted of three phases.
Phase one was a hydrocensus on the farm properties.
Phase two consisted of borehole yield determination by conducting pumping tests on the boreholes (where possible) identified in the hydrocensus phase.
Phase three included a visit to the underground mine workings, where water samples were collected at different groundwater inflow locations (especially water flowing in at the ventilation shaft). The monthly groundwater monitoring data of the underground coal mine was also incorporated for interpretation purposes. It appears that the water levels of the boreholes outside the mining boundaries are not affected. The water levels of the monthly monitored boreholes stabilized or even started recovering over the last few years. It also seems as though the larger streams in the area drains the groundwater as most of the deeper water level areas coincides with the presence of the streams. Most of the boreholes have typical borehole yields that is to be expected from Karoo formations i.e. between 0.5 and 1.5 L/s. An interesting observation is that a number of the boreholes with deep water levels are situated along dolerite contact zones at the western side of the mine. This may also be a geological structure resulting from the impact of a meteorite? From the available data it appears that the boreholes along this structure have the same chemical character as the water flowing down the ventilation shaft, strengthening the belief that the water from the shaft originates from this structure (or structures).

To determine the origin of the water flowing down the ventilation shaft, a detailed study of the structure to the west of the shaft is recommended. The farmers in the area should carefully monitor their water use in the boreholes, as over-abstraction can result in total failure of some of the boreholes.

Abstract

Arsenic is a common contaminant typically found in effluent from gold mine operations and copper smelters throughout the world. The geochemical behaviour of arsenic in contact with dolomite underlying an arsenic containing waste rock pile was investigated. The interaction between the arsenic and the dolomite is an important control in the subsequent transport of the arsenic in the dolomitic aquifer. Rocks with varying dolomite content were tested to investigate the interaction between the arsenic and dolomite. From the modelling and test results it was estimated that in the aquifer, between 60 - 90% of arsenic is present in the solid phase under oxidation conditions at >50 mV. At 50 to -25 mV about 40 - 60% of the arsenic is estimated to be present in the solid phase and below -25 mV about 0 - 10% of arsenic will be present in the solid phase. Although some arsenic is removed by the dolomite in the aquifer the arsenic would still be present above acceptable guidelines for drinking water. The arsenic in the solid phase will be in equilibrium with the aquifer water and could be remobilised 1) under more reducing conditions or 2) with a decrease in arsenic in the aquifer.

Abstract

Artificial Intelligence (AI) has been used in a variety of problems in the fields of science and engineering in particular automation of many processes due to their self-learning capabilities as well as their noise-immunity. In this paper, we describe a study of the applicability of one of the popular branch of AI (Artificial Neural Network (ANN)) as an alternative approach to automate modelling of one-dimensional geoelectrical resistivity sounding data. The methodology involves two ANNs; first one for curve type identification and the other one for model parameter estimation. A three-layer feedforward neural network that was trained from geoelectrical resistivity data taken at boreholes with geology logs was used to predict earth models from measured data without the need to guess the initial model parameters or use synthetic data as is done with most conventional inversion approaches. The motivation for using the ANN for geophysical inversion is that they are adaptive systems that perform a non-linear mapping between two sets of data from a given domain. For network training, we use the back-propagation algorithm. An example using data from southern Malawi shows that the ANN results outperforms the conventional approaches as the results after adequate training, produce reasonably accurate earth models which are in agreement with borehole log data.

Abstract

Groundwater is an essential source of water worldwide. The increased reliance on groundwater has caused the mining of many aquifers, a situation compounded by climate change, rising surface-air temperature, declining precipitation, and reduced groundwater recharge in many regions. The global annual intensity of groundwater use rose from 128 to 155 m3 per capita between 1950 (when the world population was 2.5 billion people) and 2021 (when the population was 7.9 billion people) and is herein projected to rise to 178 m3 per capita by 2050 as the world’s population is projected to increase (to 9.7 billion people by 2050) throughout the rest of the 21st century and beyond. This study projects a global annual groundwater depletion of 1,008 km3 by 2050, representing a 256% rise from the estimated 2010 depletion. This projection is most likely a lower bound of the actual groundwater depletion that would be realized considering environmental flows, historical trends of global economic growth, and climate-change impacts, thus being a harbinger of rising environmental degradation (e.g., land subsidence, seawater intrusion, streamflow reduction, aridification). Measures to achieve groundwater sustainability are herein identified.

Abstract

Groundwater levels in E33F quaternary catchment are at their lowest level ever. The impact of climatic variation and increasing abstraction were determined to be the main factor. There are 115 registered groundwater users in E33F and the monthly abstraction volumes are not being measured. There is a need to use land use activities as well as the population to estimate groundwater use. The main objective is to use non-groundwater monitoring data to estimate groundwater use in order to protect the aquifer and ecosystem in general in varying climatic condition. Land use activities information was used to estimate groundwater use in E33F quaternary catchment. The estimated groundwater use volumes were compared to allocated and measured volumes. For domestic groundwater use estimation, population data and an estimation 100 litre per person per day were used. The water requirements for the types of crops being cultivated together with the area (m2) were used to estimate groundwater use volumes for irrigation. The number and type of live stocks were used with the water requirements for each livestock type to estimate the groundwater use volumes. 96 % of groundwater users are using groundwater for irrigation purposes with 9 966 105 m3/a allocated for irrigation. Mining, industries, domestic and livestock are allocated 100 200 m3/a. The estimated groundwater use volume for irrigation is 30 960 000 m3/a, which is three times higher than the allocated volume. Groundwater use volume for domestic use is estimated to be 38 225 m3/a which is higher than the 31 000 m3/a allocated. The total estimated groundwater use volume in E33F is estimated to be 30 998 225 m3/a, which is three times higher than the allocated groundwater use volume of 10 066 305 m3/a. This estimation could be higher as only registered boreholes were used and estimations from mining, Industries and live stocks were excluded due to lack of data

Abstract

Globally, cumulative plastic production since 1950 is estimated to have reached 2500 Mt of plastic. It is estimated up 60% of this plastic is either resting in landfills or the natural environment, including groundwater settings. Microplastics are small pieces of plastic ranging between 1μm – 5mm in size and have been found in every ecosystem and environment on the planet. Much of the available literature on microplastics is focused on marine environments with few in comparison focused on freshwater environments, and even fewer on groundwater settings.

The aim of this study is therefore to investigate the attenuation process responsible for influencing microplastic transport in saturated sands. This research will adapt colloid transport theory and experiments to better understand the movement of microplastics through sandy media. Saturated aquifer conditions will be set up and simulated using modified Darcy column experiments adapted from Freeze & Cherry (1979). Modified microplastics will be injected into the columns as tracers and the effluent concentrations measured by Fourier-transform infrared spectroscopy (FTIR). Breakthrough curves will then be plotted using the effluent concentrations to determine the attachment efficiency (α). It is expected the attachment efficiency will vary by microplastic type and size range. The Ionic strength of the solution flowing through the column and the surface charges of both microplastics and sandy surfaces are likely to influence the degree of attenuation observed. The relationship between different types of microplastics and collector surfaces from a charge perspective and their influence on the degree of attenuation will be evaluated.

Given the lack of literature, its ubiquitous presence and postulated effects on human health, this research is significant. Through this research, the transport and attenuation of microplastics through sandy aquifers can be better understood, and in the process inform future research and water resource management.

Abstract

The uncertainties associated with both the sampling process and laboratory analysis can contribute to the variability of the results. In most cases, it does appear that if the water samples have been analysed by an accredited laboratory, the results are acceptable. While the accreditation of analytical laboratory and therefore its credibility is very important to uphold quality and integrity, the same should be said about the sampling process. The quality and credibility of a sampling process is typically left to the responsibility of the appointed groundwater practitioner without any criteria to evaluate the quality and integrity of the sampling process. Perhaps the quality and integrity of the sampling process is evaluated based on trust or experience of the practitioner. However without any form of scientific criteria to evaluate the quality and integrity of the sampling process, it is difficult for the sampling process to be scrutinized. The quality and integrity of both the sampling process and laboratory analysis must be scientifically evaluated based on the uncertainty of measurements in line with the monitoring goals/requirements. This presentation discusses the aspects of evaluation of measurement uncertainties associated with groundwater sampling as an important component of quality assessment of groundwater sampling processes. The potential implications of the uncertainties on the final results and their use in decision making is also discussed. The credibility of the decisions made also depends on the knowledge about the uncertainties of the final results

Abstract

The generation of acid mine drainage (AMD), as a result of mining activities, has led to the degradation of groundwater quality in many parts of the world. Coal mining, in particular, contributes to the production of AMD to a large extent in South Africa. Although a vast number of remediation methods exist to reduce the impacts of AMD on groundwater quality, the use of a coal fly ash monolith to act as a reactive and hydraulic barrier has not been extensively explored. This study, therefore, aims to investigate how different ways of packing ash affect the hydraulic conductivity of ash and influence leachate quality when acid-mine drainage filters through the ash. Coal ash is highly alkaline due to the existence of free lime on the surface of the ash particles. Previous studies that investigated alternative uses of coal ash, particularly in AMD treatment, suggest that coal ash has the potential to neutralise pH in acid water and remediate acidic soils. To test the effects of different packing methods of coal ash on the hydraulic conductivity and quality of acid mine leachate flowing through it, several Darcy column tests will be conducted. During the course of these experiments, the following parameters will be measured, electrical conductivity, pH discharge, lime (CaCO3) and selected elements of environmental concern.

Abstract

The advent of the 'Big Data' age has fast tracked advances in automated data analytics, with significant breakthroughs in the application of artificial intelligence (AI). Machine learning (ML), a branch of AI, brings together statistics and computer science, enabling computers to learn how to complete given tasks without the need for explicit programming. ML algorithms learn to recognize and describe complex patterns and relationships in data - making them useful tools for prediction and data-driven discovery. The fields of environmental sciences, water resources and geosciences have seen a proliferation of the use of AI and ML techniques. Yet, despite practical and commercial successes, ML remains a niche field with many under-explored research opportunities in the hydrogeological sciences. Currently physical-process based models are widely applied for groundwater research and management, being the dominant tool for describing and understanding processes governing groundwater flow and transport. However, they are limited in terms of the high data requirements, costly development and run time. By comparison, ML algorithms are data-driven models that establish relationships between an input (e.g. climate data) and an output (e.g. groundwater level) without the need to understand the underlying physical process, making them most suitable for cases in which data is plentiful but the underlying processes are poorly understood. Combining data-driven and process-based models can provide opportunities to compensate for the limitations of each of these methodologies. We present applications of ML algorithms as knowledge discovery tools and explore the potential and limitations of ML to fill in data gaps and forecast groundwater levels based on climate data and predictions. Results represent the first step in on-going work applying ML as an additional tool in the study and management of groundwater resources, alongside and enhancing conventional techniques such as numerical modelling.

Abstract

Unicef is the WASH sector lead globally and is, present at the country level, the main counterpart of government, especially regarding the component of the water balance utilised for potable safe water supplies. This mandate means that Unicef then has a role in looking at water resources nationally and not just as individual projects, and in doing so, contributes to good water governance as an integral part of system strengthening. Ensure this is done in partnership with other ministries and stakeholders that support them through advocacy for humanitarian and developmental access and support in technical areas such as groundwater assessments and monitoring. The focus on groundwater is especially linked with the fact that groundwater plays a major role due to its buffering capacity to climate variations, easier access and global coverage. Since groundwater is the most significant component of accessible freshwater resources, it is in the interest of UNICEF to make this resource more visible to meet both development and humanitarian goals, strengthen national systems and ultimately build resilience in mitigating water scarcity to scale or at the National level. Therefore, examples will be presented where Unicef has engaged on this journey with nations such as Afghanistan, Yemen, Mozambique and Rwanda to understand their water resources better. The overall objective at the National level is to adapt the capacity to withstand and recover as quickly as possible from external stresses and shocks or build resilience.

Abstract

Well-established engineered systems for depth-discrete monitoring in fractured rock boreholes (referred to as a Multilevel System or MLS) are commercially available and offer much diversity in design options, however, they are used infrequently in professional practice and have seen minimal use in groundwater research. MLSs provide information about hydraulic head and hydrochemistry from many different depths in a single borehole and, therefore, magnify greatly the knowledge value of each borehole. Conventional practice globally is devoted to standard monitoring wells, either alone as longer single screened wells or in clusters or nests with a few wells screened at different depth intervals. These are the mainstay of the groundwater science and engineering community and severely limit prospects for each borehole to provide the information needed to solve the complex problems typically posed by fractured rock. This paper outlines the nature and evolution of MLS technologies and points to recent literature showing how MLSs add important insights that cannot be obtained using conventional wells. Also, it reviews commercially available MLS technologies, which present a range of robust options with each system having different characteristics and niches depending on characterization and monitoring goals and site conditions. The paper also describes refined MLS criteria aimed at improving the cost effectiveness and expanding capabilities of MLSs, so as to improve their accessibility for high resolution data acquisition in the context of both groundwater system characterization and long-term monitoring.

Abstract

Precision agriculture continuously seeks improved methods to enhance productivity whether it is for greater crop yields or economic viability regarding labour inputs and satisfying the demand in a shorter time span. Soil moisture is one important factor that drives the agricultural industry and is therefore of utmost importance to manage it correctly. A shortage of water may result in reductions in yield, while excess irrigation water is a waste of water resources and can also have a negative impact on plant growth. Knowledge of the spatial distribution of soil moisture is important for determining soil moisture storage and soil hydraulic transport properties. Capturing field heterogeneity without exhaustive sampling and costly sample analysis is difficult. Electromagnetic induction, Frequency Domain Reflectometry, Neutron Scattering and conventional soil sampling have been utilised to determine the spatial variability of soil moisture within a field. Emphasis has been placed on practicality and accuracy of all the methods. Electromagnetics have proven itself to be the primary method to determine soil moisture within the field by comparing the results of the volumetric soil water content present in the field together with a combination of various soil properties such as clay and silt content, sand fraction, concretions, density and soil depth that contribute towards the accumulation of soil water. Electromagnetic induction has the highest resolution of data collected for a specific time period of all considered methods making it economically the best option for soil moisture management within a variable rate irrigation system. Electromagnetic induction has proven to be successful in delineating a field into management zones consisting of different classes based on observed conductivity values. Higher conductive zones are considered with small water demand. Lower conductive zones are considered with a greater water demand through a variable rate irrigation system. These water management zone maps could be informative for modelling, experimental design, sensor placement and targeted zone management strategies in soil science, hydrogeology, hydrology, and agricultural applications.

Abstract

Simple and cost-effective techniques are needed for land managers to assess and quantify the environmental impacts of hydrocarbon contamination. During the case study, hydrocarbon plume delineation was carried out using hydrogeological and geophysical techniques at a retail filling station located in Gauteng.

Laboratory and controlled spill experiments, using fresh hydrocarbon product, indicate that fresh hydrocarbons generally have a high electrical resistivity, whilst biodegraded hydrocarbons have a lower resistivity. This is attributed to the changes from electrically resistive to conductive behaviour with time due to biodegradation. As such, it should be possible to effectively delineate the subsurface hydrocarbon plume using two-dimensional (2D) Electrical Resistivity Tomography (ERT). As part of the case study, two traverses were conducted using an Electric Resistivity Tomography (ERT) survey with an ABEM SAS1000 Lund imaging system. The resultant 2D tomographs were interpreted based on the resistivity characteristics and subsurface material properties to delineate the plume. Localised resistivity highs were measured in both models and are representative of fresh hydrocarbons whereas areas of low resistivity represented areas of biodegraded hydrocarbons.

More conventional plume delineation techniques in the form of intrusive soil vapour and groundwater vapour surveys as well as hydrochemical anlayses of the on-site monitoring wells were used to compare the results and to construct the detailed Conceptual Site Model. During the investigation, four existing monitoring wells located on the site and additional two wells were installed downgradient of the Underground Storage Tanks (USTs) in order to determine the extent of the plume.

In conclusion, a comparison was found between the groundwater results and geophysical data obtained during the case study and it was concluded that ERT added a significant contribution to the Conceptual Site Model.

Abstract

Water monitoring is a key aspect in the mining industry, in terms of gathering baseline data during the pre-construction stage, identifying potential areas of concern and mitigating source pollutants during the operational stage. A proper water monitoring program assists in the monitoring of plume development and water level rebound during the closure phase. The data made available through consistent long term monitoring should not be underestimated. Monitoring the effect that coal mine operations have on the water quality and quantity of surface and groundwater resources is a complex and multidisciplinary task. Numerous methodologies exist for monitoring of this kind. This paper will supply an overview of the water- rock chemistry associated with coal mine environments and the key indicator elements that should be focused on for water monitoring as well a review of the Best Practice Guidelines requirements in terms of water monitoring. Two case studies of coal mines in KwaZulu Natal will be reviewed, the key challenges outlined and mitigation measures implemented. The impact of requirements such as those set out by the Department of Water and Sanitation in terms of strict water quality limits for water containment and waste facilities as specified by Water Use Licences has also created unrealistic non-compliance conditions. The initial approach to creating a water monitoring programme should involve first identifying gaps in previous datasets and delineating potential sources of contamination. The sampling frequency will depend primarily on the water resource being monitored and the water quality analysis will depend on the type of facility. The facilities required for a specific situation will depend on the type and amount of waste generated, potential for leachate formation, vulnerability of groundwater resources and potential for water usage or resource sensitivity.

Abstract

The so-called apparent increase of transmisivity (T) or hydraulic conductivity (K) with scale is an artifact and does not exist in the field. The reason for the apparent increasing of T with scale is due to the use of the "not applicable" random log Gaussian stochastic models that are used by geohydrologists. In the petroleum field, which uses deterministic methods, the apparent increase of T with aquifer volume does not occur. Groundwater practitioners have to change their view and use models that do not show this effect.

By using intuitive inspection of geological, fracture and connectivity data as well as real pumping test data, this paper shows that up-scaling must be performed with an exponential decaying function, where T always decreases with scale
.
Two types of heterogeneities exists namely a.) horizontal and b.) vertical. Connectivity between fractures is extremely important in both cases, but it is only in semi-confined and watertable aquifers that the vertical heterogeneities are really important (typical case of fracture dewatering)
{List only- not presented}

Abstract

Approximately 982 km3 /annum of the world’s groundwater reserve is abstracted, providing almost half of all drinking water worldwide. Globally, 70% is used for agricultural purposes while 38% for irrigation.

Most water resources of South Africa are threatened by contamination caused by industrial, agricultural, and commercial activities, and many parts of the country face ongoing drought with an urgent need to find alternative freshwater sources, such as groundwater. Groundwater constitutes approximately 15% of the total volume consumed, hence it is an important resource that supplements insufficient surface water supplies across South Africa.

Very little attention has been afforded to understanding the anthropogenically altered vadose zone as a potential source or buffer to groundwater contamination. This is evident from few research studies that has applied multiple isotopic tracers to characterise this zone. Most subsurface systems in South Africa are characterised by fractures, whereby flow and transport are concentrated along preferential flow paths.

This study aims to evaluate the performance of different tracer classes (environmental and artificial) with one another, and create a better understanding of the hydraulic properties, mean residence time and transport mechanisms of these tracers. The influence of unsaturated zone thickness on recharge mechanisms will also be evaluated.

Site visits will be conducted for the proposed study areas, and the neighbouring sources of contamination will be assessed. The matric potential and unsaturated hydraulic conductivities will be measured using various techniques. Water samples will be collected and analysed for the various tracers from the vadose zone using gravity lysimeters including suction cups. Several tracers will also be injected into boreholes where samples will be collected to calculate tracer residence times (BTC’s) and further constrain the hydraulic properties of the vadose zone. All samples will be analysed, interpreted, and simulated using the numerical finite-element modelling code SPRING, developed by delta h. The software derives quantitative results for groundwater flow and transport problems in the saturated and unsaturated zones of an aquifer.

The research is expected to provide more insight into the selection and use of environmental and artificial tracers as markers for detecting, understanding the transport processes and pathways of contaminants in typical altered South African subsurface environments. The impact derived improved characterisation of the pathways, transport, and migration processes of contaminants, leading to groundwater protection strategies and appropriate conceptual and numerical models. The output from this study will determine the vertical and horizontal flux for both saturated and unsaturated conditions.

Abstract

Stringent drinking water standards for constituents like chromium, arsenic, and nitrates, combined with continually higher demand for groundwater resources have led to the need for more efficient and accurate well characterization. Many boreholes are screened across multiple aquifers to maximize groundwater production, and since these aquifers can have different water qualities, the water produced at the wellhead is a blend of the various water qualities. Furthermore, the water entering a well may not be distributed equally across the screened intervals, but instead be highly variable based on the transmissivity of the aquifers, the depth of the pump intake, the pumping rate, and whether any perforations are sealed off due to physical, chemical, or biological plugging. By identifying zones of high and low flows and differing water qualities, well profiling is a proven technology that helps optimize operational groundwater production from water supply boreholes or remediation systems. This frequently results in increased efficiencies and reduced treatment costs. By accurately defining groundwater quantity and quality, dynamic profiling provides the data needed to optimize well designs. Conventional exploration methods frequently rely on selecting well screen intervals based on performing and analyzing drill stem tests for one zone at a time. Using dynamic flow and water quality profiling, the transmissivity and water quality can be determined for multiple production zones in a matter of one to two days. It also allows the location and size of the test intervals to be adjusted in the field, based on real-time measurements.

In this paper we discuss dynamic well profiling techniques with project case examples of characterization different types groundwater boreholes for a variety of applications and industries resulting in significant cost saving and sustainable water abstraction.

Abstract

Implementation of a mining project in South Africa involved dewatering of a fractured rock aquifer at considerable depth below ground level. Groundwater quality within this aquifer is not suitable for domestic use due to high levels of salinity. Numerous geological investigations in the area indicate that the target aquifer is confined, with a different piezometric head to the shallower aquifers. However, regulators and other interested and affected parties expressed concern regarding the potential mixing of more saline groundwater from the deeper aquifer to be dewatered with groundwater from shallower aquifers, which are extensively used for farming and domestic purposes.
A large database of groundwater quality monitoring data collected over 16 years was available to investigate the degree of mixing between the deeper more saline and shallower freshwater aquifers. The groundwater chemistry of selected boreholes with known geological profile, depth and construction was used to develop groundwater fingerprinting criteria for each of the aquifers in the area. These fingerprinting criteria were then applied to private and exploration boreholes in the area in order to identify the main aquifer from which groundwater was being sourced. Once the boreholes were classified in terms of groundwater origin, an attempt was made to identify indicators of mixing with deeper, more saline groundwater from the aquifer being dewatered.
Groundwater fingerprinting allowed identification of impacts related to the mining operations. The data showed that there was no upward mixing of water related to dewatering operations, but rather that surface spillages and disposal schemes may have resulted in minor changes in shallow groundwater quality. {List only- not presented}

Abstract

South Africa has committed to achieving the United Nations Sustainable Development Goals (SDG's) by 2030. But what does this mean and how does groundwater fit in to this? SDG 6 in particular focuses on ensuring universal access to safe and affordable drinking water for all by 2030. SDG 6 requires that the country protects and restores water-related ecosystems such as forests, mountains, wetlands, aquifers and rivers which are essential if we are to mitigate water scarcity. To accomplish this, South Africa has proceeded to align various plans, strategies, and policies to encompass the targets of the SDG's. This paper will focus on SDG sub-goal 6.3 which incorporates improvement of water quality and sub-goal 6.6 which involves protection and restoration of ecosystems. The methodology given by the UN for the groundwater in indicator 6.3.2 stipulates that countries are required to report on "proportion of water with good ambient water quality", in South Africa however we had to domesticate the indicator i.e. render it suitable for South African conditions so we changed the methodology to "proportion of water the conforms to the Water Quality Objectives (WQO's)" but there are virtually no WQO's developed for groundwater. Four core groundwater quality parameters (Electrical Conductivity, pH, Nitrate and Sulphate) are available through ZQM stations categorized through 65 hydrogeological (Vegter) regions. Groundwater water quality baseline is calculated as a reference period/range per hydrogeological region. For SDG 6.6, the indicator required for groundwater is "Quantity of groundwater within aquifers" The methodology received by the UN for "Quantity of groundwater within aquifers" required a baseline (average reference period of five years) in meters per hydrogeological region. This indicator is again domesticated for South Africa and based on the 40-60 percentiles of groundwater levels per hydrogeological region. There are a number of future indicators that can be included for aquifers under SDG 6.6, but the groundwater sector needs to come together and decide what is important to report on. These SDG targets reporting has given the Water and Sanitation sector a new look at data. It has forced us to critically think of concepts such as baseline and performance monitoring. We now know where our data gaps and targets are, and we have to provide an action plan to address these.

Abstract

Gold Mining activities the past 60 years at AngloGold Ashanti polluted the groundwater underlain by 4000 ha of land at the Vaal River and West Wits operations in South Africa. Sulphide material in Tailings Storage Facilities, Waste Rock Dumps and extraction plants produce Saline Mine Drainage with Sulphate, minor salts and metals that seep to the groundwater and ultimately into surface water resources. Water regulation requires mines to prevent, minimise/ reduce or eliminate pollution of water resources. The waste philosophy has matured from tolerate and transfer to treat and termination of pollution sources. The impact of the pollution was determined and possible technologies to treat the impact were evaluated. Source controls of proper water management by storm water management, clean dirty water separation, lined water conveyance structures and reduced deposition of water on waste facilities is crucial. The aquifer character determines the possible remediation technology. From the possible technologies phytoremediation, physical interception and re-use of this water was selected. In future possible treatment of the water would be considered. This paper explain the strategy and report on the phased implementation of these plans and the expected results. The establishment of 750 ha of woodlands as phytoremediation, interception trenches of 1250 m, 38 interception boreholes and infrastructure to re-use this water in 10 water management areas is planned. The total volume of 15 Ml/day would be abstracted for re-use from the boreholes and trenches. The woodlands can potentially attenuate and treat 5 ml/day. The established woodlands of 150 ha proof successful to intercept diffused seep over the area of establishment and reduce the water level and base flow. The 2 implemented trenches of 1000 m indicate a local decline in the water level with interception of shallow groundwater within 1-2 m from surface. The 2 production interception well fields abstracting 50 and 30 l/s respectively indicate a water level decline of between 2 to 14 m with regional cones of depression of a few hundred meters to intercept groundwater flow up to 20 meter. Predictions from groundwater modelling indicate that these schemes can minimise pollution during the operational phase and protect downstream water resources. Predictions from modelling indicate that the pollution sources need to be removed to ensure long term clean-up to return the land to save use. The gold and uranium prize is securing the removal of the sources through re-processing of the tailings and waste rock dumps. After removal of the sources of pollution the remediation schemes would have to be operated for 20 years to return the groundwater to an acceptable standard of stock watering and industrial water use. The water quality is observed by a monitoring network of approximately 100 observation boreholes.

Abstract

This paper presents data obtained from sampling and geochemical analysis of gold tailings and associated pool and drain water. Inverse geochemical modelling using PHREEQC indicated geochemical processes operating in the tailings between the pool and drains. These included pyrite oxidation and dissolution of various minerals identified in the tailings. The processes were incorporated into an ensemble geochemical model to calculate post-closure sulphate concentration in tailings seepage.

The ensemble model included a spreadsheet model to calculate oxygen flux at various depths in the tailings column, and a one-dimensional transport model in PHREEQC. The calculated oxygen flux was applied to each cell in the tailings column to determine the amount of sulphide oxidation and the release of acidity into the tailings pore water. The rate of vertical transport of pore water in the column was determined from physical characterisation of the tailings particle size and saturated hydraulic conductivity.

The model results indicate elevated sulphate concentrations and acidity moving as a front from the top of the column downwards. The modelled sulphate concentration of 1 500 to 1 900 mg/L 8-16 years after closure compared well with measurements of drain water quality at a tailings dam decommissioned approximately 16 years ago. This provided some credibility to the modelled result of 2 300 mg/L sulphate up to 50 years post-closure. However, the tailings moisture content, infiltration rate, and pyrite oxidation rate in the model were based on literature values, rather than site-specific measurements.

Abstract

The Gravity Recovery and Climate Experiment (GRACE) satellites detect minute temporal variation in the earth's gravitational field at an extraordinary accuracy, in order to make estimation of the total water storage (TWS). GRACE provides a unique opportunity to study and monitor real time water variation in the hydrologic stores (snow, groundwater, surface water and soil moisture) due to increases or decreases in storage. The GRACE monthly TWS data are being used to estimate changes in groundwater storage in the Vaal River Basin for a period (2002 to 2014). The Vaal River Basin has been selected, because it is one of the most water stressed catchments in South Africa; it is well-renowned for its high concentration of industrial activities and urbanized zones. Therefore, in order to meet future water demands, it is critical to monitor and calculate changes in groundwater storages as an important aspect of water management, where such a resource is a key to economic development and social development. Previous studies in the Vaal River Basin were mostly localised focusing mainly on groundwater quality and to a lesser extent groundwater assessment. Hydrological models have been generated for the whole of South Africa, but many of these models do not take into account the groundwater component. Thus, there is a significant gap in the understanding of surface and ground water dynamics in the Vaal River Basin. The paucity of data and monitoring networks are often the limitation in calculating changes in water storage over a large area, particularly in Africa. In this scenario GRACE is a good approach to estimate changes in hydrological storages as it covers large areas and generates real time data. It does not require information on soil moisture, which is often difficult to measure. The accuracy of calculating change in groundwater storage lies in the processing of GRACE data and smoothing radii. For this study, smoothing radii of 1500, 900, 500, 300, 150 and 1 km are used. Currently the associated error with different smoothing radii is unknown. The preliminary results indicate that the study area experienced a loss in TWS of -31.58 mm equivalent water height over a period of 144 months in TWS at 300 km smoothing radius. The change in groundwater storage is calculated by incorporating hydrologic components to the TWS (work in progress). The results obtained from this study will be compared to existing hydrological models and results generated from models applicable to the semi-arid region of South Africa. It is anticipated that this satellite observation technique, GRACE, will provide an accurate estimate of change in groundwater storage. Furthermore, it will show the usefulness of satellite based techniques for improving our understanding of groundwater dynamics, which will improve water management practices.

Abstract

Artesian boreholes are a common feature worldwide in confined aquifers, but the hydraulic testing of these boreholes, and estimation of aquifer properties from such tests, still poses a challenge for hydrogeologists. Common hydraulic tests, such as step-drawdown or constant-discharge rate tests require a static water level at the start of the test, and the measurement of drawdown (increasing over time) and abstraction rate (fixed for a period of time). Usually, when undertaking a pumping test in an artesian borehole, the drawdown is measured from ground level, and the drop in hydraulic head between static pressure and ground level is often ignored. This procedure also implies that the starting time of the test is not at the static water level. A constant-head test, set at ground level, is the other option. However, the decrease in flow rate is not only dependent on the hydraulic properties of the aquifer, but also masked by pipe hydraulic effects within the well. This kind of test would also limit the available drawdown to be utilized for the test. Hence, it was required to develop a method for undertaking hydraulic tests in strong artesian boreholes, allowing for the drawdown to fluctuate between levels both above and below ground and avoiding the pitfalls described above. The solution is a specially designed and constructed wellhead for the installation of the pump and monitoring equipment prior to the hydraulic test. The standard tests are slightly modified and are only undertaken after sealing the wellhead and reaching static hydraulic pressure. The recommended wellhead construction and subsequent hydraulic tests were implemented at a strong artesian borehole in the Blossoms Wellfield, south of Oudtshoorn in the Western Cape province of South Africa.

Abstract

Identifying and characterising the vertical and horizontal extent of chlorinated volatile organic compound (CVOC) plumes can be a complex undertaking and subject to a high degree of uncertainty as dense non-aqueous phase liquid (DNAPL) movement in the subsurface is governed most notably by geologic heterogeneities. These heterogeneities influence hydraulic conductivity allowing for preferential flow in areas of higher conductivity and potential pooling or accumulation in areas of lower conductivity. This coupled with the density-induced sinking behaviour of DNAPL itself and the effects of groundwater recharge in the aquifer result in significant challenges in assessing the distribution and extent of CVOC plumes in the subsurface. It has been recognized that high resolution site characterization (HRSC) can provide the necessary level of information to allow for appropriate solutions to be implemented to mitigate the effects of subsurface contamination. Although the initial cost of HRSC is higher, the long-term costs can be substantially reduced and the remedial benefits far greater by obtaining a better understanding of the plume characteristics upfront. The authors will discuss a case study site in South Africa, where ERM has conducted HRSC of a CVOC plume to characterise the distribution of the source area and plume architecture in order to assess the potential risk to receptors on and off-site. The source of impact resulted from the use of a tetrachloroethene (PCE)-based solvent in an on-site workshop. The following methods of characterization were employed:
- Conducting a passive soil gas survey to identify and characterise potential source zones and groundwater impacts;
- Vertical characterisation of the hydrostratigraphy, contaminant distribution and speciation in real time using a Waterloo Advanced Profiling System (APS) with a mobile on-site laboratory;
- Using the Waterloo APS data to design and install groundwater monitoring wells to delineate the vertical and lateral extent of contamination; and
- Conducting a vapour intrusion investigation including sub-slab soil gas, indoor and outdoor air sampling to estimate current risk to on-site employees.
In less than a year, the risk at the site is now largely understood and the strategies for mitigating the effects of the contamination can be targeted and optimised based on the information gained during the HRSC assessment.

Abstract

The most used methods for the capturing of shallow groundwater contamination are the use of abstraction wells and infiltration trenches. The use of trenches for the interception of shallow groundwater contamination has become a popular choice of remediation method due to the lower cost than a comparable pump-and-treat system. Trenches have large surface areas which limits the tendency of filter media clogging with suspended media as well as only a single pump and lower maintenance requirements. An important consideration of the use of trenches is determining the effectivity before design and construction. To date, limited information on the effectivity of trench designs are available, therefore a method to determine the effectivity of a trench was devised. This paper will discuss this evaluation method and look at some cases where planned trenches were successful and some cases where they were not.

Abstract

The hydrogeological setting of a proposed mine site can significantly influence the viability of the mining venture. The management of groundwater inflows, costs of the dewatering technology, construction and maintenance of storage facilities, discharge strategies and anticipated environmental impacts are vital factors for consideration. It is fundamental to assess the hydrogeological setting at an early stage of the mine life cycle and should involve the collection of sufficient hydrogeological data, conceptualisation of the hydrogeological setting and an assessment of planned mine operations and anticipated impacts. Ambient hydrogeological conditions at the deposit area may be identified by conducting a hydrocensus and utilising existing ore exploration drilling data. Information from the hydrocensus and ore exploration drilling can provide valuable preliminary data on groundwater risks, dewatering and available groundwater resources. Potential groundwater/surface water interactions and receptors sensitive to environmental impacts can be identified during a hydrocensus. Similarly, water strikes and fracture density recorded during exploration drilling provide valuable insight to the subterranean environment. It is also possible to obtain aquifer hydraulic properties through packer testing of exploration boreholes. Geochemical test work on exploration borehole-cores could provide valuable information regarding contamination risks from ore deposit and waste material storage. The installation of piezometers within available and accessible exploration holes that extend below the regional groundwater level can pioneer the collection of monitoring data crucial for consideration during the mine life cycle and provide an understanding of the interaction between hydrogeological units and recharge characteristics. Ultimately, mine operations and associated potential impacts on the surrounding groundwater environment can be simulated with the application of numerical hydrogeological flow and contaminant transport models. The numerical models can simulate the regional groundwater flow system and complexities of the mine environment, the accuracy of which is influenced by the type, spatial and temporal distribution of the data collected. It is accordingly suggested that the collection of hydrogeological data and information during the exploration phase would facilitate the timely conceptualisation of potential groundwater risks and effective planning of hydrogeological investigations required during upcoming phases while assisting in the budget optimisation of these future studies.

Abstract

The groundwater quality of the Orange Water Management Area (OWMA) was assessed to determine the current groundwater status. Groundwater is of major importance in the Orange Basin and constitutes the only source of water over large areas. Groundwater in the OWMA is mainly used for domestic supply, stock watering, irrigation, and mining activities. Increase in mining and agricultural activities place a demand for the assessment of groundwater quality. The groundwater quality was assessed by collecting groundwater samples from farm boreholes, household boreholes, and mine boreholes. Physical parameters such as pH, temperature and Electrical Conductivity (EC) were measured in-situ using an Aquameter instrument. The groundwater chemistry of samples were analysed using Inductively Coupled Plasma Mass Spectrometry, Ion Chromatography, and Spectrophotometer for cations, anions and alkalinity respectively. The analyses were done at Council for Geoscience laboratory. The results obtained indicated high concentration of Nitrate (NO3), EC, sulphate (SO4), Iron (Fe), and dissolved metals (Chromium, Nickel, Copper, Zinc, and Lead). The concentrations were higher than the South African National Standards (SANS) 241 (2006) drinking water required guideline. The OWMA is characterised by the rocks of the Karoo Supergroup, Ventersdorp Supergroup, Transvaal Supergroup, Namaqua and Natal Metamorphic Province, Gariep Supergroup, and Kalahari Group. Groundwater is found in the sandstones of the Beaufort Group. Salt Mining occurs in the Namaqua Group, hence the high concentration of EC observed. High EC was also found in the Dwyka Group. The salt obtained from the pans underlain by the Dwyka Group rocks has relatively high sodium sulphate content, this probably results from oxidation of iron sulphate to sulphate. Therefore, high concentration of SO4 is due to the geology of the area. High concentration of NO3 is due to agricultural activities, whereas high concentration of EC, Fe, SO4 and dissolved metals is due to mining activities.

Abstract

Resources required for groundwater sampling includes but not limited to pumping equipment, trained manpower and technical resources specific to the sampling function. Bearing these expenses in mind, choosing a laboratory for testing the water samples collected should be a carefully considered purchase. Choosing a testing facility that cannot deliver an efficient, reliable and technically sound service could render the sampling futile.

Water samples submitted to a laboratory for testing are received from third party sources more than ninety percent (90%) of the time and sampling techniques and sample integrity cannot be verified by the laboratory. However, the validity, reliability and integrity of the laboratory testing are within the control of the testing facility. These aspects of a laboratory are usually controlled within a quality management system where established policies and procedures form the basis of such a system. This system maintains a foundation for technical competence and customer service at the laboratory.

There are numerous testing facilities available to Consultants requiring chemical and microbiological groundwater testing, each with varying levels of integrity and technical ability. It is imperative to maintain confidence in the validity of results of analyses from a laboratory and this assurance can be understood through an examination of a facility's management system.

An established quality management system would comprise a policy statement, associated technical methods and technical and administrative procedures. This system would be formally documented and audited as part of the on-going laboratory's management system. In some instances, laboratories formalise this into an accreditation of the laboratory to an international standard, such as ISO 17025:2005.

The assurance that the results of analyses from any laboratory are of sound technical integrity would depend on factors such as
- personnel training,
- accommodation and environmental conditions under which the tests are carried out,
- validation of the methodology applied (including the uncertainty of measurement),
- the calibration and maintenance of the equipment used,
- understanding the traceability of and measurement undertaken,
- handling and preservation of the sample on receipt and while in the laboratory.

Each of these factors plays a critical role in the integrity of results of analyses and should be interrogated when trying to understand the reliability and competence of the laboratory of choice.{List only- not presented}

Abstract

Water scarcity is a growing issue in South Africa. The consumption of water is rising and as such, water is becoming a scarce and valuable resource. Given the circumstances that South Africa is facing, improving the use of ground water could help tackle water scarcity in South Africa. Groundwater has been an important source of water and it can bring socio-economic benefits if properly used. Studies have proved that groundwater resources play a fundamental role in the security and sustainability of livelihoods and regional economies throughout the world. However, in South Africa, groundwater still remains a poorly managed resource and this hinders socio-economic development. This paper examines the current state of ground water management in South Africa. The paper also examines how ground water in South Africa is currently allocated and used, and explores some of the consequences of current water management arrangements. {List only- not presented}

Abstract

In South Africa, the use of stochastic inputs in surface water resources assessments has become the norm while this is rarely done for groundwater resources. Studies that have applied multi-site and multi-variate methods that incorporate stochastic generation of groundwater levels are limited. Stochastic based inputs account for uncertainties attributed to inherent temporal and spatial variability of hydrologic variables and climatic conditions. This study applied variable length block (VLB) stochastic generator for simultaneous generation of multi-site stochastic time series of rainfall, evaporation and groundwater levels. In the study, 100 stochastic sequences with record length of 34 years (1980-2013), similar to the historic one were generated. Performance of VLB was assessed by comparing single statistics of historic time series located within box plots of the 100 annual and monthly stochastically generated time series. The statistics used include mean, median, 25th and 75th percentiles, lowest and highest values, standard deviation, skewness, and serial and cross correlation coefficients. Majority (9 out of 10) of the historical statistics were mostly well preserved by VLB, except for skewness. Historic highest groundwater levels were mostly underestimated. Historic statistics below interquartile range (overestimation) is a common problem of weather generators which can be reduced by including additional covariates that influence atmospheric circulation. The generation of multi-site stochastic sequences support realistic assessment of groundwater resources and generation of groundwater operating rules.

Abstract

Estimating groundwater recharge response from rainfall remains a major challenge especially in arid and semi-arid areas where recharge is difficult to quantify because of uncertainties of hydraulic parameters and lack of historical data. In this study, Chloride Mass Balance (CMB) method and Extended model for Aquifer Recharge and soil moisture Transport through unsaturated Hardrock (EARTH) model were used to estimate groundwater recharge rates. Groundwater chemistry data was acquired from the Department of Water and Sanitation (DWS) and Global Project Management consultants, while groundwater samples were collected to fill-in the identified gaps. These were sent to Council for Geoscience laboratory for geochemical analysis. Rainfall samples were also collected and sent for geochemical analysis. An average value of rainfall chloride concentration, average groundwater chloride concentration and mean annual precipitation (MAP) were used to estimate recharge rate at a regional scale. Local scale recharge was also calculated using chloride concentration at each borehole. The results were integrated in ArcGIS software to develop a recharge distribution map of the entire area. For EARTH model, long term rainfall and groundwater levels data were acquired from the South Africa Weather Services and DWS, respectively. Soil samples were collected at selected sites and analysed. These were used to determine representative values of specific yield to use on EARTH model. 60% of the groundwater levels data for 5 boreholes was used for model calibration while the remaining 40% was used for model validation. The model performance was evaluated using coefficient of determination (R2), correlation coefficient (R), Root Mean Square Error (RMSE) and Mean square error (MSE). Regional recharge rates of 12.1 mm/a (equivalent to 1.84% of 656 mm/a MAP) and 30.1 mm/a (equivalent to 4.6% MAP) were calculated using rainfall chloride concentrations of 0.36 and 0.9 mg/L, respectively. The estimated local recharge rates ranged from 0.9-30.2 mm/a (0.14 - 4.6%) and 2 - 75 mm/a (0.3 - 11.4%) using chloride concentration of 0.9 and 0.36 mg/L, respectively. The average recharge rate estimated using EARTH model is 6.12% of the MAP (40.1 mm/a). CMB results were found to fall within the same range with those obtained in other studies within the vicinity of the study area. The results of EARTH model and CMB method were comparable. The computed R2, R, RMSE and MSE ranged from 0.47-0.87, 0.68-0.94, 0.04-0.34, 0.16-3.16, and 0.50-0.79, 0.68-0.89, 0.07-0.68, 0.15-8.78 for calibration and validation, respectively. This showed reasonable and acceptable model performance. The study found that there is poor response of groundwater levels during rainy season which is likely to be due to lack of preferential flows between surface water and groundwater systems. This has resulted in poor relationship between estimated and observed groundwater levels during rainfall season.

Key words: ArcGIS, CMB, EARTH, Groundwater recharge, rainfall

Abstract

Model calibration and scenario evaluations of 2D and 3D groundwater simulations are often computationally expensive due to dense meshes and the high number of iterations required before finding acceptable results. Furthermore, due to the diversity of modelling scenarios, a standardised presentation of modelling results to a general audience is complicated by different levels of technical expertise.

Reducing computational time
In this presentation we look briefly at the use of Reduced Order Models (ROM's), which is one of the recent developments in groundwater modelling. The method allows significant speed-up times in model calibration and scenario evaluation studies. In saturated flow for example, these approaches show speed-up times of >1000 when compared to full models created with Finite Element of Finite Difference methods. These methods are demonstrated to a case study in the Table Mountain Group, in which we show a simplified parameter calibration and scenario evaluation study.

Standardising presentation
In order to present the results to as wide an audience as possible, the use of a web-browser as a GUI is proposed, where the web-page is coupled to a geo-spatial database and data is presented in a spatial and numeric format. The use of the spatial database manager PostgreSQL with PostGIS is proposed. Through a browser interface, users can run modelling scenarios using the ROM, which is evaluated in near real-time. Following the evaluation of the model, we show how PostGIS can spatially present data on a base-map such as google maps. In keeping with the current trends in online map customisation, viewers can interactively choose to overlay the base-map with a data-type (such as pressure or hydraulic head contours or flow direction) that is most intuitive for their level of familiarity with the data.

Conclusion
In using advanced modelling techniques and a simplified browser based presentation of results, high-level decisions in water resource management can be significantly accelerated with the use of interactive scenario evaluations. Furthermore, by reaching a broader audience, public participation will be significantly enhanced.