Stream-flow and storm-flow in four small catchments were analysed by the paired catchment method for a response to fire. Two of the catchments were vegetated with over-mature fynbos (the indigenous scrub vegetation of the south-western Cape Province, South Africa), one was afforested to Pinus radiata and the fourth to Eucalyptus fastigata. One of the fynbos catchments was burned in a prescribed fire in the late dry season. The other catchments burned in wildfires. Neither of the fynbos catchments showed a change in storm-flows. Annual total flow increases of around 16% were in line with predictions, being related to the reductions in transpiration and interception. The manner of stream-flow and storm-flow generation appeared to have remained unaltered despite the fire. The two timber plantation catchments experienced large and significant increases in stormflow and sediment yields, while total flow increased by 12% in the pine catchment and decreased marginally in the eucalypt catchment. After fire, storm hydrographs were higher and steeper though their duration was little changed. These fire effects are considered to be due to changes in storm-flow generation consistent with an increased delivery of overland flow to the stream channel. This was caused, in part, by reduced infiltration resulting from water repellency in the soils of the burned catchments. The inherent wettability of a wide range of soil types and textures from beneath timber plantations and other vegetation types over a broad geographic distribution in South Africa was measured by four methods. Soils with high repellency ratings, unrelated to fire, are common and are most likely to occur beneath plantations of Eucalyptus and Acacia spp. and indigenous forest. Water repellent soils played a role, at two of the three locations, in the generation of overland flow from small plots exposed to simulated rainfall. However, the inherent repellency of the dry soils was extreme, such that fire-induced water repellency was not a factor in the response of the plots. The important role of fire in this experiment was in burning-off of repellency in the surface layer of the soil and in removing ground cover. / Thesis (Ph.D.)-University of Natal, Pietermaritzburg, 1994.

Changes in climate due to anthropogenic influences are expected to affect both hydrological and agricultural systems in southern Africa. Studies on the potential impacts of climate change on agrohydrological systems had been performed previously in the School of Bioresources Engineering and Environmental Hydrology (School of BEEH). However, refinement of these modelling tools and restructuring of the databases used was needed to enable more realistic and dynamic simulations of the impacts of changes in climate. Furthermore, it was realised that modifications and linkages of various routines would result in a faster processing time to perform climate impact assessments at the catchment scale. Baseline ("present") climatic information for this study was obtained from the School of BEEH's database. Scenarios of future climate were obtained from six General Circulation Models (GCMs). Output from the five GCMs which provided monthly climate output was used in the climate impact assessments carried out. Potential changes in variability of rainfall resulting from climate change was assessed using the daily climate output from the sixth GCM. As the spatial resolution of the climatic output from these GCMs was too coarse for use in climate impact studies the GCM output was interpolated to a finer spatial resolution. To assess the potential impact of climate change on water resources in southern Africa the ACRU hydrological modelling system was selected. The ACRU model was, however, initially modified and updated to enable more dynamic simulation of climate change. In previous hydrological studies of climate change in southern Africa Quaternary Catchments were modelled as individual, isolated catchments. To determine the potential impact of changes in climate on accumulated flows in large catchments the configuration of the Quaternary Catchments needed to be determined and this configuration used in ACRU. The changes in hydrological responses were calculated both as absolute differences between future and present values and the ratio offuture hydrological response to the present response. The large degree of uncertainty between the GCMs was reflected in the wide range of results obtained for the water resources component of this study. In addition to the climate impact studies, sensitivity and threshold studies were performed using ACRU to assess the vulnerability of regions to changes in climate. Potential change in the yields and distributions of parameters important to agriculture, such as heat units, crops, pastures and commercial tree species were assessed using simple crop models at a quarter ofdegree latitude / longitude scale. Most species were simulated to show decreases in yields and climatically suitable areas. There are many sources of uncertainties when performing climate impact assessments and the origins of these uncertainties were investigated. Lastly, potential adaptation strategies for southern Africa considering the results obtained are presented. / Thesis (Ph.D.)-University of Natal, Pietermaritzburg, 2001.

South African water management areas could find themselves without enough water for its users due to new methods of performing water allocation as stipulated in the National Water Act of 1998. A water accounting system would address the need for accurate metering, monitoring and auditing of South Africa’s water resources to ensure that users are complying with their allocations. Such a system should be able to provide information such as comparisons between the simulated and observed flow of water at a point, comparisons between the amount of water allocated to a user and the actual water used by that user, and the source and destination of water at a point. This document contains a literature review, an explanation of the methods used to develop a prototype water accounting system and a discussion of the results from testing the system. A literature review was undertaken which covered topics in water resources planning, water resources operations, local legislation for water allocation and new technologies which could be applied to aid the management of water resources in South Africa. The results from the literature review indicated real time water accounting systems can give effect to water allocation rules. The water accounting system is comprised of two simulation models and a database. The models used for the study were the ACRU2000 model and the MIKE BASIN model. These models require data as well as a means to automate the transfer of data between the models and thus a database was developed. The database was developed in Microsoft Access and, in addition to the construction of a number of tables required to house the data, a database dashboard was made to control the functions of the database. An assessment of the ACRU2000 and MIKE BASIN models was performed in order to determine if they are suitable for use as water accounting tools. ACRU2000 was used for its process based, daily rainfall-runoff modelling capabilities. Due to the process based modelling capabilities of ACRU2000, forecasts of rainfall can be used as input to the simulations. Hot starting is the storing of internal model state variables at a particular time and the use of these variables in a different simulation to start the model up again. It was expected that, due to long simulation run times for ACRU2000, it would be beneficial to enable ACRU2000 to be hot started and an attempt to hot start ACRU2000 is presented. This would have allowed for significantly decreased simulation run times as the model can be warmed up for two years and thereafter hot started to run only for one day at a time. An assessment of the MIKE BASIN network allocation model to be used as a water accounting system was performed by attempting to meet the project objectives through building a fictional water supply network. The network is composed of a small catchment containing six runoff generating regions, a reservoir and ten water users. Three network allocation scenarios were constructed in order to fully test the rule sets and allocation capabilities currently available in the MIKE BASIN model. The study has shown that the tools and models used are capable of forming a rudimentary water accounting system. This is encouraging as it shows that there is the potential to improve the water resources management in South Africa using tools that already exist. / Thesis (M.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2010.

The early efforts of Booth, Johnson, Rubidge, Catuneanu, de Wit, Chevallier, Stankiewicz, Weckmann and many other scientists in studying the Karoo Supergroup has led to comprehensive documentation of the geology on the main Karoo Basin with regards to understanding the age, sedimentology, sedimentary facies and depositional environments. In spite of these studies, the subsurface structure, variations in thickness of various formations in large parts of the basin, the location and orientation of subsurface dolerite intrusions, and the depth to magnetic and gravity sources remains poorly documented. A geological study with the aid of geophysical techniques, magnetic and gravity, was conducted in the southwestern part of the main Karoo Basin. The objectives of the study were to construct numerous models of the main basin that image the crust to a depth of 45 km, to determine thicknesses of various formations, to relate observed geophysical anomalies with geological bodies and lineaments, to estimate the depth of existing anomalous bodies, to determine densities and porosity of various formations, as well as to determine the mineralogy of various rocks in the Karoo Basin.

This thesis develops a technique for the identification, classification and quantification of instream flow environments. These features have been traditionally referred to as 'habitats' by lotic ecologists, in this research they are termed 'hydraulic biotopes'. The hydraulic biotope is the lowest of six nested levels of a hierarchical geomorphological model. This model has been developed as a tool to assist river managers, researchers and conservationists to categorise or classify rivers with respect to their geomorphic characteristics. Each level of the model provides data at a different level of resolution. This ranges from the broad scale catchment data to the site specific 'habitat' or hydraulic biotope data. Although this thesis is primarily concerned with the development of the hydraulic biotope, the interaction of all catchment variables needs to be recognised. Detailed analysis of hydraulic biotope data in the Buffalo River are presented within the broader hierarchical model. Consultation with lotic ecologists, together with a review of ecological literature, emphasised the need for a standardised terminology for the classification of ecologically significant instream flow environments. At present a fairly haphazard 'habitat' classification tends to be carried out by most researchers, this often leads to confusion in the identification and naming of different hydraulic biotopes ('habitats'). This confusion is exaggerated by the sharing of terminology between lotic ecology and fluvial geomorphology, usually for the categorisation of different types of features. A review of the ecological literature emphasises the importance of flow hydraulics within a river to describe the distribution of biota. The hydraulic variables considered to be most significant include velocity and depth. As river morphology directly determines the prevailing distribution of depth, velocity and substratum, it is obvious that there are important links to be made between fluvial geomorphology and lotic ecology. This thesis explores the potential of the hydraulic biotope as a tool to help develop those links. This thesis presents a standardised classification matrix for the identification of hydraulic biotopes. The matrix is simply based on water surface characteristics together with channel bed substratum. The validity of this matrix is tested by statistical analysis of hydraulic variables quantifying flow conditions within the various hydraulic biotope classes. Data is presented from four different river systems, each representing a different sedimentological environment. Where possible the influence of discharge has been considered. Results from more than 3000 data points show that hydraulic biotopes have distinct hydraulic characteristics in terms of velocity-depth ratio, Froude number, Reynolds number, 'roughness' Reynolds number and shear velocity. These hydraulic indices represent flow conditions both as an average within the water column, and near the bed. Statistical analysis shows that the hydraulic characteristics of the various hydraulic biotope classes are relatively consistent both within different fluvial environments and at different stages of flow. Unlike the morphological unit in which the hydraulic biotope is nested, in stream flow environments are shown to be temporally dynamic. Using the classification matrix as a tool for identification, hydraulic biotopes identified at one discharge are shown to be transformed from one class to another as a response to change in stage. The pattern of transformation is shown to be consistent within different sedimentological environments. An examination of the associations between hydraulic biotopes and morphological units demonstrates that, although some hydraulic biotopes are common to all morphological units (backwater pools, pools and runs), some features have specific associations. In this study rapids were found to be prevalent in bedrock pavement, bedrock pool and plane bed morphology, while cascades, chutes and riffles were common to plane bed, step and riffle morphology. Results from this research indicate that the hydraulic biotope, within the hierarchical geomorphological model, has the potential to aid the prediction of channel adjustment and associated 'habitat' (hydraulic biotope) transformation in response to changes in flow and sediment yield. These are likely to become increasingly important issues as South Africa strives to maintain a balance between the development of water resources to meet the needs of the rapidly expanding population, whilst at the same time maintaining the fluvial environment for sustainable use.

The supply of water to 'disadvantaged' areas of KwaZulu-Natal has in the past received low priority. Local government is now faced with supplying water to large, sometimes dispersed, rural populations. Groundwater has been utilised informally as a water supply for some years, but the impetus provided by the Department of Water Affairs and Forestry's White Paper has compelled those responsible for water supply to seriously consider groundwater as a sustainable option. The development potential or success of groundwater in its role as a reliable water supply is dependent on acceptance of the resource by the communities, appropriate level of service, yield sustainability and safe quality. Apart from the social aspects, the yield and groundwater quality characteristics of the aquifers under consideration must be used to determine the best hydrogeological features to target during exploration. A total of 993 borehole records, from a recent government drought relief programme, were used to compare the yield, water quality and best geophysical exploration and drilling techniques of the main hydrolithologies in rural KwaZulu-Natal. The results of comparative analysis shows that the competent rocks of the Natal Group and Natal Metamorphic Province and the karstic Uloa Formation of the Maputaland Group have the best overall potential for water supply. The unconsolidated sediments of the Maputaland Group also have good potential, but have some salinity problems. The Karoo Supergroup sediments and volcanics have moderate potential, with the argillaceous rocks having the worst potential of the Karoo rocks. The contacts between the Ecca Group shales and sandstone have the best potential of the Karoo Supergroup sedimentary hydrolithologies. Fractures clearly enhance the groundwater potential of most hydrolithologies, with fractured Dwyka Group tillites rated as having one of the best development potentials of all the hydrogeological targets in KwaZulu-Natal, despite the hydrolithology's poor water-bearing characteristics. Dolerite contacts with sedimentary rocks are commonly targeted features in groundwater development. However, the results from this research showed that, apart from the Natal Group's contact with dolerite, these targets have poor development potential. In general, contacts between different hydro lithologies. Health related quality was found to be adversely affected in argillaceous hydrolithologies, such as the majority of Karoo rocks which had high levels of sodium and chloride and Natal Metamorphic Province schists which had elevated sodium, chloride and fluoride. Crystalline and arenaceous hydrolithologies generally exhibited good quality groundwater. A comparison between the different geophysical methods for each target feature indicates that there are appropriate methods to use to detect anomalies related to water-bearing features. The large number of dry boreholes drilled in locations with recorded geophysical anomalies can be either a function of the water-bearing characteristics of the formation, human error or background noise. The cost of using the different geophysical methods vary considerably. The order of increasing cost is magnetics, VLF, EM-34, electrical resistivity profiling followed by vertical electrical sounding. Drilling has a large influence on the development potential of certain aquifers due to the high costs involved. Most of the secondary aquifers will require percussion drilling which is the cheapest method of drilling commonly used. Some very unstable formations within fractured or highly weathered rock may need ODEX drilling to enable drilling to advance. ODEX drilling in these conditions is very costly and can double the cost of drilling compared to air percussion. The unconsolidated sediments of the Maputaland Group can only be drilled by mud rotary or ODEX techniques. The relative costs of these two methods arer very similar with ODEX being slightly cheaper. Because of the high expense of drilling in the sands it is recommended that alternative sources, possibly from shallow hand-dug wells, be considered as appropriate methods of accessing groundwater. The aspects of groundwater yield and quality of aquifers, appropriate geophysical siting and drilling methods, together with social considerations will all contribute to the success of groundwater development in rural KwaZulu-Natal.

Salt marshes are some of the most productive ecosystems in the world and have been the centre of attention over the past few decades, due to their decline as a result of global climate change and anthropogenic impacts. The growth of salt marshes is determined by substrate type, soil conductivity and elevation. The permanently open Keurbooms Estuary along the south-east coast of South Africa is subjected to occasional fluvial flooding and its intertidal area lacks well developed salt marshes, with Spartina maritima restricted to the lower reaches of the Bitou tributary and a few sections of the Keurbooms tributary. Presumeably because of fine sediment habitat in the confluence and lower Bitou tributary. The salinity of the estuarine water ranges between 0.1 – 26.9 and 3.2 – 35.3 in the Bitou and Keurbooms tributaries respectively. A typical salt wedge salinity pattern is common in the Keurbooms tributary where saline water often intrudes underneath the freshwater, especially during high river flows. The following hypotheses were developed and tested in this study: The limited spatial distribution of S. maritima in the Keurbooms Estuary is due to limited availability of fine sediment habitat; and the source of the fine sediment in the estuary is the Bitou tributary rather than the Keurbooms tributary or the sea. It was further postulated that after sediment characteristics, floods are the major hydrological driver determining the distribution of S. maritima in the Keurbooms Estuary. The results of the surveys of the estuarine channel bottom sediments showed that the Keurbooms tributary was mostly characterized by the sand-size sediment fraction derived from the feldspathic and sandstone with evidence of fine sediment fractions restricted to the upper reaches at the confluence with Whiskey Creek. The Bitou was almost always composed of coarse sized sediments in the upper reaches, fine sediment deposits in the middle and lower reaches and medium sorted sand with almost no clay or calcium carbonate in the estuarine component below the confluence of the tributaries. These findings were further supported by the surface sediment deposited within the S. maritima intertidal salt marsh, which showed finer sediment deposits in the Bitou marsh compared to the Keurbooms marsh surface. Similar results were also found in the sediment cores, with the Keurbooms marsh sediment becoming finer with increasing depth whereas fine sediments reduced with depth in the Bitou marsh. The results of the sediment mineralogy indicated that the increased concentrations of clay minerals in the S. maritima surface sediments are derived from the Bokkeveld shale, siltstone and clay slate exposed above the N2 Bridge in the Keurbooms Estuary. GIS mapping shows that S. maritima has been declining over the past two decades, with rapid decreases especially evident after big flooding events. The GIS mapping also indicates that the patches of the S. maritima in the Keurbooms tributary are more exposed to big floods than the Bitou marsh. Despite showing an overall decline, S. maritima area coverage remained more consistent in the lower reaches of the Bitou tributary than in the Keurbooms tributary. Despite the larger and more persistent area cover, the S. maritima plants were shorter and less dense than the plants growing in the sandy substrate. The black/grey colouration of soil with increasing depth in the Bitou tributary was an indication of the reduced state of the soil caused by prolonged waterlogged conditions. The roots of S. maritima in both tributaries were mostly restricted to the sub-surface substrate layer (i.e. 0 – 0.25 m), although the Bitou populations showed more vegetative propagation than the Keurbooms populations. This mechanism of reproduction was also demonstrated during the transplant experiment which showed a greater number of new stem production in the fine sediment substrates compared to the sandy silt substrates. Although accretion rates were not determined in this study, the short-term sediment deposition rates revealed that sedimentation is active in the marshes of the Keurbooms Estuary. Therefore, in spite of showing a decline in area cover, the production of viable seed and observed vegetative propagation suggest that the S. maritima is likely to colonize open stable intertidal mudflats / sandflats, thus maintaining its distribution as an intertidal species in the salt marshes of the Keurbooms Estuary.

Water quality characteristics of the heavily urbanised and industrialised Swartkops River catchment in the Eastern Cape has been the focus of several studies since the 1970s. Overloaded and poorly maintained wastewater treatment works (WWTW), polluted stormwater runoff and solid waste have had a negative impact on the water quality status of the Swartkops River and estuary. Past studies have revealed that a distinct relationship exists between land use activities and the water quality of the Swartkops Estuary, which in turn has raised concerns pertaining to the ecological, economic, recreational, and cultural value of the estuaryThe Swartkops Estuary has a Present Ecological State (PES) of Category D(1) and a Recommended Ecological Category (REC)(2) of a C (Van Niekerk et al., 2014). It is known that effective management of anthropogenic impacts on coastal systems requires a thorough understanding of the system’s biological responses to wastewater discharges and to hydrologic changes. For this reason, the objective of this study was to determine the current water quality status of the Swartkops Estuary, and to gain greater insight into factors controlling eutrophication. This was important as outdated water quality information was used in the Swartkops Integrated Environmental Management Plan (Enviro-Fish Africa, 2011) due to a lack of current data. To determine the current water quality status of the Swartkops Estuary the present study investigated spatial and temporal variability in physico-chemical parameters, nutrients, phytoplankton biomass and community composition, faecal bacteria, and “where possible” related this to historical water quality data. In general, points of entry into the estuary were investigated for their impact on nutrient enrichment and the bacteriological status of the estuary. Water quality surveys were completed in September 2012, November 2012, February 2013, May 2013 and August 2013The present study found evidence to suggest that water is not flushed as efficiently from the estuary as was previously the case, and that the natural hydrology of the estuary has been modified. These changes appear to be the effect of large volumes of wastewater discharges from the wastewater treatment works (WWTW), which has led to the additional stresses of increased vertical stratification and reduced vertical mixing. A build-up of dissolved inorganic nutrients has given rise to persistent eutrophic conditions and phytoplankton blooms occurring from the middle reaches to the tidal limit of the estuary. These findings were associated with a generally well oxygenated estuary; however, bottom water hypoxic conditions were recorded in the upper reaches of the estuary on two occasions and were generally associated with bloom.forming flagellates. Elevated concentrations of inorganic nutrients stimulated phytoplankton to attain high biomass ranging from 0 to 248 g l-1 (31.8 6.56 g l-1). All nutrients displayed positive linear gradients from the mouth to the tidal limit, showed significant (p < 0.05) temporal and spatial variability, and were significantly (p < 0.05) correlated with phytoplankton biomass. Phytoplankton blooms (> 10 000 cells ml-1) of several different groups were recorded from the middle reaches of the estuary to the tidal limit. Diatoms were the dominant group during increased freshwater inflow (at mean daily flow rate of 2.14 m3 s-1) and low DIP levels, whereas flagellates were generally the dominant group during reduced flow and under higher nutrient levels. Although the different tidal stages had no effect on phytoplankton biomass per se, it did support co-existance between phytoplankton groups. This was noted during the spring ebb tide in September 2012 (i.e. flagellates, diatoms and dinoflagellate) and in February 2013 (i.e. dinoflagellates, diatoms and chlorophytes). Phytoplankton blooms have become persistent in the middle to upper reaches of the estuary where chlorophyll-a was > 20 µg l-1 and cell density exceeded 10 000 cells ml-1; a situation not reported in previous studies. The Motherwell Canal was and still is the main source of nitrogen (generally in the form of ammonium) to the estuary, whereas the Swartkops River is still the primary source of phosphorus to the estuary. Since the stormwater canal services the large residential area of Motherwell where leaks in the sewer system, the dumping of night soil buckets, and faulty pumps are often reported, polluted discharges from the Motherwell Canal can enter the canal at any given point. In contrast to the canal, DIP loading from the Swartkops River to the estuary generally occurred under conditions of low flow, whereas nitrogen showed no apparent relationship. Faecal bacteria originating from the Motherwell Canal had the most profound effect on the bacteriological status in the middle reaches of the estuary, whereas the Swartkops River had an intermediate effect due to bacteria die-offs occurring between the point of release from the WWTW to the riverine reaches and the tidal limit of the estuary. Nevertheless, Escherichia coli and enterococci levels are still high, especially in the summer months rendering the estuary unsafe for recreation during this season. Historical data on trace metals in the water column were limited and thus observations from the present study could not be concluded with much confidence. However, preliminary data suggest that levels of copper, zinc, iron and cadmium have increased by at least 90% in the estuary, at the tidal limit of the estuary and in the Markman and Motherwell canals. High inputs of nutrients, trace metals and faecal bacteria to the estuary from land-use activities indicate the necessity for remedial actions with the main objective being to conserve and protect the estuary’s recreational, ecological and economic functions.

South African commercial plantations occupy an estimated 1.5 million hectares of the country and as the demands for timber products increase, this area is expected to increase. However, further expansion is limited, not only by the suitability of land, but also by the pressures from other water users. As a result the need has arisen for simulation models that can aid decisionmakers and planners in their evaluation of the water requirements of forestry versus competing land uses at different spatial scales. Different models exist to perform such tasks and range from simple empirical models to more complex physically-based models. The policies of the National Water Act (1998) relating to forestry serve to highlight the requirements of a model used for the assessment of afforestation impacts and these are discussed in this document. There is a perception that physically-based distributed models are best suited for estimation of afforestation impacts on a catchment's water yield since their physical basis allows for extrapolation to different catchments without calibration. Furthermore, it is often stated that the model parameters have physical meaning and can therefore be estimated from measurable variables. In this regard, a review of physically-based modelling approaches and a comparison of two such hydrological models forms the main focus of this dissertation. The models evaluated were the South African ACRU model and the Australian topography-based Macaque model. The primary objective of this research was to determine whether topography-based modelling (Macaque model) provides an improved simulation of water yield from forested catchments, particularly during the low flow period, compared to a physically-based model (ACRU model) that does not explicitly represent lateral sub-surface flow. A secondary objective was the evaluation of the suitability of these models for application in South Africa. Through a comparison of the two models' structures, the application of the models on two South African catchments and an analysis of the simulation results obtained, an assessment of the different physically-based modelling approaches was made. The strengths and shortcomings of the two models were determined and the following conclusions were drawn regarding the suitability of these modelling approaches for applications on forested catchments in South Africa:• The ACRU model structure was more suited to predictive modelling on operational catchments, whilst the more complex Macaque model's greatest limitation for application in South Africa was its high input requirements which could not be supported by the available data. • Despite data limitations and uncertainty, the Macaque model's topography-based representation of runoff processes resulted in improved low flow simulations compared to the results from the ACRU simulations, indicating that there are benefits associated with a topographically-based modelling approach. • The Macaque model's link to the Geographic Information System, Tarsier, provided an efficient means to configure the model, input spatial data and view output data. However, it was found that the ACRU model was more flexible in terms of being able to accurately represent the spatial and temporal variations of input parameters. Based on these findings, recommendations for future research include the. verification of internal processes of both the ACRU and Macaque models. This would require the combined measurement of both catchment streamflow and processes such as evapotranspiration. For the Macaque model to be verified more comprehensively and for its application in operational catchments it will be necessary to improve the representation of spatial and temporal changes in precipitation and vegetation parameters for South African conditions. / Thesis (M.Sc.)-University of Natal ,Pietermaritzburg, 2002.

The hydrology of the Kudumela Wetland, Limpopo Province of South Africa was studied from November 2005 to April 2007, involving both fieldwork and laboratory analyses. This study presents the results of an investigation of the hydrology of the Kudumela Wetland in South Africa, and its contribution to dry season flow in the Mohlapitsi and Olifants Rivers. Initially, 40 Piezometers were installed along seven transects and water levels monitored in order to understand water table level characteristics (fluctuations) with time. Water levels in transects one, three, the right bank portion of transect four and transect six showed fluctuations. Transect two, the left bank portion of transect four and transect five did not show significant temporal changes. The relationships between piezometer water levels, rainfall in the study area and stream flow observed at a river gauging station are not clear. The river within the wetland is a gaining stream because the water table level elevation is above that of the river. This indicates that the wetland is feeding the river. The northern part of the wetland (T1 and T2) is affected by artificial drains and most of the piezometers closest to the river channel showed the lowest variations. The relationships between rainfall, groundwater, and surface water at this site shows that stream flow did not respond quickly to precipitation as expected, even in months when rainfall increased (for example, 74 and 103mm during 08/02/06 and 18/02/06 respectively), and the groundwater levels did not show fluctuations, indicating that groundwater responds gradually to precipitation, and that the relationship between rainfall, groundwater and surface water is complex. The environmental stable isotopes (deuterium and oxygen-18) and the radioactive isotope (tritium) were analyzed, along with field observations of electrical conductivity (EC), pH, total alkalinity (Talka) and some major and minor dissolved ion analyses for tracing water dynamics in the study area. A total of 39 water samples was taken and analyzed from boreholes, auger holes, right bank and left bank drains, various points along the river and springs in four sampling visits to the wetland. The results did not clearly provide a temporal record of isotope and chemical variations in the various sources. Results from the most extensive sampling survey in April 2007 provide the most comprehensive overview of hydrological relationships. Clustering of the stable isotope data suggests that the water samples of upstream and downstream river, auger holes further south and most drains clustered together suggesting a common water source and almost all samples fall above the global (GMWL) and local (Pretoria MWL) meteoric water lines, while some fall between the global and Pretoria meteoric water lines. Six representative water samples were analyzed for major ion concentration. Both cation (Ca, Mg, K, and Na) and anion (HCO3, SO4, Cl, and NO3) analyses in November 2007 confirmed conclusions reached from field observations. The analysis shows that a single type of water (Ca, Mg-HCO3) is involved in the study area. In almost all major ion plots, the right bank drains, upstream river and downstream river samples grouped together in a single cluster. As the means for reliable river flow measurements were not available, except for the gauging station at the outlet of the valley, rough, semi-quantitative estimates were made during several field visits. These, suggest considerable losses of river flow into the gravel/boulder beds at and below a gabion dam at the head of the valley. Three major and several other left bank springs and right bank drains at transects T1 and T2 contributed to the river flow at all times. Along with the isotopic and chemical evidence, these observations have lead to a hypothesis that river water enters the wetland and flows back to the Mohlapitsi River through boulder beds underlying the wetland and through drains on the surface of the argillaceous aquitard covering the more conductive boulder beds. Deeper dolomitic groundwater does not appear to contribute to the water balance at least in the northern half of the wetland. Although environmental isotope and hydrochemistry results may not unequivocally prove this hypothesis they do not contradict it.