The safety of dams is crucial in ensuring the continual availability of water, safety of the surrounding communities and infrastructure. Surveillance systems are implemented to monitor the structural integrity of certain dams which have a safety risk. The components and extent of the surveillance systems adopted depends on many factors, which include the type of dam wall structure used to impound the reservoir, geotechnical and environmental conditions. The case study used for this thesis is Kouga Dam located in the Eastern Cape Province of South Africa. It is a double curvature, concrete arch dam which supplies water for domestic, irrigation, and industrial use to the Gamtoos River Valley and Nelson Mandela Metropolitan. During construction the stability of the right flank was questioned and subsequently remedial measures were taken in order to increase the shear resistance of this flank. Previous dam safety evaluations also noted the possibility of Alkali Silica Reaction (ASR) occurring within the structure which resulted in concrete swelling and loss of strength. Due to these factors and the large hazard potential rating associated with this dam, an intensive surveillance system has been used to monitor the dam's behaviour during operation. In this thesis the results of the surveillance system is analysed. A strong linear relationship exists between the temperature loading and displacement response of the dam wall. Changes in temperature initiate the response of the structure almost instantaneously. A more complex relationship exists between hydrostatic loading and the displacement response of the structure. A phase lag of approximately one to three months is evident between these two variables. Since construction the displacement and strain rates in the upstream (y) and upward (z) directions are 0.3mm/annum and 8.6με/annum respectively. However, since 1989 there has been a reduction in the average displacement and strain rates in all directions by approximately 70%. This may suggest that the ASR has stabilized. The vertical construction joints, especially the central and upper joints, are relatively open during low water levels. The structure is found to transfer the imposed loading mainly to the central foundation via dominant cantilever action. As a result the reaction forces on the upper foundation have been found to be relatively low, lowering the risk of potential shear failure of the right foundation. Small foundation movements of less than 0.3mm have been observed within the foundation downstream of the dam wall on the right flank. These movements are between 10 and 40m within the foundations.

The problem of water loss from water distribution systems is an issue that faces municipalities worldwide. A large proportion of water loss is a result of leakage. With increasing water scarcity across the globe, it is imperative to conserve water resources, and hence reduce leakage in water distribution systems as best we can. Leaks develop in various different forms, and they form in pipes of all materials. It has been observed in numerous cases around the world that pipe material has been removed from the pipe surfaces adjacent to leaks in excavated failed water distribution pipe specimens. It was proposed by various researchers that this pipe material was being removed as a result of abrasive soil action caused by the pipe leak itself. When pressurised water distribution pipes fail, they emit high velocity jets of water into the surrounding soil bed. Research has shown that high velocity jets of water entering a granular soil bed have the ability to fluidise the bed, allowing the granular particles to move freely. This fluidisation mechanism is known as internal fluidisation. The concept of internal fluidisation offers an explanation for the erosion of pipe material adjacent to pipe leaks. In this study, the removal of pipe material adjacent to leaks due to internal fluidisation has been termed “leakage-induced pipe erosion.” This phenomenon has received minimal attention from researchers in the past. Leakage-induced pipe erosion has the potential to aggravate small existing leaks. There are two main implications of aggravating small leaks; firstly, where water authorities do not utilise active leak detection programmes, aggravating the initial leak conditions of small leaks can result in increased long term water losses. Secondly, in water distribution systems where the water authority does implement active leak detection programmes, aggravating small leaks increases the probability of finding and repairing them. This aim of this study was to experimentally investigate the influence of various factors on the leakage-induced erosion process. Five main factors were investigated, namely bedding material grain size, cover depth, leakage flow rate, initial leak orientation and pipe material. An experimental setup was designed and manufactured in order to provide a controlled environment in which to investigate the factors affecting leakage-induced pipe erosion. It was demonstrated in this study that small leaks have the ability to develop into larger leaks due to the erosion process. It was found in this study that of the five aforementioned factors, the orientation of the initial leak has the greatest influence on the rate of leakageinduced pipe erosion. It was also found that larger grain sizes and larger flow rates increased the rate of leakage-induced pipe erosion significantly. It was further demonstrated that of the three pipe materials that were tested, uPVC was the most susceptible to the erosion, while steel and HDPE exhibited more resistance to the soil abrasion. It was also found in this study that the cover depth of the soil bed had a small effect on the rate of leakage-induced pipe erosion within the tested range of bed heights.

Universal access to sustainable and equitable sanitation is a Sustainable Development Goal on the 2030 Agenda for Sustainable Development. The South African government has taken strides to try and meet both international and domestic development goals with its Free Basic Sanitation policy, for which a national implementation strategy was developed in 2008. Although the policy was formulated at a national level, municipal governments are delegated the authority to ensure service delivery at the local level. Municipalities have adapted and interpreted the policy to suit their own contexts. In particular, they have attempted to address the challenge of providing sanitation services to informal settlements using different approaches with varying degrees of success and often without explicit consideration or guidance for how to incorporate sustainability and equity principles. The aims of this thesis are thus to explore how the concepts of sustainability and equity can be applied to improve municipal sanitation services in South African informal settlements and to explore how various dimensions of sanitation and equity relate to sanitation. A comparative case study method using the lens of sustainability and equity was used to critique the approaches to providing sanitation services to informal settlements in three of South Africa’s largest municipalities: eThekwini (Durban), Johannesburg and Cape Town. Each municipal case study incorporated an embedded case study that was used to examine sanitation services in selected informal settlements at a programme, project or settlement level. Primary data was collected using interviews and field visits. Secondary data was obtained from national and municipal records such as water and sanitation department reports, census data from Statistics South Africa, and municipal geographical information system databases. Findings from the thesis indicate that there is a need to better incorporate multiple stakeholders’ perspectives on what sustainable and equitable sanitation services should be like. Strengths and weaknesses of each municipality’s approach to sanitation service provision were compared and used to identify factors relating to sustainability and equity. A major conceptual gap identified in sanitation service delivery approaches is the need to emphasise equity as a core tenet of sustainability, especially in a socio-economic context of extreme inequality. This thesis makes a contribution towards knowledge by highlighting the importance of equity to support sustainable sanitation service delivery in South African informal settlements, adding new perspective into different dimensions of equity in sanitation and a suggested framework for how they could be incorporated into M&E practices. Keywords: informal settlements; urban; South Africa; sanitation; sustainability; equity

Construction of roads over soft soils can lead to design and construction related problems linked to the soil’s compressibility characteristics and low strength. Failure, in terms of bearing capacity can occur when pavements are constructed over such soft soils. When road pavements, which are constructed over soft soils, are subjected to cyclic traffic loading (dynamic in nature), rapid deterioration of the base layer material and progressive permanent deformation of the surface will occur. This not only reduces the serviceability of the pavement structure but also its design life. In this study, reinforcement geosynthetics (geogrids and geotextiles) were used as reinforcement inclusions within a granular base overlying a soft subgrade of California Bearing Ratio (CBR) less than 2% in a 1.0 m3 steel test box. Firstly, a geotextile/geogrid was placed at the interface between the base layer and subgrade. Thereafter, a combination of the geotextile at the interface (of the base and subgrade) and geogrid within the base layer. Bench scale plate load tests (static and cyclic) were conducted on a 305 mm diameter circular steel plate on the two layer system using a Universal Compression Machine. Static loading was applied at a rate of 1.2 mm/min. Dynamic sinusoidal load wave was applied with a 4 kN seating load that was linearly increased with an incremental load of 4 kN for every 8 cycles at a frequency of 0.2 Hz on a 305 mm circular plate. For both tests, settlement failure of the composite system was considered at a deformation of 75 mm as defined for unpaved roads. The results obtained from the pavement model showed that there was a significant improvement in bearing capacity and reduction in settlement accruing from geosynthetic inclusion as shown by the Bearing Capacity Ratio (BCR) of 1.21, 1.29 and 1.63 for geogrid, geotextile and geogrid-geotextile combinations respectively. Additionally, a Settlement Reduction Factor (SRF) of 18% for geogrid, 23% for geotextile and 31% for the geogridgeotextile combination resulted. There was also an improvement in extended pavement life as depicted by the Traffic Benefit Ratio (TBR) greater than 1 for all reinforced base layers. An improved performance was realised with the double combination of geotextile at the interface, geogrid at the base. The observed benefits were considered in using the AASHTO pavement design equation, and the resulting geosynthetic reinforced pavements were capable of supporting more than twice the Equivalent Standard Axles (ESALs) of an unreinforced pavement. Furthermore, cost savings of up to 55% in base thickness reduction were realised with the use of geotextile and geogrid in pavements.

Concrete structures, in particular reinforced concrete structures, have been designed and built for many years. Many previously built structures are now being compromised with regard to their condition and structural integrity. There has developed a need to maintain these structures and protect them in order to protect the users and minimise the expenses associated with repair and maintenance. Instead of resorting to demolishing and rebuilding, engineers are becoming increasingly capable of restoring and enhancing existing structures in order to maximise the original structures lifespan. One of the ways in which this is achieved is through the implementation of various repair and maintenance strategies. These strategies can range from cathodic protection of the steel reinforcement to increasing concrete cover of the section. The easiest and often most commonly used method, although often not effective when used on its own, is to coat the concrete surface with a protective coating. There are many coatings available, ranging from simple to complex formulations of inorganic and organic materials. These coatings each perform a unique function and will often be designed to combat a specific problem with which the structure is to be exposed. The formulation of the coatings are almost always designed correctly when they are manufactured, however, premature bond failure due to poor substrate surface preparation and various substrate moisture conditions still seem to occur after coating application. This investigation will give insight into the effects a sound, clean and profiled concrete substrate that is subjected to different moisture conditions will have on the bond strength achieved when using cementitious based coatings. The reason for differing the moisture conditions of the substrate is that often on site the moisture condition of an existing substrate is not known prior to application of the coating. This investigation shows that epoxy modified cementitious coatings provide acceptable bond strength. In some cases, failure within the coating instead of the acceptable failure within the substrate has resulted, due only to the level of dryness of the substrate. A thorough investigation on the theory behind surface coatings, their properties and performance requirements was done and subsequently followed by a detailed experimental programme that was performed and analysed to provide insight to the said coating behaviour. Inorganic coatings behave more predictably when applied to moist substrates as this is the ideal condition for the coating to adequately hydrate. In almost all the specimens with fully dried substrates, the resulting bond strength of the coating was less than that observed with the saturated surface dry substrate condition. It is therefore advised that the moisture condition stated by the coating manufacturer be adhered to as this will give acceptable results with failure taking place within the concrete substrate and not within the coating as a failure of the latter is completely unacceptable when referenced to the European and American set of standards. The European standard BS EN 1504: Products and systems for the protection and repair of concrete structures provide a more complete and reliable framework for the concrete repair industry and hence is preferred over any South African national standard currently available. Further research into the viability of a South African based standard will need to be investigated before being considered for use in this research paper.

strength of soil is characterized by cohesion, angle of internal friction and dilatation. The first two parameters mentioned primarily define the soil’s ability to resist shear stress under specified load. These parameters can be determined by tests conducted either in the laboratory or the field for use in design of geotechnical structures. Some of the tests in the laboratory to determine the shear strength of soil, include triaxial, ring shear, torsional shear, direct shear and direct simple shear. Direct shear test is the most widely used geotechnical shear device due to its simplicity, however, the test suffers from stress inhomogeneity. Direct simple shear apparatus was developed because of the shortcomings in the direct shear test. In these two tests, different shearing conditions are applied to soil samples. For the direct shear test, shearing occurs at a predetermined center of the specimen which may not be the weakest plane of the soil while in direct simple shear, the entire specimen distorts without the formation of single shearing surface. The mode of shearing established in the direct simple shear device is similar to that which occurs around the shaft of a pile. In contrast to the extensive geotechnical application of direct shear test, limited information exists on direct simple shear test. This thesis endeavours to establish the relationship between the two tests by undertaking extensive testing to obtain a better understanding of direct simple shear test as used for testing local soils and to determine a correlation with the results from direct shear tests. A series of shear tests were undertaken on Klipheuwel sand, Kaolin clay and composite of the sand and clay using universal shear device. The soils were mixed with water in percentages of 5, 10, 15, 20 and 25%, and the composite with clay percentages of 10, 25, 50 and 75%, to investigate the impact of water and clay on the shear parameters determined from the two tests. The results showed that direct shear test gives higher shear strength when compared to direct simple shear test under the same soil condition. The addition of water, and clay, generally reduced the internal friction angle of sand for both tests. Furthermore, increase in cohesion was observed with the addition of water to Kaolin clay for the direct simple shear test and the reverse was true in the direct shear test. The correlation factors developed in this study for the direct simple shear test could be used to refine the results from the direct shear test.

The study of the improvements to transport controlling facilities, defined in this dissertation as operational transport upgrades, aims to ascertain its influence on a growing urban traffic demand. This dissertation assesses the influence that the conversion of Werner List Street into a one-way street has on the performance of traffic accommodation. Werner List Street links traffic commuting from the south to the north of Windhoek’s Central Business District (CBD). The CBD traffic demand is estimated to breach network capacity by 2015, a scenario prevalent in many small middleweight cities. The research conducted a literature review on aspects related to the development and implementation of operational upgrades; thereby gaining an understanding on the relevance that such improvements have on small middleweight cities. Studied literature suggests that with the reduction in the allocation of funds to develop transport systems, transport authorities resort to innovative methods of improving transport network utilising minimal capital expenditure. The limited funding is prevalent in small middleweight cities due to the current lack of major traffic impediment. The research studied aspects of Windhoek’s activity system, as presented in the city’s Household Survey of 2004 that primarily focused on the income status, transport mode use and transport mode preference. The origin of trips would assist in determining the direction, in relation to Windhoek’s CBD, peak traffic commutes. The City of Windhoek embarked on establishing a base CBD traffic scenario, from which future improvements could be measured. The assessment of the network demarcated for the base study formed part of the input information used to establish the Strategic Transport Master Plan (STMP). The STMP aims to address the promotion of access to land-use areas (Phase 1), the modal composition distortion (Phase 2) and CBD traffic circulation improvement (Phase 3). The assessment of Windhoek’s transport network has primarily consisted of visual inspections, with the one-way conversion of Werner List Street being the only upgrade completed. The conversion of Werner List Street into a one-way street is an implemented improvement borne from the development of the STMP. The analysis of performance measures along the upgraded street entailed the micro simulation of peak traffic during morning and evening periods. Visual inspections, using the manual traffic count method, were conducted to validate turning movements and queue length build-up at the three major streets intersecting Werner List Street; Fidel Castro Street, Dr Frans Indongo Street and John Meinert Street. Research questions were developed to investigate the influence that operational upgrades had on traffic network performance. The questions catered to:  The incorporation of operational upgrades in the STMP’s methodology,  Operational upgrades as a traffic network improvement measure, and  The future use of output data. The research concluded that the STMP, fundamentally, aims to develop Windhoek’s transport network. The proposed techniques, primarily, focus on operational improvements with limited infrastructure expansion required during the implementation phase. The operational upgrade improves the traffic network’s accommodation transport demand, however, the wholesome implementation of the STMP’s proposed operational upgrades could provide a better measure of improvement. The output data, obtained from the micro simulation of Werner List Street, was used to identify areas needing further attention. It was, therefore, concluded that output data formed a critical role in providing a platform to continually update the CBD traffic network’s performance. Through the assessment of Windhoek’s activity system, it was noted that a significant percentage (40%) of residents have no access to private vehicle ownership use creating a potential market for public transport utilisation. Public transport occupancy per trip far outmatches that attained through public transport. The analysis results, obtained from the micro simulation of the upgraded network and substantiated through results obtained from the conducted visual inspections, depict a slight improvement in the network performance. The John Meinert-Werner List intersection experienced significantly lower queue length levels than the levels established in the base study. The results, further, depict areas of concern along Werner List Street. The queue length measurements for the eastern and western approaches of the Fidel Castro-Werner List intersection and the western approach of the Dr Frans Indongo-Werner List intersection depict relatively high levels of queue length build-up. Further, the analysis of the traffic commuting between the Fidel Castro-Werner List intersection and the Dr Frans Indongo-Werner List intersection depicts a low average speed, as compared with the average speeds measured for the other links. The implementation of additional upgrades, preferably operational upgrades, should, therefore, aim to reduce the impact at the assessed areas of concern whilst reducing overall traffic volumes within the CBD. The research recommends a pragmatic implementation of the STMP, specifically the network continuity and ring route formations. These methods of network upgrading aim to reduce vehicle volumes within the CBD by providing alternate route choices for traffic that commutes across the CBD to reach the destinations outside the CBD. Additionally, the incorporation of modern technology would provide updated count data and monitors the network’s performance. Modern technology, such as the lane occupancy detectors and tolltag transponders, provides continually updated quantitative data at traffic intersection. This dissertation primarily focuses on the operational aspects of the transport network. The possibility of attending to pedestrian movement and public transport regulation within the CBD would, further, improve the academic understanding of Windhoek’s transport system and promote a holistic approach to improving the network’s capacity.

Background With the relatively recent emergence of non-petroleum-derived aviation gas turbine fuels, it was appropriate to review the complete list of jet-fuel specifications to assess whether they were sufficiently robust to ensure fit-forpurpose within the new paradigm. Although this has been an industry-wide endeavour, there were some particular research areas that were identified for special in-house attention by Sasol, as the world’s first commercial producer of approved and certified semi-synthetic and fully synthetic jet fuel. The project described in this report formed part of one of these research areas, which pertained to ignition and combustion stability in gas turbines and the role played by various fuel attributes and properties. The project was conducted at the Sasol Advance Fuels Laboratory based at the University of Cape Town. Objectives The project entailed the design and construction of a combustion test facility for conducting synthetic jet fuel research. The primary intended focus of the facility was the investigation of ignition and combustion stability behaviour of various test fuels, ranging from commercial jet fuel to single component model fuels. The scope of the project also included the design of both a basic homogeneous and a heterogeneous combustor which served to verify the facility’s suitability for investigating the influence of fuel chemistry and combustor inlet conditions on ignition and combustion stability limits. Test facility design Design criteria, such as the required test condition range, facility scale, cost and safety, were considered during the generation of design concepts and the selection of the final facility design. The facility design was approached as a number of integrated subsystem designs. The final facility design employed a single positive displacement blower that allowed testing to be conducted under both vacuum and pressurised combustor inlet conditions depending on the configuration of the flow control valves. An absolute pressure range of 70kPa to 150kPa was attainable over a temperature range of 263K to 340K. The fuel system allowed primary zone equivalence ratios of 0.3 to 1.5 over the full air mass flow range of 0.85kg/min to 18kg/min.. This allowed the study of both temperature and pressure influences on ignition and combustion stability limits. A homogeneous and a heterogeneous combustor were designed to allow the study of both fuel chemistry influences in isolation and in conjunction with mixing and evaporation effects. Test programme As the sign-off acceptance criterion for the commissioning of the test facility, a test programme was conducted with a small selection of single component model fuels and some petroleum-derived Jet A-1. These tests were used to provide not only proof of the facility’s capabilities, but also to confirm the sensitivity of the equipment to detect and measure the expected influence of autoignition chemistry on threshold combustion performance. Tests with single component model fuels were performed using the premixed homogeneous combustor to assess the measurement capability of the test facility in terms of the influence of autoignition chemistry on lean ignition and lean blowout behaviour. This was followed by tests with petroleum-derived Jet A-1 in the heterogeneous combustor to assess the temperature and pressure dependence of ignition and combustion stability behaviour. Finally in order to determine how the results obtained in the homogeneous combustor translated to a heterogeneous environment, the lean blowout behaviour of two single component model fuels were compared with that of petroleum-derived Jet A-1. Conclusions The test programme provided conclusive evidence of the successful commissioning of the test facility. The results of the pressure and temperature influence evaluation clearly illustrated the repeatability of test results and the suitability of the test facility and the heterogeneous combustor design for investigating the ignition and extinction behaviour of practical synthetic jet fuel alternatives. The results of the fuel autoignition chemistry evaluation, using both combustor designs, revealed evidence of the influence of fuel chemistry and physical property effects. These results were seen to validate the motivation for designing and constructing a facility that would enable further study of the influence of fuel chemistry on ignition and extinction behaviour, and its particular relevance to synthetic gas turbine fuel formulation.

This thesis starts from the position that the enterprise of transport planning is failing to deliver a transport system which fully meets the needs of urban dwellers. Criticisms of the process are presented and it is argued that one reason for the lack of delivery is the insufficient attention paid by transport planning professionals, and researchers, to the decision-making process. The nature of decision-making is explored by reference to various theoretical models of decision-making, and it is concluded that decision-making in an existing situation must be comprehensively described, before recommendations for its improvement can be made. Cape Town is chosen as the city for investigation. Checkland's Soft Systems Methodology is used for describing urban transport decision-making in Cape Town, and the approach is informed by seventeen in-depth interviews and forty-five questionnaires with decisionmakers. The analysis yields insights which are categorised into four areas: issues of institutional re- . organisation and funding; issues of politician involvement; issues of change in officials; issues of public involvement. An action plan is developed from the insights. It is concluded that the Soft Systems Methodology was useful in highlighting problematic areas in the decision-making process. The Cape Town experience is briefly compared with that of six other case studies. It is concluded that the Cape Town experience, although unique due to political upheaval in South Africa, also has some similar traits to that of other case studies. Finally some general conclusions about urban transport decision-making are made, which could be useful in considerations of how to improve the delivery of urban transport.

This thesis is concerned with rapid chloride testing of concrete. In particular, it describes the development of the rapid chloride conductivity test which measures the physical resistance of concrete to the transport of chloride ions. The test was developed to be used in engineering practice for, inter alia:- • marine concrete durability research • predicting long-term chloride ingress into concrete • construction quality control. The test is of low cost, has a sound theoretical basis, and as far as can be ascertained, is the most rapid of all chloride tests developed to date. Due to the short test duration, it is possible to test concretes cured for very short periods. Samples can also be left immersed in chloride solution and re-tested at later ages to study time-dependent phenomena such as chloride binding. The chloride conductivity test involves the standardisation of the concrete pore solution by vacuum-saturating a concrete sample with a 5 M NaCl solution. A single conductivity measurement of the sample is then performed. Different concrete samples yield different conductivities primarily because of differences in their pore and micro-structure. The measured conductivity is related to the diffusibility ratio as well as to the theoretical steady state chloride diffusivity of concrete. Trial tests were performed to determine the optimum conditions for the test (sample conditioning. test procedure) and to verify the test relationships. The conductivity test apparatus was designed to: • • • accurately measure DC conductivity be inexpensive and easy to use allow rapid changeover from one sample to another. The chloride conductivity test was standardised by means of a ruggedness test (ASTM E 1169-89). which determines the allowable ranges in the critical variables involved in the test method. Proper control over these variables was subsequently set out in the test method specifications. The single-operator coefficient of variation of the test is estimated to be 6 % . while the range of typical test results may span approximately two orders of magnitude. The test is therefore very sensitive to the measured property. An extensive laboratory- and site-based experimental programme was carried out to demonstrate the possibility of using the test for the purposes stated. The laboratory tests involved the durability characterisation of a range of concrete mixes using the chloride conductivity test, an oxygen permeability test and a water sorptivity test. Toe site tests involved correlating the test results with chloride ingress in-situ in a marine environment. The chloride conductivity test was shown to be suitable for durability research due to the following characteristics: • high precision • sound theoretical basis • short test duration. The good correlation obtained between the chloride conductivity results and in-situ chloride ingress parameters verified the potential of the test to predict chloride ingress. The test should ideally be coupled with a chloride ingress model if it is to be used for the prediction of long-term chloride ingress. Such a chloride ingress model based on the chloride conductivity test has recently been developed in the UCT laboratory. The chloride conductivity test can be used for construction quality control as it is sensitive to most of the quality-related parameters: • • • • water/cement ratio cement type cement content curmg • compaction For quality control purposes the test can be thought of as an index or characterisation test. since it yields a reproducible measure of the physical resistance of concrete to the passage of chloride ions. The use of such a durability index test in conjunction with a covermeter survey after construction would help ensure compliance with specifications. The work done has indicated the possibility of developing a further test that would give a rapid indication of the chloride binding characteristics (chemical resistance to chloride ingress) of concrete. The test would involve saturating concrete with a lower concentration chloride solution ( 1 - 3 M NaCl) and then measuring the reduction in conductivity with time. With lower concentration chloride solutions, the test becomes very sensitive to a reduction in chloride concentration in the pore solution due to chloride binding.