Submitted in fulfilment of the requirements of the degree of Master of Technology: Environmental Health, Durban University of Technology, 2014. / The study entailed the determination of key water quality parameters significantly influencing treatment cost in uMngeni Basin. Chemical dosage was used as a substitute for treatment cost as the study indicated that cost, in its monetary value, is influenced by market forces, demand and supply, which are both not directly linked to water quality. Chemical dosage is however, determined by the quality of water and thus provides a clear illustration of the effect of pollution on treatment cost. Three specific objectives were set in an effort to determine key water quality parameters influencing treatment costs in uMngeni Basin. The fourth objective was to develop a model for predicting chemical dosages. The first approach was analysis of temporal and spatial variability of water quality in relation to chemical dosage during production of potable water. The trends were explained in relation to river health status. For this purpose, time-series, box-plot, and the Seasonal-Kendal test were employed. The results showed that the quality of water significantly deteriorated from upstream to downstream in relation to algae, turbidity and Escherichia coli (E. coli). High mean range of E. coli (126-1319 colony count/100mL) and turbidity (2.7-38.7 NTU) observed indicate that the quality of water along the basin is not fit for human consumption as these parameters exceeded the target range stipulated in South Africa’s guidelines for domestic use. For water intended for drinking purpose, turbidity should be below 5 NTU, while zero E. coli count is expect in 100 mL. Among the six sampling stations considered along the uMngeni Basin, three dam outflows (Midmar, Nagle and Inanda) showed an improved quality compared with their respective inflow stations. This was expected and could be attributed to the retention and dilution effects. These natural processes help by providing a self-purification process, which ultimately reduces the treatment cost. While considering the importance of disseminating water quality information to the general public and non-technical stakeholders, the second objective of the study was to develop two water quality indices. These were; (1) Treatability Water Quality Index and (2) River Health Water Quality Index. The Treatability Water Quality Index was developed based on the Canadian Council Minister of Environment Water Quality Index (CCME-WQI). The technique is used to determine fitness of water against a set of assigned water quality resource objectives (guidelines). The calculated Harmonised Water Quality Resource Objectives (HWQRO) were used to compare the qualities of the raw water being abstracted at Nagle and Inanda Dam for the purpose of treatment. The results showed that Nagle Dam, which supplies Durban Heights, is significantly affected by E. coli (42% non-compliance), turbidity (20% non-compliance) and nitrate (18% non-compliance) levels. Wiggins Water Treatment Plant which abstracts from Inanda Dam has a problem of high algae (mean 4499 cell/mL), conductivity (mean 26.21 mS/m) and alkalinity (mean 62.66 mg/L) levels. The River Health Water Quality Index (RHWQI) was developed using the Weighted Geometric Mean (WQM) method. Eight parameters, namely, E. coli, dissolved oxygen, nitrate, ammonia, turbidity, alkalinity, electrical conductivity and pH were selected for indexing. Rating curves were drawn based on the target ranges as stipulated in South Africa’s guidelines for freshwater ecosystems. Five classes were used to describe the overall river health status. The results showed that the water is still acceptable for survival of freshwater animals. A comparison of the RHWQI scores (out of 100) depicted that dam inflow station (MDI(61.6), NDI(74.6) and IDI(63.8)) showed a relatively deteriorated quality as compared with their outflows (MDO(77.8), NDO(74.4) and IDO(80)). The third objective was to employ statistical analysis to determine key water quality parameters influencing chemical dosage at Durban Heights and Wiggins Water Treatment Plants. For each of the two treatment plants, treated water quality data-sets were analysed together with their respective raw water data-set. The rationale was to determine parameters showing concentration change due to treatment. The t-test was used to determine the significance of concentration change on each of the 23 parameters considered. Thereafter, the correlations between water quality parameters and the three chemicals used during treatment (polymer, chlorine and lime) were analysed. The results showed that the concentrations of physical parameters namely, algae, turbidity and total organic carbon at both treatment showed a significant statistical (p<0.05) reduction in concentration (R/Ro<0.95). This results implies that such parameters were key drivers for chemical dosage. From the results of the first three objectives, it is recommended that implementing measures to control physical parameter pollution sources, specifically sewage discharges and rainfall run-off from agricultural lands along the uMngeni Basin should assist in reducing the chemical dosage and ultimately cost. The fourth objective was to develop chemical dosage models for prediction purposes. This was achieved by employing a polynomial non-linear regression function on the XLStat 2014 program. The resultant models showed prediction power (R2) ranging from 0.18 (18%) up to 0.75 (75%). However, the study recommends a comparative study of the developed models with other modelling techniques.

Submitted in fulfillment of the requirements of the degree of Master of Engineering: Chemical Engineering, Durban University of Technology, Durban, South Africa, 2016. / River water is a life supporting watercourse to most communities in rural areas. It is used for both human and animal consumption, and is well becoming a collection channel for defecation and urination due to shortage or lack of access to running water and sanitation facilities. This has resulted to the contamination of water sources, which poses a great risk to human health. This has motivated researchers to study simple but yet robust systems to produce safe drinking water. Photocatalysis is one of such emerging disinfection technologies. Titanium dioxide (TiO2) which is one of the basic materials used for paint manufacturing has emerged as an excellent photocatalyst material for water purification. TiO2 was selected in this study because it is locally available with a potential to open a new market in water purification for the manufacturers. The setback in previous studies is the recovery of nano-scaled TiO2 (nTiO2) after purification when used as a suspension in treated water. Thus this study evaluates the performance of four grades of micro-scaled TiO2 (mTiO2) on the degradation of organic matters, Escherichia coli (E. coli) and total coliform in river water and to investigate the percentage recovery of the mTiO2 using a locally manufactured Polyester Woven Fabric Microfiltration (PWFMF) membrane. The PWFMF though uncharacterized has been used in a number of studies for treating domestic and industrial waste waters. The best-performing grade was used to optimize the degradation efficiency of E. coli in river water using the Design of Experiments (DOE) methodology. Grade 2 of the mTiO2, which is hydrated titanium dioxide with additions (ahTiO2) of particle size range of 0.2 – 53 µm at a concentration of 2.5 g/l displayed an advantageous photocatalytic activity. The results show that 80 % of the organics were removed in 3 hours and increased to 93% after 6 hours. Two particle size ranges of 0.2 – 53 µm and 54 – 75 µm at a concentration of 5 g/l degraded organic matters to 90 % and 77 % in 3 hours respectively. The particle size range of 0.2 – 53 µm at a concentration of 5 g/l was then filtered using a PWFMF and turbidities went below 1 NTU after 20 minutes from feed turbidity of 470 NTU for all three trials. The average percentage recovery in 2 hours was 98.91 %. The four grades of mTiO2 were analyzed for E. coli and total coliform for 4 hours at concentrations of 2, 5 and 7 g/l. Grade 2 achieved the E. coli specification of 0 count/ 100 mL at 5 g/l in 2 hours and at 7 g/l in 0.5 hours. Grade 4 E. coli specification was achieved with 7g/l in 4 hours. Grades 2 and 4 performed better since they both achieved the E. coli and total coliform specifications. Grade 2 was the best performing grade and was considered for statistical studies. Grade 2 was then used on a comparative study between the Central Composite Design (CCD) and Box-Behnken Design (BBD), which are two of the major Response Surface Methodologies (RSM). The CCD compared to BBD provides high quality predictions over the entire design space. The CCD obtained optimum results for concentration of mTiO2 (X1), temperature (X2), initial pH (X3) and aeration (X4) which were 6.94 g/l, 28.75 OC, pH = 6.04, and 13.35 L/min for the maximum degradation efficiency of 99.85 % which showed comparable optimum results to the BBD that were 6.45 g/l, 28.28 OC, pH = 6.02 and 12.21 L/min for the maximum degradation efficiency of 99.80%. These theoretical model results were validated by practical experiments that produced the maximum degradation efficiency for CCD and BBD of 99.67 and 99.26 % respectively. Grade 2 of the mTiO2 can be used as a photocatalyst for river water purification due to its strong ability for the removal of E. coli. The additions used in grades 2 and 4 during production improved the photocatalytic activity. The PWFMF membrane showed a great performance of above 98 % particle recovery of mTiO2 from treated water, although there was an indication that the smallest particles were passing through the membrane. The RSM results gave approximately the same optimum results that were well within the limits, which were experimentally validated and showed that the models were sustainable. It is recommended that the effect of additions be studied on the structures or the charge stability of the two grades. / M

M. Tech. (Chemical Engineering, Faculty of Engineering and Technology), Vaal University of Technology. / A novel technology in the treatment of river water, which involves an electrochemical treatment technique to produce domestic or drinking water is being investigated using aluminium and iron electrodes in an electrochemical circuit. Coagulation and flocculation are traditional methods for the treatment of polluted water. Electrocoagulation presents a robust novel and innovative alternative in which a sacrificial metal anode treats water electrochemically. This has the major advantage of providing mainly active cations required for coagulation and flocculation, without increasing the salinity of the water. Electrocoagulation is a complex process with a multitude of mechanisms operating synergistically to remove pollutants from the water. A wide variety of opinions exist in the literature for key mechanisms. A lack of a systematic approach has resulted in a myriad of designs for electrocoagulation reactors without due consideration of the complexity of the system. A systematic, holistic approach is required to understand electrocoagulation and its controlling parameters. An electrocoagulation-flotation process has been developed for water treatment. This involved an electrolytic reactor with aluminium and/or iron electrodes. The water to be treated (river water) was subjected to coagulation, by Al(III) and Fe(II) ions dissolved from the electrodes, resulting in floes floating after being captured by hydrogen gas bubbles generated at the cathode surfaces. Apparent current efficiencies for AI and Fe dissolution as aqueous Al(III) and Fe(II) species at pH 6.5 and 7.8 were greater than unity. This was due to additional chemical reactions occurring parallel with electrochemical AI and Fe dissolution: oxygen reduction at anodes and cathodes, and hydrogen evolution at cathodes, resulting in net (i.e. oxidation plus reduction) currents at both anodes and cathodes. Investigation results illustrate the feasibility of ferrous and aluminium ion electrochemical treatment as being a successful method of water treatment. Better results were achieved under conditions of relatively high raw water alkalinity, relatively low raw water turbidity, and when high mixing energy conditions were available.

South Africa is currently experiencing water shortage crisis, a challenge that has been attributed not only to the scarcity of freshwater, but also to fast degrading water quality. Factors such as rapid urbanisation, population and economic growth, climate change as well as poor operational and maintenance of many of the exisiting water/wastewater treatment facilities have been acknowledged as important contributors to degrading water quality in the country. Untreated or inadequately treated discharged wastewater effluents constitute point source pollution to many freshwater environments in South Africa. Hence, it becomes imperative to evaluate wastewater discharges in other to protect the scarce freshwater resource, the environment and public health. Over a twelve-month sampling period (September 2012 to August 2013), we assessed the bacteriological, virological and physicochemical qualities of the discharged final effluents of two wastewater treatment facilities in the Eastern Cape Province of South Africa. For the physicochemical assessment, a total of 144 final effluent samples were collected from both the final effluent tanks (FE) and the discharge points (DP) of the treatment facilities. Physicochemical parameters including pH, temperature, turbidity, total dissolved solids (TDS), dissolved oxygen (DO), electrical conductivity (EC) and free chlorine concentration were determined on site while biological oxygen demand (BOD), nitrate (NO3-), nitrite (NO2-), phosphate (PO4-) and chemical oxygen demand (COD) were determined in the laboratory. The bacteriological analysis of the samples was done using standard membrane filtration (MF) technique. Bacterial group assessed included: faecal indicator bacteria (faecal coliforms and E. coli) and Vibrio species, while the antibiotic susceptibility profiles of selected E. coli and Vibrio species isolates against some selected antibiotics commonly used in human therapy and veterinary medicine were determind using the standard agar-disc diffusion method. The occurrence and concentrations of human enteric viruses including: human adenovirus (HAdV), hepatitis A virus (HAV) and rotavirus (RoV) in the samples were determined by TaqMan-based real-time polymerase chain reaction (qPCR) following concentration by adsorption-elution method. The physicochemical characteristics of the samples ranged as follows: pH (6.5 – 7.6), TDS (95 – 171 mg/L), EC (134 – 267 μS/cm), temperature (12 – 27 °C), turbidity (1.5 – 65.7 mg/L), free chlorine (0.08 – 0.72 mg/L), DO (2.06 – 9.81 mg/L), BOD (0.13 – 9.81 mg/L), NO3- (0 – 21.5 mg/L), NO2- (0 – 0.72 mg/L), PO4- (0 – 18.3 mg/L) and COD (27 – 680 mg/L). Some of the characteristic such as pH, TDS, EC, temperature, nitrite and DO (on most instances) complied with recommended guidelines. Other characteristics, however, including turbidity, BOD, nitrate, phosphate and COD fell short of the recommended guidelines. All the 48 samples analysed for bacteriological qualities tested positive for the presence of the bacterial groups with significant (P≤0.05) seasonal variation in their densities. Faecal coliforms were detected in counts ranging from 1 CFU/100ml to 2.7 × 104 CFU/100ml. Presumptive E. coli counts ranged generally between 1 CFU/100ml – 1.4 × 105 CFU/100ml while counts of presumptive Vibrio species ranged between 4 CFU/100ml – 1.4 × 104 CFU/100ml. Molecular identification of the presumptive isolates by polymerase chain reactions PCR gave positive reaction rates of 76.2 percent (381/500) and 69.8 percent (279/400) for E.coli and Vibrio species respectively. The antibiotic susceptibility profiling of 205 PCR-confirmed Vibiro isolates against 18 commomly used antibiotics showed resistance frequencies ranging from 0.5 percent (imipenem) to 96.1 percent (penicillin G) at recommended breakpoint concentrations. Eighty-one percent (166/205) of the Vibrio isolates showed multidrug resistance (resistance to 3 or more antibiotics) with the most common multiple antibiotic resistance phenotype (MARP) being AP-T-TM-SMX-PG-NI-PB, occurring in 8 isolates.

Submitted in fulfillment of the requirements of the Degree of Master of Technology: Chemistry, Durban University of Technology, 2014. / This project arose out the need for a simple method to analyse NOM on a routine basis. Water samples were obtained from the Vaal dam, which is one of the dams used by a hydroelectric power station. Analysis was preceded by separation of NOM into the humic and non-humic portions. The humic portion was separated into two fractions by employing a non-ionic resin (DAX-8) to separate humic acid from fulvic acid. High performance size exclusion chromatography (HPSEC), equipped with an Ultraviolet( UV) detector and an Evaporative Light Scattering (ELS) detector connected in series, was used to obtain molecular weight distribution information and the concentration levels of the two acids. Mixed standards of polyethylene oxide/glycol were employed to calibrate the selected column. Suwanee River humic acid standard was used as a certified reference material. The molecular weight distributions (MWDs) of the isolated fractions of humic and fulvic acids were determined with ELSD detection as weight-average (Mw), number-average (Mn) and polydispersity (ρ) of individual NOM fractions. The Mw/Mn ratio was found to be less than 1.5 in all the fractions, indicating that they have a low and narrow size fraction. An increase in Mn and Mw values, with increasing wavelength for all three humic substances (HS) examined was observed. The HS, isolated from the dam water, was found to be about the same molecular weight as the International Humic Acid Standard (IIHSS). For the fulvic acid standard, the molecular weight was estimated to be around 7500 Da. Characterization of NOM was done to assist in the identification of the species present in the water. FTIR-ATR was used to as a characterization tool to identify the functional groups in the structure of the humic and fulvic acid respectively present in the Vaal Dam. Analysis of the infrared (IR) spectra indicated that the humic acids of the Vaal dam have phenolic hydroxyl groups, hydroxyl groups, conjugated double bond of aromatic family (C=C), and free carboxyl groups. The isolation method has proved to be applicable and reliable for dam water samples and showed to successfully separate the humic substances from water and further separate the humic substances into its hydrophobic acids, namely, humic and fulvic acids. It can be concluded that the Eskom Vaal dam composes of humic substance which shows that the technique alone gives a very good indication of the characteristics of water. The HPSEC method used, equipped with UV and ELSD was able to identify the molecular weight range of NOM present in source water as it confirmed that the Eskom Vaal dam contains humic substances as humic acid and fulvic acid and these pose a health concern as they can form disinfectant byproducts in the course of water treatment with chemicals. FTIR characterization was successful as important functional groups were clearly assigned. Lastly, the use of the TOC and DOC values to calculate SUVA was also a good tool to indicate the organic content in water. It is recommended to use larger amounts of water must be processed to obtain useful quantities of the humic and fulvic acid fractions.

Water quality from surface sources is fast deteriorating due to pollution from organic compounds. Among the organic compounds are chlorophenols, which are described as priority pollutants because of their detrimental effects. One way of removing them from water is by using membranes. However direct removal of chlorophenols using membranes is limited due to the inherent problem of membrane fouling. The thesis describes fabrication of thin film composite membranes modified with Ag-TiO2 and Ag-ZnO for enhancing filtration properties of the membranes for removal of 2-CP and 2,4-DCP and improving the antifouling properties of the modified membranes. Chlorophenols, 2- CP, 2,4-DCP and 2, 4, 6-TCP were determined from Vaal and Klip River using SPE- HLPC method. The SPE - HPLC method was validated by determining breakthrough volume, repeatability, reproducibility, linearity, MDL and LOQ. Nanoparticles (NPs), Ag, ZnO and TiO2 and nanocomposites (NCs), Ag-TiO2 and Ag-ZnO were synthesized using precipitation method and chemical reduction for Ag. The NPs and NCs were characterised using UV-Vis, FTIR, XRD, SEM and EDX. The synthesised NPs and NCS were evaluated for photocatalytic degradation of 2-CP and 2,4-DCP, antimicrobial activity against E.coli. and toxicity against Daphnia magna. Nanocomposites were then embedded into the PA thin film membrane surface using interfacial polymerisation and PES as a support material to produce the antifouling Ag-TiO2/PA-TFC and Ag-ZnO/PA-TFC membranes. The control PATFC membrane was prepared with no added NCs to the membrane. The membranes were characterised using ATR-FTIR, contact angle, SEM and AFM. The performance of the membranes was tested using permeation flux (using pure water and 2-CP / 2,4-DCP solutions as feed) against the neat PA-TFC membrane. Membranes were further tested for rejection of 2- CP and 2, 4 – DCP, antifouling properties and flux recoveries. The stability of the antifouling properties of the membrane was evaluated through silver release test. The performance of the membranes was tested using real water samples from Vaal and Klip Rivers. The SPE-HPLC method was repeatable, reproducible with % RSD less than 5%. Linearity range of (0.1-50 µg/ L) and recoveries of spiked water samples of more than 97% for 2-CP and 2,4-DCP but lower at 64 and 75% for 2.4.6-TCP were achieved. The Ag, TiO2 and ZnO NPs showed characteristic peaks of NPs with UV-Vis. The absorption peaks were all blue shifted due to quantum confinements. The crystalline structures were confirmed as face centred cubic, anatase and hexagonal wurzite for Ag, TiO2 and ZnO respectively. The morphology as observed from SEM showed spherically shaped nanoparticles with average sizes of 68.25 ± 4.7 and 50.92 ± 3.39 nm for Ag and TiO2 respectively. The ZnO NPs were rod -like shaped with average length = 603 nm ± 50.4 and a width = 82.92 ± 5. 40nm. Successful incorporation of silver into the TiO2 and ZnO structures was confirmed by elemental analysis, EDX. From SEM images, silver particles were distributed around TiO2 particles and ZnO rods. The presence of silver showed a remarkable improvement in photodegradation of 2-CP and 2,4-DCP from less than 40% to 86% with 2, 4- DCP. Silver modified TiO2 and ZnO showed antibacterial activity against E.coli. with minimum concentration of inhibition as low as 1.56 mg/L for both Ag-ZnO (5) and Ag-TiO2 (5). Silver was more toxic against Daphnia magna than Ag-ZnO (5) and AgTiO2 (5). The polyamide layer was confirmed by the presence of the amide I peak at 1650 cm1 and 1670 cm-1 in the Ag-TiO2/ PA-TFC and Ag-ZnO/ PA-TFC membranes. The appearance of NCs particles spread across the surface of the thin layer of the membranes as observed from surface SEM images confirming their incorporation into the PA layer. The presence of the NCs in the membranes improved water flux, water permeation, rejection of 2- CP, and 2,4-DCP, antifouling properties of the membranes and flux recoveries of more than 93 % was achieved. Silver release test revealed that Ag-ZnO/PA-TFC membrane performed better than AgTiO2/PA-TFC membrane because of the steady release of silver, which shows long lasting antifouling properties. When applied to real water samples from Vaal and Klip River, the prepared membranes showed better antifouling properties than the neat PA-TFC membrane

PUMPS mimics natural geothermal vents as a conceptual model designed for the remediation of acid mine water (AMW) in voids of abandoned gold mines of the Witwatersrand Basin in South Africa. In this system, a reaction chamber containing Desulfotomaculum kuznetsovii sulfate reducing bacteria will be set at the bottom of a 3-4 km deep mine that will be flooded. A geothermal system with at least one (1) doublet will be drilled from the bottom of the mine to the depth of 8km, where the temperatures are sufficient for geothermal energy harvesting. AMW, used as a geothermal fluid, will be pumped down the injection well and circulate through hot rock. The hot water is then used to generate electricity and then channelled into the reaction chamber to undergo bio-remediation. Following treatment, the water flows back into the mine voids where it will improve the quality of untreated AMW through dilution. Eventually, the mine will be flooded with clean water that can be stored underground and/or pumped up to surface for social and ecosystem services. Following an introduction and proof of concept for the PUMPS, the research builds further on the technical and economic evaluation of the PUMPS in order to assert its viability and sustainability. The technical viability includes testing the ability for Desulfotomaculum kuznetsovii to survive in high pressure condition; quantifying the amount of energy that can be drawn from the geothermal reservoir; determining the placement and scheme of the geothermal wells; and, finally, developing a robust economic model of the system. Experiments show that Desulfotomaculum Kuznetsovii can tolerate high pressure conditions in of at least 100bar at their ideal sulfate reducing temperature of 63°C. Geochemical modelling shows that AMW can be used effectively as a geothermal fluid for PUMPS. To achieve highest efficiency and minimal fluid loss, the geothermal wells should be placed along the SSE-NNW direction, based on the known stress field across the Witwatersrand Basin. With a flow rate of 30l/s the energy drawn from the geothermal reservoir is sufficient to drive PUMPS and the surplus energy is determined by the volume of AMW treated per day. All results indicate that the PUMPS is technically and economically viable. The economic model shows that the value and viability of the PUMPS is best reflected with a comprehensive inclusion of potential revenue (for example from chemical solution mining of deep seated gold) and financial/environmental incentives.

Submitted in fulfillment of the requirements for the degree of Master of Engineering, Department of Industrial Engineering, Durban University of Technology, Durban, South Africa, 2017. / Access to clean and adequate water is a universal and basic human right that feeds into the 6th of the 17 Sustainable Development Goals (SDGs). This goal aims at ensuring availability and sustainable management of water and sanitation for all. Clean water is referred to as potable water, which is safe for human consumption and offer low risk of immediate or long term harm. Raw water undergoes rigorous processing which consists of coagulation, sedimentation, filtration, disinfection and storage, to produce potable water. Each module or stage consumes chemicals and energy resources and thus incurs costs. To achieve the aim of the study, which was to synthesize an optimised potable water treatment network and a water usage analysis model, the Umzinto Water Treatment Plant (UWTP) and its distribution system was used as the study area. This treatment plant is located within Umdoni, a local municipality of the Ugu District Municipality in KwaZulu-Natal Province, South Africa. This study’s objectives were fourfold and the first objective was to identify and quantify key raw water quality parameters affecting treatment at the UWTP. The second objective was to design a genetic algorithm for the potable water treatment process control. The third objective was to evaluate the Umzinto Water Distribution System’s Non-Revenue Water (NRW) while the fourth objective was to develop a model for water usage analysis. For the first objective, data for water quality parameters for the water treatment from July 2006 to June 2013 were statistically analysed. This data were collected from the UWTP’s historical records. To improve the data’s integrity it was pre-processed using cubic hermite interpolation. After the pre-processing trend lines and box plots were used to determine the parameters’ significance compared to the standard values stipulated in the South African National Standard (SANS 241). The trend lines were used to analyse the frequency of observations that were higher than the standard values according to SANS 241. The box plots were used to determine the minimum, median, maximum and mean of the data sets. The mean values for each parameter were compared to the SANS 241 value to determine their significance. The raw water quality parameters were then correlated to the chemical dosages for lime, polymer, potassium permanganate and chlorine. The key parameters selected from the correlation analysis were algal count, manganese (Mn), iron (Fe), Escherichia coli, total coliforms, colour, odour, conductivity, turbidity, suspended solids (SS), pH, temperature, total organic carbon (TOC,) and Hardness. A number of methods can be used to achieve such optimisation, including artificial neural networks, dynamic programming, linear and non-linear programming, and this study utilised a genetic algorithm as an optimisation tool to achieve the second objective of optimising water treatment at the UWTP. For the model development, data from the correlations obtained for objective 1 were used. The model was aimed at reducing the cost of chemical dosage and four chemical dosage prediction models were developed using genetic algorithms and these were then used to produce a combined chemical dosage cost prediction model. The programming interface utilised for these models was Matlab. In developing these models, the data were first pre-processed to remove outliers and fill in the blanks using a Microsoft Excel Add-in that was developed for this particular purpose. The next step involved a curve fitting exercise in Microsoft Excel 2013. Matlab was then used to code the genetic algorithm that combined and optimised the solutions obtained from the curve fittings. The results showed that genetic algorithms can be reliably used to predict the chemical dosages and hence reduce water treatment costs. After treatment, water is pumped into the distribution system for consumption. It is therefore important to ensure that all the pumped out treated water reaches the consumer. The third objective therefore assessed the NRW for the Umzinto Water Distribution System for the period between July 2013 and June 2014. The data used for this objective was provided by the Ugu District Municipality. The method used combined the top-down approach and the component-based approach. This combined approach was modified to enable the calculation of all the components that are required in a standard South African Water Balance. The results showed that the distribution system had a high value of NRW, which was 27.9% of the System Input Volume. The major component of NRW was Real Losses, that is, losses that can be mitigated by improving maintenance. The fourth objective was to develop a model for water usage analysis that would reduce the time taken to evaluate NRW and also improve the analysis of the NRW components using Microsoft Visual Basics 2012 and Microsoft SQL Server 2012 development interfaces. The Visual Basics enabled the development of a graphic user interface that was user-friendly and minimised the time taken to learn the software. The software platform developed was able to import the data required to construct a standard International Water Asssociation (IWA) Water Balance, calculate all the components of NRW, store historical data for the water distribution systems and report on a rolling year basis. A model for water usage analysis was developed and made available for usage by practitioners in Ugu District. The model was developed for the specific study area and further studies would be required in order to validate it in a different setting. The results obtained for the first objective led to the conclusion that, there was very high pollution emanating from communities and activities close to the raw water sources, especially the EJ Smith Dam. The results from the first objective were also used to determine parameters for the models developed in the second objective. From objective two it was concluded that genetic algorithms can be reliably used to predict chemical dosages and hence reduce water treatment costs. The third objective’s results showed that 27.9% of treated water pumped into the distribution system is NRW. Which is a concern because 65% of this are real losses which have maintenance related problems. The fourth objective’s results showed the practicality of designing model that could be used determine all the important components of NRW that would take time to evaluate manually. It would also store historical data for the water distribution system and report on a rolling year basis. Implementation of this software would help minimise the errors associated with manual calculation of NRW and improve the availability of data for research and analysis. From the research findings, it is recommended that the treatment plant should change the way it is dosing chemicals in the balancing tank. The method currently being used is prone to error. The analysis of NRW showed that Real Losses were a major challenge in the Umzinto Distribution System. There is need to develop a maintenance program to cater for leakage. Communities also need to be educated on the importance of reporting leakage in the network. / M

Thesis (M.Sc. (Microbiology)) -- University of Limpopo, 2016 / Access to safe drinking water is a major problem globally and it mostly affects people living in low-income countries. The lack of potable water leads to the use of raw water from surface or ground water sources for drinking and other household purposes. A water filtration unit was designed and constructed using fabric, gravel and sand, which were wet-packaged into a 20 L bucket. The efficiency of the filter unit to improve the bio-physicochemical properties of contaminated water was tested using surface waters from rivers in the Sekhukhune area. Physico-chemical parameters tested included turbidity, colour, total suspended solids (TSS), total dissolved solids (TDS), total hardness and pH. Turbidity and colour were the most improved characteristics, where turbidity improved by 69% and colour by 80%. Other parameters such as total hardness, TSS and TDS were non-significantly reduced following treatment with the filtration unit. The amount of soluble solids in raw water was well within allowable limits by WHO standards. Microbiological tests included heterotrophic bacteria, total coliform and faecal coliform counts. The bacterial load was too numerous to count for the untreated water, however, after treatment with the filtration unit, heterotrophic bacterial load decreased to 15 x 10103 CFU/ml, total coliforms to 14 x 10 2 CFU/100ml and faecal coliforms to 11 x 10 2 CFU/100ml. Further treatment with UV sterilization unit resulted in undetectable amount of bacteria. The unit designed in this study can be beneficial in those rural communities were clean water is not available, it is easy to construct and simple to operate and most importantly it reduced contaminants in surface water. The UV unit incorporated in this study is not cost effective, therefore, other household treatment options such as chlorination and boiling which are easily accessible to most communities can be used to further eliminate remaining microorganisms after filtration. The effective shortest boiling time and minimum dosage of Na(OCl2) have been previously tested in our laboratory and total elimination of bacteria was achieved within 2 minutes of rolling boil and after 30 minutes following addition of 5ml/20L of liquid chlorine.

This study is a component of a research project on the economic costs of eutrophication in the Vaal River system. Its objective is to investigate the relationship between raw water quality and the chemical costs of producing potable water at two water treatment plants: Zuikerbosch Station #2 (owned by Rand Water) in the Upper Vaal Water Management Area (UVWMA), and Balkfontein (owned by Sedibeng Water) in the Middle Vaal Water Management Area (MVWMA). Time series data on raw water quality and chemical dosages used to treat raw water were obtained for Zuikerbosch Station #2 (hereafter referred to as Zuikerbosch) for the period November 2004 – October 2006 and for Balkfontein for the period January 2004 to December 2006. Descriptive statistics reveal that raw water in the Vaal River is of a poorer quality at Balkfontein compared to that at Zuikerbosch. Furthermore, the actual real chemical water treatment costs (measured in 2006 ZAR) averaged R89.90 per megalitre at Zuikerbosch and R126.31 at Balkfontein, indicating that the chemical water treatment costs of producing potable water tend to increase as raw water quality declines. Collinearity among water quality (WQ) variables at both water treatment plants was analysed using Principal Component Analysis (PCA). The dimensions of water quality identified in the analysis are similar to those reported in Pieterse and van Vuuren’s (1997) study of the Vaal River. For both water treatment plants, Ordinary Least Squares (OLS) regression was used to identify the relationship between real chemical costs of water treatment and the dimensions of water quality identified through the respective Principal Components Analyses. The estimated regression models account for over 50.2% and 34.7% of variation in real chemical water treatment costs at Zuikerbosch and Balkfontein, respectively. The coefficient estimated for PC1 at Zuikerbosch is statistically significant at the 1% level of probability with high negative loadings of total alkalinity and turbidity. Increases in the levels of total alkalinity and turbidity in raw water treated at Zuikerbosch is negatively related to the chemical costs of water treatment. An increased total alkalinity level was found to reduce the chemical costs of treating potable water. PC2 is statistically the most important variable in the estimated explanatory model for Balkfontein. The estimated regression coefficient for PC2 is statistically significant at the 5% level of probability. The estimated relationship between chemical water treatment costs and PC2 shows that there is a positive relationship between the raw water temperature and chemical water treatment costs. However, increases in the levels of chlorophyll and pH in raw water treated at Balkfontein is negatively related to the chemical costs of water treatment. Total hardness, magnesium, calcium, sulphate, conductivity, and chloride, being the highest positive loadings in PC1, relate negatively to the chemical cost of treating water. For predictive rather than explanatory purposes, a partial adjustment regression model was estimated for each of the two water treatment plants. Using this model, real chemical water treatment costs were specified as a function of real chemical water treatment costs in the previous time period, and of raw water quality variables in the current period. The R2 statistics for the two regression models were 61.4% using the data for Zuikerbosch and 59.9% using the data for Balkfontein, suggesting that both models have reasonable levels of predictive power. The chemical cost of water treatment for Zuikerbosch and Balkfontein are predicted at R96.25 and R90.74 per megalitre per day respectively. If raw water nitrate in the UVWMA increases by 1% per megalitre a day while other factors remain constant, chemical water treatment costs at Zuikerbosch can be expected to increase by 0.297% per megalitre and the cost accompanied this change is (R0.285*1998ML*365days) R207,841.95 provided that Zuikerbosch treats an average of 1998 megalitres per day. Likewise, if Zuikerbosch maintains its daily average operating capacity and is able to maintain an optimal level of total alkalinity in UVWMA, the estimated saving on chemical water treatment cost will be R150.063.78 per annum. At Balkfontein, chemical water treatment cost is expected to increase on average by 0.346% per megalitre per day for a 1% per megalitre per day increase in the level of chlorophyll-a, and the cost accompanied this change is R41,128.20 per annum. The prediction also shows a 2.077% per megalitre per day increase chemical water treatment cost for a 1% increase in turbidity and this accompanied with a chemical water treatment cost of R 249,003 per annum, provided that Balkfontein operates at its full capacity (i.e., 360 megalitres per day). / Thesis (M.Sc.Agric.)-University of KwaZulu-Natal, Pietermaritzburg, 2009.