The freshwater dinoflagellate species, Ceratium hirundinella (C. hirundinella) possesses unique characteristics, such as a thecal-plate cell covering of cellulose, spines and flagella. Unlike most other algae and cyanobacteria, C. hirundinella cells are relatively large in size (up to 450 μm in length and 50 μm in width). These unique characteristics (e.g. cell covering and flagella) and adaptations (e.g. spines) give the dinoflagellate cells the ability to reduce their sinking rate from the euphotic zone and to migrate easily through the water column. When source water contains high concentrations of C. hirundinella cells, water treatment problems and poor aesthetic water quality can be expected. These water treatment problems may include 1) the disruption of coagulation and flocculation, 2) clogging of sand filters and 3) taste and odour problems when cells penetrate into the final water. In Chapter 9 of this study, a list of operational guidelines (including alert levels) and recommendations to assist managers and operators of plants when C. hirundinella cells are causing water treatment problems. During events of high C. hirundinella concentrations in source water, managers and operators of conventional water treatment plants need strategies to optimize coagulants and unit processes. Thus when source water contains motile nuisance algae, such as C. hirundinella, in moderate or abundant quantities, it is advisable to conduct jar stirring test experiments using both turbidity and total photosynthetic pigment (or chlorophyll-a) analyses as indicators of appropriate coagulant choice and dosages. The aims of this study are summarized as follows:  To optimize coagulants and conventional water treatment processes by implementing relevant algal removal strategies and indicators during jar stirring test experiments,  To investigate the changes in surface charge (known as zeta potential) on C. hirundinella cells before and after adding coagulants as part of the treatment processes,  To investigate the physical and chemical impacts on the morphology of C. hirundinella cells after coagulation, flocculation and sedimentation,  To identify organic compounds that may be responsible for taste and odour problems associated with C. hirundinella,  To investigate the efficiency of pre-chlorination on the removal C. hirundinella cells when dosing various coagulants, and  Give recommendations and operational guidelines relevant for a conventional water treatment plant to improve C. hirundinella removal A combined water treatment system (Phipps and Bird Model), consisting of a six paddle jar test apparatus and six sand filter columns, was used to simulate conventional processes (coagulation, flocculation, sedimentation and rapid sand filtration). Source water samples containing relatively high C. hirundinella concentrations (> 500 cell/mℓ) were collected from Benoni Lake (26º10’50.40’’S; 28º17’50.11’’ E) in plastic containers and stored as a homogenous sample in a 200 litre container under laboratory conditions (± 22 °C). Samples were collected from the source water as well as after sedimentation (from the supernatant or sludge) to determine turbidity, total photosynthetic pigment analyses (chlorophyll) and for phytoplankton analyses. Flocs (containing C. hirundinella cells) were collected from the sludge or sediment for scanning electron microscopy investigations and to perform zeta potential analyses. Concentrated C. hirundinella samples were frozen at -80 °C according to the proposed sampling protocol for organic compound analyses. Results obtained from this study proved that using the relevant indicators to determine the appropriate coagulant dosages during jar stirring tests may generally improve the removal of problem-causing algae, such as C. hirundinella cells. Improved algal removal efficiencies will subsequently ensure final water with good aesthetic quality. The surface charge (zeta potential) on C. hirundinella cells can be used to evaluate the best coagulation conditions within an operating window of -10 mV to +3 mV when dosing various coagulants. Scanning electron microscopy investigations revealed major damaging effects to C. hirundinella cells when dosing high Ca(OH)2 concentrations. However, when dosing lower Ca(OH)2 concentrations, in combination with organic polymer, better C. hirundinella cell removal efficiencies with less damaging effects to cells was observed. This study also indicated that the pre-chlorination, without causing cell lyses, can be applied to render the highly motile cells immobile which will subsequently assist the coagulation unit process. The aesthetic quality (e.g. tastes and odours) of drinking water may be influenced when C. hirundinella cells release organic material into the water as a result of cell lyses. Organic compounds, such as fatty acids and dicarboxylic acids can lead to taste and odour problems which associate with the presence of C. hirundinella. Organic compounds also serve as precursors for the formation of harmful chlorine by-products formed during chlorination. / PhD (Environmental Sciences), North-West University, Potchefstroom Campus, 2015

The freshwater dinoflagellate species, Ceratium hirundinella (C. hirundinella) possesses unique characteristics, such as a thecal-plate cell covering of cellulose, spines and flagella. Unlike most other algae and cyanobacteria, C. hirundinella cells are relatively large in size (up to 450 μm in length and 50 μm in width). These unique characteristics (e.g. cell covering and flagella) and adaptations (e.g. spines) give the dinoflagellate cells the ability to reduce their sinking rate from the euphotic zone and to migrate easily through the water column. When source water contains high concentrations of C. hirundinella cells, water treatment problems and poor aesthetic water quality can be expected. These water treatment problems may include 1) the disruption of coagulation and flocculation, 2) clogging of sand filters and 3) taste and odour problems when cells penetrate into the final water. In Chapter 9 of this study, a list of operational guidelines (including alert levels) and recommendations to assist managers and operators of plants when C. hirundinella cells are causing water treatment problems. During events of high C. hirundinella concentrations in source water, managers and operators of conventional water treatment plants need strategies to optimize coagulants and unit processes. Thus when source water contains motile nuisance algae, such as C. hirundinella, in moderate or abundant quantities, it is advisable to conduct jar stirring test experiments using both turbidity and total photosynthetic pigment (or chlorophyll-a) analyses as indicators of appropriate coagulant choice and dosages. The aims of this study are summarized as follows:  To optimize coagulants and conventional water treatment processes by implementing relevant algal removal strategies and indicators during jar stirring test experiments,  To investigate the changes in surface charge (known as zeta potential) on C. hirundinella cells before and after adding coagulants as part of the treatment processes,  To investigate the physical and chemical impacts on the morphology of C. hirundinella cells after coagulation, flocculation and sedimentation,  To identify organic compounds that may be responsible for taste and odour problems associated with C. hirundinella,  To investigate the efficiency of pre-chlorination on the removal C. hirundinella cells when dosing various coagulants, and  Give recommendations and operational guidelines relevant for a conventional water treatment plant to improve C. hirundinella removal A combined water treatment system (Phipps and Bird Model), consisting of a six paddle jar test apparatus and six sand filter columns, was used to simulate conventional processes (coagulation, flocculation, sedimentation and rapid sand filtration). Source water samples containing relatively high C. hirundinella concentrations (> 500 cell/mℓ) were collected from Benoni Lake (26º10’50.40’’S; 28º17’50.11’’ E) in plastic containers and stored as a homogenous sample in a 200 litre container under laboratory conditions (± 22 °C). Samples were collected from the source water as well as after sedimentation (from the supernatant or sludge) to determine turbidity, total photosynthetic pigment analyses (chlorophyll) and for phytoplankton analyses. Flocs (containing C. hirundinella cells) were collected from the sludge or sediment for scanning electron microscopy investigations and to perform zeta potential analyses. Concentrated C. hirundinella samples were frozen at -80 °C according to the proposed sampling protocol for organic compound analyses. Results obtained from this study proved that using the relevant indicators to determine the appropriate coagulant dosages during jar stirring tests may generally improve the removal of problem-causing algae, such as C. hirundinella cells. Improved algal removal efficiencies will subsequently ensure final water with good aesthetic quality. The surface charge (zeta potential) on C. hirundinella cells can be used to evaluate the best coagulation conditions within an operating window of -10 mV to +3 mV when dosing various coagulants. Scanning electron microscopy investigations revealed major damaging effects to C. hirundinella cells when dosing high Ca(OH)2 concentrations. However, when dosing lower Ca(OH)2 concentrations, in combination with organic polymer, better C. hirundinella cell removal efficiencies with less damaging effects to cells was observed. This study also indicated that the pre-chlorination, without causing cell lyses, can be applied to render the highly motile cells immobile which will subsequently assist the coagulation unit process. The aesthetic quality (e.g. tastes and odours) of drinking water may be influenced when C. hirundinella cells release organic material into the water as a result of cell lyses. Organic compounds, such as fatty acids and dicarboxylic acids can lead to taste and odour problems which associate with the presence of C. hirundinella. Organic compounds also serve as precursors for the formation of harmful chlorine by-products formed during chlorination. / PhD (Environmental Sciences), North-West University, Potchefstroom Campus, 2015

The Ceratium species occurring in the Vaal River since 2000, was identified as Ceratium hirundinella (O.F. Müller) Dujardin as proposed by Van Ginkel et al (2001). Ceratium hirundinella is known to cause problems in drinking water purification and has been penetrating into the final drinking water of Rand Water since 2006. Ceratium hirundinella is associated with many other water purification problems such as disrupting of the coagulation and flocculation processes, blocking of sand filters and algal penetration into the drinking water. Ceratium hirundinella also produce fishy taste and odorous compounds and causes discolouration of the water. The aims of this study were to determine the main environmental factors which are associated with the bloom formation of C. hirundinella in the source water and to investigate the influence of C. hirundinella on the production of potable water. In order to optimise treatment processes and resolve problems associated with high C. hirundinella concentrations during the production of potable water, jar testing and chlorine exposure experiments were performed. Multivariate statistical analyses were performed to determine the main environmental variables behind C. hirundinella blooms. Ten years data (2000 - 2009) from the sampling point C–VRB5T in the Vaal River, (5 km upstream from the Barrage weir) were used for this investigation, because C. hirundinella occurred there frequently during the ten year period. In this study, it was found that C. hirundinella was favoured by high pH, Chemical Oxygen Demand (COD), orthophoshapte (PO4), and silica concentrations, as well as low turbidity and low dissolved inorganic nitrogen (DIN) concentrations. No correlation was found between C. hirundinella and temperature, suggesting that this alga does not occur during periods of extreme warm or extreme cold conditions, but most probably during autumn and spring. The results of the multivariate statistical analysis performed with historical data from Vaalkop dam, indicate that the dinoflagellate C. hirundinella seems to be favoured by low temperature and turbidity, and high DIN, Fe, Methyl–orange alkalinity, Cd, PO4, Conductivity, pH, hardness and SO4 concentrations. In order to optimise treatment processes such as coagulation, flocculation and sedimentation, jar testing experiments were performed to investigate different coagulant chemicals namely: cationic poly–electrolyte only, cationic poly–electrolyte in combination with slaked lime (CaO) and CaO in combination with activated silica. Water from four different sampling localities were chosen to perform the different jar testing experiments: 1) sampling point M–FOREBAY (in the Forebay, connecting the canal to the Zuikerbosch Purification plant) near Vereeniging due to its proximity to the Zuikerbosch treatment plant, 2) M–CANAL_VD (upstream from the inflow of the recovered water from Panfontein) to determine the influence of (if any) the recovered water from Panfontein on Forebay source water, 3) source water from Vaalkop Dam (M–RAW_VAALKOP) and 4) source water from Rietvlei Dam (water from both Vaalkop and Rietvlei Dams contained high concentrations of C. hirundinella at that time of sampling) to determine which coagulant chemical is the most effective in removing high concentrations of C. hirundinella cells during the production of drinking water. The jar testing experiments with Vaalkop Dam and Rietvlei Dam source water (rich with C. hirundinella) indicated that using cationic poly–electrolyte alone did not remove high concentrations of C. hirundinella efficiently. However, when CaO (in combination with cationic poly–electrolyte or activated silica) were dosed to Vaalkop Dam source water a significant decrease of C. hirundinella concentration was observed. This indicates that the C. hirundinella cells were “shocked or stressed” when exposed to the high pH of the CaO, rendering it immobile and thereby enhancing the coagulation and flocculation process. However, when 10 mg/L CaO in combination with poly–electrolyte was dosed to Rietvlei Dam source water the turbidity and chlorophyll–665 results indicated that this coagulant chemical procedure was ineffective in removing algal material from the source water. The jar testing experiments using the cationic poly–electrolyte alone or cationic poly–electrolyte in combination with CaO on M–FOREBAY and M–CANAL_VD source water, showed a decrease in turbidity, chlorophyll–665 concentration, and total algal biomass, with an increase of coagulant chemical. When CaO in combination with activated silica was dosed, the inherent turbidity of the lime increased the turbidity of the Vaalkop Dam, M–FOREBAY and M–CANAL_VD source water to such an extent that it affected coagulation negatively, resulting in high turbidity values in the supernatant. Regardless of the turbidity values, the chlorophyll–665 concentration and total algal biomass (C. hirundinella specifically in Vaalkop Dam source water) decreased significantly when CaO was dosed in combination with activated silica. Therefore it was concluded that a cationic poly–electrolyte alone is a good coagulant chemical for the removal of turbidity, but when high algal biomass occur in the source water it is essential to add CaO to “stress” or “shock” the algae for the effective removal thereof. However, when CaO in combination with activated silica was dosed to Rietvlei Dam source water a decrease in turbidity and chlorophyll–665 concentration was found with an increasing coagulant chemical concentration. These results confirm the fact that coagulant chemicals may perform differently during different periods of the year when water chemistry changes and that certain coagulant chemicals may never be suitable to use for certain source waters. For the effective removal of algae during water purification, it is recommended that cationic poly–electrolyte in combination with CaO are used as coagulant chemical at the Zuikerbosch Water Purification Plant. Turbidity is not a good indication of algal removal efficiency during jar testing experiments. If problems with high algal concentrations in the source water are experienced it is advisable to also determine the chlorophyll–665 concentrations of the supernatant water during the regular jar testing experiments, since it will give a better indication of algal removal. Chlorine exposure experiments were performed on water from Vaalkop Dam (M–RAW_VAALKOP) and Rietvlei Dam source water, to determine the possibility of implementing pre– or intermediate chlorination with the aim to render the cells immobile for more effective coagulation. The chlorine exposure experiments with Vaalkop Dam and Rietvlei Dam source water showed similar results. The chlorine concentration to be dosed as part of pre– or intermediate chlorination will differ for each type of source water as the chemical and biological composition of each water body are unique. When the effect of chlorine on the freshwater dinoflagellate C. hirundinella was investigated, it was found that the effective chlorine concentration where 50 % of Ceratium cells were rendered immobile (EC50) was approximately 1.16 mg/L for Vaalkop Dam source water. For the source water sampled from Rietvlei Dam, it was found that the EC50 was at approximately 0.87 mg/L. Results of analyses to determine the organic compounds in the water after chlorination revealed that an increase in chlorine concentration resulted in increase in total organic carbon concentration (TOC), as well as a slight increase in MIB and trihalomethanes (CHCl3). Pre– or intermediate chlorination seem to be an effective treatment option for the dinoflagellate C. hirundinella to be rendered immobile and thereby assisting in its coagulation process. The use of pre– or intermediate chlorination to effectively treat source waters containing high concentrations of C. hirundinella is a viable option to consider. However, the organic compounds in the water should be monitored and the EC50 value for each source water composition should be determined carefully as to restrict cell lysis and subsequent release of organic compounds into the water. / Thesis (M.Sc. (Environmental Science))--North-West University, Potchefstroom Campus, 2012.

The Ceratium species occurring in the Vaal River since 2000, was identified as Ceratium hirundinella (O.F. Müller) Dujardin as proposed by Van Ginkel et al (2001). Ceratium hirundinella is known to cause problems in drinking water purification and has been penetrating into the final drinking water of Rand Water since 2006. Ceratium hirundinella is associated with many other water purification problems such as disrupting of the coagulation and flocculation processes, blocking of sand filters and algal penetration into the drinking water. Ceratium hirundinella also produce fishy taste and odorous compounds and causes discolouration of the water. The aims of this study were to determine the main environmental factors which are associated with the bloom formation of C. hirundinella in the source water and to investigate the influence of C. hirundinella on the production of potable water. In order to optimise treatment processes and resolve problems associated with high C. hirundinella concentrations during the production of potable water, jar testing and chlorine exposure experiments were performed. Multivariate statistical analyses were performed to determine the main environmental variables behind C. hirundinella blooms. Ten years data (2000 - 2009) from the sampling point C–VRB5T in the Vaal River, (5 km upstream from the Barrage weir) were used for this investigation, because C. hirundinella occurred there frequently during the ten year period. In this study, it was found that C. hirundinella was favoured by high pH, Chemical Oxygen Demand (COD), orthophoshapte (PO4), and silica concentrations, as well as low turbidity and low dissolved inorganic nitrogen (DIN) concentrations. No correlation was found between C. hirundinella and temperature, suggesting that this alga does not occur during periods of extreme warm or extreme cold conditions, but most probably during autumn and spring. The results of the multivariate statistical analysis performed with historical data from Vaalkop dam, indicate that the dinoflagellate C. hirundinella seems to be favoured by low temperature and turbidity, and high DIN, Fe, Methyl–orange alkalinity, Cd, PO4, Conductivity, pH, hardness and SO4 concentrations. In order to optimise treatment processes such as coagulation, flocculation and sedimentation, jar testing experiments were performed to investigate different coagulant chemicals namely: cationic poly–electrolyte only, cationic poly–electrolyte in combination with slaked lime (CaO) and CaO in combination with activated silica. Water from four different sampling localities were chosen to perform the different jar testing experiments: 1) sampling point M–FOREBAY (in the Forebay, connecting the canal to the Zuikerbosch Purification plant) near Vereeniging due to its proximity to the Zuikerbosch treatment plant, 2) M–CANAL_VD (upstream from the inflow of the recovered water from Panfontein) to determine the influence of (if any) the recovered water from Panfontein on Forebay source water, 3) source water from Vaalkop Dam (M–RAW_VAALKOP) and 4) source water from Rietvlei Dam (water from both Vaalkop and Rietvlei Dams contained high concentrations of C. hirundinella at that time of sampling) to determine which coagulant chemical is the most effective in removing high concentrations of C. hirundinella cells during the production of drinking water. The jar testing experiments with Vaalkop Dam and Rietvlei Dam source water (rich with C. hirundinella) indicated that using cationic poly–electrolyte alone did not remove high concentrations of C. hirundinella efficiently. However, when CaO (in combination with cationic poly–electrolyte or activated silica) were dosed to Vaalkop Dam source water a significant decrease of C. hirundinella concentration was observed. This indicates that the C. hirundinella cells were “shocked or stressed” when exposed to the high pH of the CaO, rendering it immobile and thereby enhancing the coagulation and flocculation process. However, when 10 mg/L CaO in combination with poly–electrolyte was dosed to Rietvlei Dam source water the turbidity and chlorophyll–665 results indicated that this coagulant chemical procedure was ineffective in removing algal material from the source water. The jar testing experiments using the cationic poly–electrolyte alone or cationic poly–electrolyte in combination with CaO on M–FOREBAY and M–CANAL_VD source water, showed a decrease in turbidity, chlorophyll–665 concentration, and total algal biomass, with an increase of coagulant chemical. When CaO in combination with activated silica was dosed, the inherent turbidity of the lime increased the turbidity of the Vaalkop Dam, M–FOREBAY and M–CANAL_VD source water to such an extent that it affected coagulation negatively, resulting in high turbidity values in the supernatant. Regardless of the turbidity values, the chlorophyll–665 concentration and total algal biomass (C. hirundinella specifically in Vaalkop Dam source water) decreased significantly when CaO was dosed in combination with activated silica. Therefore it was concluded that a cationic poly–electrolyte alone is a good coagulant chemical for the removal of turbidity, but when high algal biomass occur in the source water it is essential to add CaO to “stress” or “shock” the algae for the effective removal thereof. However, when CaO in combination with activated silica was dosed to Rietvlei Dam source water a decrease in turbidity and chlorophyll–665 concentration was found with an increasing coagulant chemical concentration. These results confirm the fact that coagulant chemicals may perform differently during different periods of the year when water chemistry changes and that certain coagulant chemicals may never be suitable to use for certain source waters. For the effective removal of algae during water purification, it is recommended that cationic poly–electrolyte in combination with CaO are used as coagulant chemical at the Zuikerbosch Water Purification Plant. Turbidity is not a good indication of algal removal efficiency during jar testing experiments. If problems with high algal concentrations in the source water are experienced it is advisable to also determine the chlorophyll–665 concentrations of the supernatant water during the regular jar testing experiments, since it will give a better indication of algal removal. Chlorine exposure experiments were performed on water from Vaalkop Dam (M–RAW_VAALKOP) and Rietvlei Dam source water, to determine the possibility of implementing pre– or intermediate chlorination with the aim to render the cells immobile for more effective coagulation. The chlorine exposure experiments with Vaalkop Dam and Rietvlei Dam source water showed similar results. The chlorine concentration to be dosed as part of pre– or intermediate chlorination will differ for each type of source water as the chemical and biological composition of each water body are unique. When the effect of chlorine on the freshwater dinoflagellate C. hirundinella was investigated, it was found that the effective chlorine concentration where 50 % of Ceratium cells were rendered immobile (EC50) was approximately 1.16 mg/L for Vaalkop Dam source water. For the source water sampled from Rietvlei Dam, it was found that the EC50 was at approximately 0.87 mg/L. Results of analyses to determine the organic compounds in the water after chlorination revealed that an increase in chlorine concentration resulted in increase in total organic carbon concentration (TOC), as well as a slight increase in MIB and trihalomethanes (CHCl3). Pre– or intermediate chlorination seem to be an effective treatment option for the dinoflagellate C. hirundinella to be rendered immobile and thereby assisting in its coagulation process. The use of pre– or intermediate chlorination to effectively treat source waters containing high concentrations of C. hirundinella is a viable option to consider. However, the organic compounds in the water should be monitored and the EC50 value for each source water composition should be determined carefully as to restrict cell lysis and subsequent release of organic compounds into the water. / Thesis (M.Sc. (Environmental Science))--North-West University, Potchefstroom Campus, 2012.