A scheduling model for the charging of electric vehicles in photovoltaic powered smart microgrids
- Authors: Nyumbeka, Dumisani Joshua
- Date: 2020
- Subjects: Electric vehicles , Photovoltaic power generation Photovoltaic power generation -- Developing countries
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/49259 , vital:41615
- Description: Electric vehicles (EVs) have emerged as a viable option to advance sustainable mobility, but adoption is still relatively low. This has been largely due to the limited range one can travel on a single charge, leading to range anxiety, longer charge cycles and long wait times at charging stations. One solution to range anxiety is to erect charging stations on major roads and urban centres. There is also a lack of real-time information regarding the state of charging stations and charging ports in existing charging infrastructure. To increase the benefit of using EVs, using renewable energy sources, such as photovoltaics (PV) to power EVs, can further increase the benefit of reduced carbon footprint. The main research objective was to design a Charge Scheduling Model for charging EVs using a PV-powered smart microgrid (SMG). The model addresses the lack of an integrated platform where EV drivers can schedule when and where to charge their EVs. The model also reduces the negative effects of the adoption of EVs, including range anxiety. The Charge Scheduling Model was developed using the Design Science Research (DSR) methodology and was the main artefact of the study. A literature study was conducted of research related to SMGs, renewable energy, EVs and scheduling, to identify shortcomings that currently exist in EV charge scheduling (EVCS), and to identify the requirements of a potential solution. The literature study also identified the hard and soft constraints that are unique to EVCS, and the available energy in the SMG was identified as one of the hard constraints. Therefore, an Energy Forecasting Model for forecasting energy generated in PV-powered SMGs was required before the Charge Scheduling Model could be designed. During the first iteration of the design and development activities of DSR, four models were designed and implemented to evaluate their effectiveness in forecasting the energy generated in PV-powered SMGs. The models were Support Vector Regression (SVR), K-Nearest Neighbour (KNN), Decision Trees, and Multilayer Perceptron. In the second iteration, the Charge Scheduling Model was designed, consisting of a Four Layered Architecture and the Three-Phase Data Flow Process. The Charge Scheduling Model was then used to design the EVCS prototype. The implementation of the EVCS prototype followed the incremental prototyping approach, which was used to verify the effectiveness of the model. An artificial-summative evaluation was used to evaluate the design of the Charge Scheduling Model, whereas iterative formative evaluations were conducted during the development of the EVCS prototype. The theoretical contribution of this study is the Charge Scheduling Model, and the EVCS prototype is the practical contribution. The results from both evaluations, i.e. the Energy Forecasting Model and the Charge Scheduling Model, also make a contribution to the body of knowledge of EVs.
- Full Text:
- Date Issued: 2020
- Authors: Nyumbeka, Dumisani Joshua
- Date: 2020
- Subjects: Electric vehicles , Photovoltaic power generation Photovoltaic power generation -- Developing countries
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/49259 , vital:41615
- Description: Electric vehicles (EVs) have emerged as a viable option to advance sustainable mobility, but adoption is still relatively low. This has been largely due to the limited range one can travel on a single charge, leading to range anxiety, longer charge cycles and long wait times at charging stations. One solution to range anxiety is to erect charging stations on major roads and urban centres. There is also a lack of real-time information regarding the state of charging stations and charging ports in existing charging infrastructure. To increase the benefit of using EVs, using renewable energy sources, such as photovoltaics (PV) to power EVs, can further increase the benefit of reduced carbon footprint. The main research objective was to design a Charge Scheduling Model for charging EVs using a PV-powered smart microgrid (SMG). The model addresses the lack of an integrated platform where EV drivers can schedule when and where to charge their EVs. The model also reduces the negative effects of the adoption of EVs, including range anxiety. The Charge Scheduling Model was developed using the Design Science Research (DSR) methodology and was the main artefact of the study. A literature study was conducted of research related to SMGs, renewable energy, EVs and scheduling, to identify shortcomings that currently exist in EV charge scheduling (EVCS), and to identify the requirements of a potential solution. The literature study also identified the hard and soft constraints that are unique to EVCS, and the available energy in the SMG was identified as one of the hard constraints. Therefore, an Energy Forecasting Model for forecasting energy generated in PV-powered SMGs was required before the Charge Scheduling Model could be designed. During the first iteration of the design and development activities of DSR, four models were designed and implemented to evaluate their effectiveness in forecasting the energy generated in PV-powered SMGs. The models were Support Vector Regression (SVR), K-Nearest Neighbour (KNN), Decision Trees, and Multilayer Perceptron. In the second iteration, the Charge Scheduling Model was designed, consisting of a Four Layered Architecture and the Three-Phase Data Flow Process. The Charge Scheduling Model was then used to design the EVCS prototype. The implementation of the EVCS prototype followed the incremental prototyping approach, which was used to verify the effectiveness of the model. An artificial-summative evaluation was used to evaluate the design of the Charge Scheduling Model, whereas iterative formative evaluations were conducted during the development of the EVCS prototype. The theoretical contribution of this study is the Charge Scheduling Model, and the EVCS prototype is the practical contribution. The results from both evaluations, i.e. the Energy Forecasting Model and the Charge Scheduling Model, also make a contribution to the body of knowledge of EVs.
- Full Text:
- Date Issued: 2020
A trait-based approach to identify indicators for measuring rocky reef ecosystem condition in South Africa
- Authors: Smit, Kaylee
- Date: 2020
- Subjects: Environmental management -- Evaluation , Environmental monitoring -- Methodology Marine ecology -- South Africa
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/49292 , vital:41618
- Description: An important yet challenging task for effective ecosystem-based management of marine ecosystems is the reliable measurement and monitoring of changes in ecosystem condition. In South Africa, the National Biodiversity Assessment (NBA) has relied on cumulative pressure mapping, based on the calculation of cumulative impact scores, as a proxy for measures of marine ecosystem condition. While this method is practical for broad scale assessments where ecological data are lacking, there is a need to groundtruth this approach at local scales using in-situ data. The aim of this thesis was to identify potential indicators that can be used to measure the ecosystem condition of subtidal rocky reefs on the East coast of South Africa to groundtruth national pressure-based condition assessments. This thesis constituted the first attempt to conduct a comprehensive assessment of rocky reef ecosystem condition to validate national pressure-based assessments in South Africa. Based on a comprehensive literature review, a traits-based approach was selected to ensure that ecosystem structure and function were included in condition measurements. Fish and benthic community data were collected using baited remote underwater stereo-video systems (144 video samples) from the Natal Shelf ecoregion. Sampling spanned a gradient of low to high human pressures, which incorporated an effective Marine Protected Area (to define reference conditions), as well as areas exposed to intermediate levels of human pressures and a highly impacted area offshore of a large metropolitan port city. Structural and functional indicators were compared among localities and suitable functional indicators were aggregated into multimetric indices (MMIs) to provide an integrated measurement of ecosystem condition. Approximately 840 benthic biota and 10 400 fish (representing 176 identified fish species) were counted. Results showed that functional indicators, particularly size-based ones, were more effective at detecting changes in ecosystem condition compared with structural indicators based on taxonomy. Reefs in good ecological condition had a higher biomass of target species and a higher proportion of large predatory fish with more adults, compared with impacted areas. Benthic communities, however, appeared less responsive to cumulative pressures. The MMIs, using fish and benthic community data, were used to compare condition categories with those from the NBA. Results showed that the pressure-based assessments provided a reasonable proxy of broad patterns of rocky reef ecosystem condition. However, fine-scale discrepancies between results from the data-driven approach versus the proxy method suggested that broad-scale proxies might overestimate condition in some areas and underestimate it in others. To build on these results, the applicability of structural and functional indicators and MMIs were tested using an independent stereo-BRUVs dataset from the Agulhas Shelf ecoregion. Findings showed that not all MPAs were in good ecological condition compared with adjacent impacted areas, as was estimated by the NBA. However, there is a need to improve the quality of trait data and to test indicators and indices in other ecoregions, ecosystem types and in response to a broader range of improved pressure data. Furthermore, there is a need to apply innovative research to guide the calculation of thresholds for condition categories to advance condition assessments.
- Full Text:
- Date Issued: 2020
- Authors: Smit, Kaylee
- Date: 2020
- Subjects: Environmental management -- Evaluation , Environmental monitoring -- Methodology Marine ecology -- South Africa
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/49292 , vital:41618
- Description: An important yet challenging task for effective ecosystem-based management of marine ecosystems is the reliable measurement and monitoring of changes in ecosystem condition. In South Africa, the National Biodiversity Assessment (NBA) has relied on cumulative pressure mapping, based on the calculation of cumulative impact scores, as a proxy for measures of marine ecosystem condition. While this method is practical for broad scale assessments where ecological data are lacking, there is a need to groundtruth this approach at local scales using in-situ data. The aim of this thesis was to identify potential indicators that can be used to measure the ecosystem condition of subtidal rocky reefs on the East coast of South Africa to groundtruth national pressure-based condition assessments. This thesis constituted the first attempt to conduct a comprehensive assessment of rocky reef ecosystem condition to validate national pressure-based assessments in South Africa. Based on a comprehensive literature review, a traits-based approach was selected to ensure that ecosystem structure and function were included in condition measurements. Fish and benthic community data were collected using baited remote underwater stereo-video systems (144 video samples) from the Natal Shelf ecoregion. Sampling spanned a gradient of low to high human pressures, which incorporated an effective Marine Protected Area (to define reference conditions), as well as areas exposed to intermediate levels of human pressures and a highly impacted area offshore of a large metropolitan port city. Structural and functional indicators were compared among localities and suitable functional indicators were aggregated into multimetric indices (MMIs) to provide an integrated measurement of ecosystem condition. Approximately 840 benthic biota and 10 400 fish (representing 176 identified fish species) were counted. Results showed that functional indicators, particularly size-based ones, were more effective at detecting changes in ecosystem condition compared with structural indicators based on taxonomy. Reefs in good ecological condition had a higher biomass of target species and a higher proportion of large predatory fish with more adults, compared with impacted areas. Benthic communities, however, appeared less responsive to cumulative pressures. The MMIs, using fish and benthic community data, were used to compare condition categories with those from the NBA. Results showed that the pressure-based assessments provided a reasonable proxy of broad patterns of rocky reef ecosystem condition. However, fine-scale discrepancies between results from the data-driven approach versus the proxy method suggested that broad-scale proxies might overestimate condition in some areas and underestimate it in others. To build on these results, the applicability of structural and functional indicators and MMIs were tested using an independent stereo-BRUVs dataset from the Agulhas Shelf ecoregion. Findings showed that not all MPAs were in good ecological condition compared with adjacent impacted areas, as was estimated by the NBA. However, there is a need to improve the quality of trait data and to test indicators and indices in other ecoregions, ecosystem types and in response to a broader range of improved pressure data. Furthermore, there is a need to apply innovative research to guide the calculation of thresholds for condition categories to advance condition assessments.
- Full Text:
- Date Issued: 2020
Data transport over optical fibre for ska using advanced modulation flexible spectrum technology
- Authors: Dlamini, Phumla Patience
- Date: 2020
- Subjects: Fiber optics
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/50666 , vital:42329
- Description: Flexible Spectrum Dense Wavelength Division Multiplexed (DWDM) optical fibre networks are next-generation technology for handling extremely high data rates of the kind produced by MeerKAT and SKA.We optimise the flexible spectrum for real-time dynamic channel wavelength assignment, to ensure optimum network performance. We needed to identify and develop novel hardware and dynamic algorithms for these networks to function optimally to perform critical tasks. Such tasks include wavelength assignment, signal routing, network restoration and network protection. The antennas of the Square Kilometre Array (SKA) network connect to the correlator and data processor in a simple point-to-point fixed configuration. The connection of the astronomer users to the data processor, however, requires a more complex network architecture. This is because the network has users scattered around South Africa, Africa and the whole world. This calls for upgrade of the classical fixed wavelength spectrum grids, to flexible spectrum grid that has improved capacity, reliable, simple and cost-effectiveness through sharing of network infrastructure. The exponential growth of data traffic in current optical communication networks requires higher capacity for the bandwidth demands at a reduced cost per bit. All-optical signal processing is a promising technique to improve network resource utilisation and resolve wavelength contention associated with the flexible spectrum. Flexible Spectrum Dense Wavelength Division Multiplexed (DWDM) optical fibre networks are next-generation technology for handling extremely high data rates of the kind produced by MeerKAT and SKA. Each DWDM channel is capable of 10 Gbps transmission rate, which is sliceable into finer flexible grid 12.5 GHz granularity to offer the network elastic spectrum and channel spacing capable of signal routing and wavelength switching for the scalability of aggregate bandwidth. The variable-sized portions of the flexible spectrum assignment to end users at different speeds depend on bandwidth demand, allowing efficient utilisation of the spectrum resources. The entire bandwidth of dynamic optical connections must be contiguously allocated. However, there is an introduction of spectrum fragmentation due to spectrum contiguity related to the optical channels having different width. Thus large traffic demands are likely to experience blocking regardless of available bandwidth. To minimise the congestion and cost-effectively obtain high performance, the optical network must be reconfigurable, achievable by adding wavelength as an extra degree of freedom for effectiveness. This can introduce colourless, directionless and contentionless reconfigurability to route individual wavelengths from fibre to fibre across multiple nodes to avoid wavelength blocking/collisions, increasing the flexibility and capacity of a network. For these networks to function optimally, novel hardware and dynamic algorithms identification and development is a critical task. Such tasks include wavelength assignment, signal routing, network restoration and network protection. In this work, we for the first time to our knowledge proposed a spectrum defragmentation technique through reallocation of the central frequency of the optical transmitter, to increase the probability of finding a sufficient continuous spectrum. This is to improve network resource utilisation, capacity and resolve wavelength contention associated with a flexible spectrum in optical communication networks. The following chapter provides details on a flexible spectrum in optical fibre networks utilising DWDM, optimising transmitter-receivers, advanced modulation formats, coherent detection, reconfigurable optical add and drop multiplexer (ROADM) technology to implement hardware and middleware platforms which address growing bandwidth demands for scalability, flexibility and cost-efficiency. A major attribute is tunable lasers, an essential component for future flexible spectrum with application to wavelength switching, routing, wavelength conversion and ROADM for the multi-node optical network through spectrum flexibility and cost-effective sharing of fibre links, transmitters and receivers. Spectrum slicing into fine granular sub-carriers and assigning several frequency slots to accommodate diverse traffic demands is a viable approach. This work experimentally presents a spectral efficient technique for bandwidth variability, wavelength allocation, routing, defragmentation and wavelength selective switches in the nodes of a network, capable of removing the fixed grid spacing using low cost, high bandwidth, power-efficient and wavelength-tunable vertical-cavity surface-emitting laser (VCSEL) transmitter directly modulated with 10 Gbps data. This to ensure that majority of the spectrum utilisation at finer channel spacing, wastage of the spectrum resource as caused by the wavelength continuity constraint reduction and it improves bandwidth utilisation. The technique is flexible in terms of modulation formats and accommodates various formats with spectrally continuous channels, fulfilling the future bandwidth demands with transmissions beyond 100 Gbps per channel while maintaining spectral efficiency.
- Full Text:
- Date Issued: 2020
- Authors: Dlamini, Phumla Patience
- Date: 2020
- Subjects: Fiber optics
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/50666 , vital:42329
- Description: Flexible Spectrum Dense Wavelength Division Multiplexed (DWDM) optical fibre networks are next-generation technology for handling extremely high data rates of the kind produced by MeerKAT and SKA.We optimise the flexible spectrum for real-time dynamic channel wavelength assignment, to ensure optimum network performance. We needed to identify and develop novel hardware and dynamic algorithms for these networks to function optimally to perform critical tasks. Such tasks include wavelength assignment, signal routing, network restoration and network protection. The antennas of the Square Kilometre Array (SKA) network connect to the correlator and data processor in a simple point-to-point fixed configuration. The connection of the astronomer users to the data processor, however, requires a more complex network architecture. This is because the network has users scattered around South Africa, Africa and the whole world. This calls for upgrade of the classical fixed wavelength spectrum grids, to flexible spectrum grid that has improved capacity, reliable, simple and cost-effectiveness through sharing of network infrastructure. The exponential growth of data traffic in current optical communication networks requires higher capacity for the bandwidth demands at a reduced cost per bit. All-optical signal processing is a promising technique to improve network resource utilisation and resolve wavelength contention associated with the flexible spectrum. Flexible Spectrum Dense Wavelength Division Multiplexed (DWDM) optical fibre networks are next-generation technology for handling extremely high data rates of the kind produced by MeerKAT and SKA. Each DWDM channel is capable of 10 Gbps transmission rate, which is sliceable into finer flexible grid 12.5 GHz granularity to offer the network elastic spectrum and channel spacing capable of signal routing and wavelength switching for the scalability of aggregate bandwidth. The variable-sized portions of the flexible spectrum assignment to end users at different speeds depend on bandwidth demand, allowing efficient utilisation of the spectrum resources. The entire bandwidth of dynamic optical connections must be contiguously allocated. However, there is an introduction of spectrum fragmentation due to spectrum contiguity related to the optical channels having different width. Thus large traffic demands are likely to experience blocking regardless of available bandwidth. To minimise the congestion and cost-effectively obtain high performance, the optical network must be reconfigurable, achievable by adding wavelength as an extra degree of freedom for effectiveness. This can introduce colourless, directionless and contentionless reconfigurability to route individual wavelengths from fibre to fibre across multiple nodes to avoid wavelength blocking/collisions, increasing the flexibility and capacity of a network. For these networks to function optimally, novel hardware and dynamic algorithms identification and development is a critical task. Such tasks include wavelength assignment, signal routing, network restoration and network protection. In this work, we for the first time to our knowledge proposed a spectrum defragmentation technique through reallocation of the central frequency of the optical transmitter, to increase the probability of finding a sufficient continuous spectrum. This is to improve network resource utilisation, capacity and resolve wavelength contention associated with a flexible spectrum in optical communication networks. The following chapter provides details on a flexible spectrum in optical fibre networks utilising DWDM, optimising transmitter-receivers, advanced modulation formats, coherent detection, reconfigurable optical add and drop multiplexer (ROADM) technology to implement hardware and middleware platforms which address growing bandwidth demands for scalability, flexibility and cost-efficiency. A major attribute is tunable lasers, an essential component for future flexible spectrum with application to wavelength switching, routing, wavelength conversion and ROADM for the multi-node optical network through spectrum flexibility and cost-effective sharing of fibre links, transmitters and receivers. Spectrum slicing into fine granular sub-carriers and assigning several frequency slots to accommodate diverse traffic demands is a viable approach. This work experimentally presents a spectral efficient technique for bandwidth variability, wavelength allocation, routing, defragmentation and wavelength selective switches in the nodes of a network, capable of removing the fixed grid spacing using low cost, high bandwidth, power-efficient and wavelength-tunable vertical-cavity surface-emitting laser (VCSEL) transmitter directly modulated with 10 Gbps data. This to ensure that majority of the spectrum utilisation at finer channel spacing, wastage of the spectrum resource as caused by the wavelength continuity constraint reduction and it improves bandwidth utilisation. The technique is flexible in terms of modulation formats and accommodates various formats with spectrally continuous channels, fulfilling the future bandwidth demands with transmissions beyond 100 Gbps per channel while maintaining spectral efficiency.
- Full Text:
- Date Issued: 2020
Database and guide for Lesotho wool and mohair production and quality
- Authors: Maqalika, Papali Elizabeth
- Date: 2020
- Subjects: Mohair -- Dissertations -- Lesotho , Wool -- Dissertations -- Lesotho Textile fabrics -- Lesotho Textile industry -- Lesotho
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/49148 , vital:41605
- Description: Lesotho produces significant quantities of Merino apparel wool and mohair, both of a quality that allows them to compete on the global market and to make a significant contribution to the economy of the country. Nevertheless, very little production and quality data and trends of these fibres have been documented. This is a serious disadvantage in terms of international trading and benchmarking as well as attempts to improve the production and quality of Lesotho wool and mohair. In the light of the aforementioned, the available production and quality data of the past 10 years have been captured and analysed for trends and also benchmarking, where considered applicable. The main focus is on fibre diameter (fineness), staple length, and yield since they largely determine fibre quality, application and price. Some quality related tests were undertaken to fill certain important gaps in the available data. In addition, prickle and medullation were evaluated on representative wool and mohair samples, respectively, because they represent important quality measures for apparel wool and mohair, respectively. It was found that Lesotho wool and mohair are of a fairly good and internationally competitive quality, with the wool having an average fibre diameter (MFD) of ≈20μm, an average staple length of about 64mm, an average VM level of about 4%, an average yield of about 57%, and the annual production being about 4 million kilograms greasy. The average staple length of the mohair was about 140mm, average MFD about 29μm and the average medullation, which unless otherwise specified, refers to the objectionable medullated fibres (kemp type) including the flat medullated fibre, relatively high at about 5.7%. The latter is certainly an area of concern which needs attention and improvement. It was found that the prickle level (Comfort Factor), of some of the wools tested was of such a level as to make the wool suitable for wearing against the skin. With respect to mohair, there is considerable scope to substantially reduce the level of objectionable medullated (kemp style) fibre level through the appropriate breeding interventions. Production of both wool and mohair has increased slightly over the ten years covered by this study. It also became apparent that Lesotho wool and mohair quality and production are greatly influenced by the farming practices and climatic conditions. Greater adoption of the merino sheep breed, sheds/barns and sheep coats are suggested as ways to reduce mortality rate (due to extremely cold temperatures), improve quality and increase yield and production. Some farming practices such as the lack of barns, supplementary feeding and veterinary care present constraints in terms of production of both wool and mohair. The districts in the Highlands region had the highest production of mostly wool, this being ascribed to better pastures, climatic and other conditions conducive to wool and mohair production. Both wool and mohair are considered to have potential for improved quality and production, which could be affected by appropriate interventions by the Small Agricultural and Development Project (SADP) and others through National Wool and Mohair Growers Association (NWMGA). Nevertheless, since the wool and mohair growers (farmers) do not form part of these and other interventions, they do not readily adopt the various strategies and decisions and do not receive the associated benefits immediately. It is therefore advised that local farmers, relevant educators and researchers be represented in policy and other decision making forums. In this way, educational campaigns will be demand driven with greater chance of adoption and success.
- Full Text:
- Date Issued: 2020
- Authors: Maqalika, Papali Elizabeth
- Date: 2020
- Subjects: Mohair -- Dissertations -- Lesotho , Wool -- Dissertations -- Lesotho Textile fabrics -- Lesotho Textile industry -- Lesotho
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/49148 , vital:41605
- Description: Lesotho produces significant quantities of Merino apparel wool and mohair, both of a quality that allows them to compete on the global market and to make a significant contribution to the economy of the country. Nevertheless, very little production and quality data and trends of these fibres have been documented. This is a serious disadvantage in terms of international trading and benchmarking as well as attempts to improve the production and quality of Lesotho wool and mohair. In the light of the aforementioned, the available production and quality data of the past 10 years have been captured and analysed for trends and also benchmarking, where considered applicable. The main focus is on fibre diameter (fineness), staple length, and yield since they largely determine fibre quality, application and price. Some quality related tests were undertaken to fill certain important gaps in the available data. In addition, prickle and medullation were evaluated on representative wool and mohair samples, respectively, because they represent important quality measures for apparel wool and mohair, respectively. It was found that Lesotho wool and mohair are of a fairly good and internationally competitive quality, with the wool having an average fibre diameter (MFD) of ≈20μm, an average staple length of about 64mm, an average VM level of about 4%, an average yield of about 57%, and the annual production being about 4 million kilograms greasy. The average staple length of the mohair was about 140mm, average MFD about 29μm and the average medullation, which unless otherwise specified, refers to the objectionable medullated fibres (kemp type) including the flat medullated fibre, relatively high at about 5.7%. The latter is certainly an area of concern which needs attention and improvement. It was found that the prickle level (Comfort Factor), of some of the wools tested was of such a level as to make the wool suitable for wearing against the skin. With respect to mohair, there is considerable scope to substantially reduce the level of objectionable medullated (kemp style) fibre level through the appropriate breeding interventions. Production of both wool and mohair has increased slightly over the ten years covered by this study. It also became apparent that Lesotho wool and mohair quality and production are greatly influenced by the farming practices and climatic conditions. Greater adoption of the merino sheep breed, sheds/barns and sheep coats are suggested as ways to reduce mortality rate (due to extremely cold temperatures), improve quality and increase yield and production. Some farming practices such as the lack of barns, supplementary feeding and veterinary care present constraints in terms of production of both wool and mohair. The districts in the Highlands region had the highest production of mostly wool, this being ascribed to better pastures, climatic and other conditions conducive to wool and mohair production. Both wool and mohair are considered to have potential for improved quality and production, which could be affected by appropriate interventions by the Small Agricultural and Development Project (SADP) and others through National Wool and Mohair Growers Association (NWMGA). Nevertheless, since the wool and mohair growers (farmers) do not form part of these and other interventions, they do not readily adopt the various strategies and decisions and do not receive the associated benefits immediately. It is therefore advised that local farmers, relevant educators and researchers be represented in policy and other decision making forums. In this way, educational campaigns will be demand driven with greater chance of adoption and success.
- Full Text:
- Date Issued: 2020
Decision analysis to inform invasive alien plant management in the Garden Route Biosphere Reserve
- Authors: Masunungure, Current
- Date: 2020
- Subjects: Invasive plants -- South Africa , Alien plants -- South Africa , Biodiversity -- South Africa -- Management , Ecosystem management -- South Africa
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/49159 , vital:41606
- Description: Invasive alien plants (IAP) pose significant threats to global economies and biodiversity and are often considered as wicked problems. With an increasing number of IAP and limited resources, their management and decision-making processes are becoming difficult because of uncertainty, multiple and conflicting objectives, and diverse stakeholder views, facts and values. This is particularly challenging given the complex interactions between economic, ecological, and social elements that exist in invaded areas. Consequently, it is important to incorporate new ways of thinking and novel methodologies to improve our understanding of IAP management and the decision-making processes around them, which are currently inadequate. Decision analysis can help with dealing with these challenges and support decision-making under uncertainty. Drawing on the systems thinking approach and the concepts of leverage points, transition management and transformational change, the aim of this thesis was to explore the effectiveness of IAP management and the decision-making process in the Garden Route Biosphere Reserve (GRBR). This was achieved using a mixed methods approach involving: social-ecological inventory (identifying relevant stakeholders); review of literature on the available decision support tools; key informant interviews (stakeholder perspectives on the current decision-making process); and stakeholder workshop and expert consultation (casual loop modelling). The results of this thesis provide evidence that application of the proposed principles of robust decision-making has the potential to overcome the weaknesses of the current decision-making process and as such, enables decision-makers to efficiently allocate resources towards IAS management. A novel causal loop diagram (CLD) was developed to highlight the interconnections between key variables in IAP management and decision-making. This revealed that to transcend ‘policy resistance’ and ‘quickfixes that fail’ archetypes, and improve IAP management, the stakeholders need to consider deep leverage points, for example, fostering trust and shared understanding among different stakeholder groups. These can be realistically maintained over the long-term and can cause a fundamental change in IAP management, rather than focusing on shallow leverage points that are relatively easy to implement but do not result in significant systemic change. The findings of this thesis are flexible and could guide various stakeholder groups at local, national, and international scales in improving the effectiveness of IAP management and decision-making.
- Full Text:
- Date Issued: 2020
- Authors: Masunungure, Current
- Date: 2020
- Subjects: Invasive plants -- South Africa , Alien plants -- South Africa , Biodiversity -- South Africa -- Management , Ecosystem management -- South Africa
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/49159 , vital:41606
- Description: Invasive alien plants (IAP) pose significant threats to global economies and biodiversity and are often considered as wicked problems. With an increasing number of IAP and limited resources, their management and decision-making processes are becoming difficult because of uncertainty, multiple and conflicting objectives, and diverse stakeholder views, facts and values. This is particularly challenging given the complex interactions between economic, ecological, and social elements that exist in invaded areas. Consequently, it is important to incorporate new ways of thinking and novel methodologies to improve our understanding of IAP management and the decision-making processes around them, which are currently inadequate. Decision analysis can help with dealing with these challenges and support decision-making under uncertainty. Drawing on the systems thinking approach and the concepts of leverage points, transition management and transformational change, the aim of this thesis was to explore the effectiveness of IAP management and the decision-making process in the Garden Route Biosphere Reserve (GRBR). This was achieved using a mixed methods approach involving: social-ecological inventory (identifying relevant stakeholders); review of literature on the available decision support tools; key informant interviews (stakeholder perspectives on the current decision-making process); and stakeholder workshop and expert consultation (casual loop modelling). The results of this thesis provide evidence that application of the proposed principles of robust decision-making has the potential to overcome the weaknesses of the current decision-making process and as such, enables decision-makers to efficiently allocate resources towards IAS management. A novel causal loop diagram (CLD) was developed to highlight the interconnections between key variables in IAP management and decision-making. This revealed that to transcend ‘policy resistance’ and ‘quickfixes that fail’ archetypes, and improve IAP management, the stakeholders need to consider deep leverage points, for example, fostering trust and shared understanding among different stakeholder groups. These can be realistically maintained over the long-term and can cause a fundamental change in IAP management, rather than focusing on shallow leverage points that are relatively easy to implement but do not result in significant systemic change. The findings of this thesis are flexible and could guide various stakeholder groups at local, national, and international scales in improving the effectiveness of IAP management and decision-making.
- Full Text:
- Date Issued: 2020
Development of MOVPE-grown InAsSb for barrier diode applications
- Authors: Dobson, Stephen R
- Date: 2020
- Subjects: Gallium arsenide semiconductors , Electronics
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/49071 , vital:41598
- Description: In this study, layers of GaSb, InAs and InAsSb are grown by metalorganic vapour phase epitaxy and characterised. Growth is conducted using the precursors of trimethylgallium, trimethylindium, trimethylantimony and tertiarybutylarsine. Focus is then placed on the characterisation of the materials, carried out by the techniques of X-ray diffraction, photoluminescence spectroscopy, Hall measurements and photocurrent spectroscopy. It is observed that V/III ratio plays a vital role in the growth of the GaSb and InAsSb layers. Epilayers of GaSb showed best crystalline quality when a V/III ratio of 1.2 was used at a growth temperature of 600 °C and a cell pressure of 600 Torr. Resultant Hall measurements indicated p-type GaSb. The Hall carrier concentrations of the p-type GaSb samples were analyzed considering electrical neutrality conditions and found to be highly compensated with evidence of band impurity conduction at low measurement temperatures. Both the donor and acceptor concentrations were determined to be of the order of 1016 cm−3 for all samples. For low temperature (< 150 K) a monovalent acceptor concentration is calculated to have an activation energy at approximately 20 meV. At high temperature (> 150 K) a divalent acceptor is extracted with an activation energy varying between samples based on compensation in a range of 90 meV to 70 meV. Photoluminescence measurements show four peaks with recombination mechanisms linked to the native acceptor identified in literature as either the gallium antisite and/or vacant gallium site. A fifth peak observed is attributed to the longitudinal phonon of the native acceptor. InAs and InAsSb epilayer are all grown at a temperature of 600 °C and cell pressure of 600 Torr. InAs is grown at a V/III ratio of 9.5 on GaAs substrate. Photoluminescence of the InAs layer shows two distinct peaks, one of which is an extrinsic band to band recombination. The other is attributed to free electron to acceptor or a donor-acceptor pair transition. An additional weak peak is also observed which is assigned to the longitudinal phonon of the band to band. InAsSb growth was conducted under a range of V/III ratios of 4.8 to 5, with a vapour phase composition of 0.4 to 0.435. Structural analysis via X-ray diffraction showed a 6 % to 12 % solid antimony content. Photoluminescence exhibited a single broad peak for all samples, with extended band tails. Temperature and power dependant analysis of luminescence indicated a convolution of extrinsic band to tail and band to band recombinations. Hall measurements indicated the InAsSb was n-type material with an apparent measured maximum mobility at 120 K of 9.5 × 103 cm2/V.s. and a room temperature apparent mobility of 7.5 × 103 cm2/V.s. Analysis of hall results using a two-layer model calculated a true bulk mobility of the epilayer at room temperature with an increased value of 15.4 × 103 cm2/V.s. The two-layer model details the effects of the surface conduction. From photoconductivity measurements and further analysis a resultant effective lifetime, at room temperature, was found to be on same order of magnitude as that of InAs materials. Application of a single Einstein oscillator extrapolated 0 K energy gaps for two samples of solid Sb contents of 6 % and 12 %, of 354 meV and 332 meV, respectively. Finally consideration was given to the growth of aluminium containing compounds, particularly AlGaSb. Multiple phases were observed under scanning electron microscope showing growth of GaSb regions surrounded by amorphous solid aluminium and/or aluminium oxide phases. The failure of the aluminium to incorporate into the desired crystal structure is speculated to be due to impure precursor introducing oxygen into the films. Additionally, the effectiveness of the gallium precursor compared to the aluminium precursor in helping the removal of the methyl groups at the growth surface, could also promote a preference for GaSb growth.
- Full Text:
- Date Issued: 2020
- Authors: Dobson, Stephen R
- Date: 2020
- Subjects: Gallium arsenide semiconductors , Electronics
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/49071 , vital:41598
- Description: In this study, layers of GaSb, InAs and InAsSb are grown by metalorganic vapour phase epitaxy and characterised. Growth is conducted using the precursors of trimethylgallium, trimethylindium, trimethylantimony and tertiarybutylarsine. Focus is then placed on the characterisation of the materials, carried out by the techniques of X-ray diffraction, photoluminescence spectroscopy, Hall measurements and photocurrent spectroscopy. It is observed that V/III ratio plays a vital role in the growth of the GaSb and InAsSb layers. Epilayers of GaSb showed best crystalline quality when a V/III ratio of 1.2 was used at a growth temperature of 600 °C and a cell pressure of 600 Torr. Resultant Hall measurements indicated p-type GaSb. The Hall carrier concentrations of the p-type GaSb samples were analyzed considering electrical neutrality conditions and found to be highly compensated with evidence of band impurity conduction at low measurement temperatures. Both the donor and acceptor concentrations were determined to be of the order of 1016 cm−3 for all samples. For low temperature (< 150 K) a monovalent acceptor concentration is calculated to have an activation energy at approximately 20 meV. At high temperature (> 150 K) a divalent acceptor is extracted with an activation energy varying between samples based on compensation in a range of 90 meV to 70 meV. Photoluminescence measurements show four peaks with recombination mechanisms linked to the native acceptor identified in literature as either the gallium antisite and/or vacant gallium site. A fifth peak observed is attributed to the longitudinal phonon of the native acceptor. InAs and InAsSb epilayer are all grown at a temperature of 600 °C and cell pressure of 600 Torr. InAs is grown at a V/III ratio of 9.5 on GaAs substrate. Photoluminescence of the InAs layer shows two distinct peaks, one of which is an extrinsic band to band recombination. The other is attributed to free electron to acceptor or a donor-acceptor pair transition. An additional weak peak is also observed which is assigned to the longitudinal phonon of the band to band. InAsSb growth was conducted under a range of V/III ratios of 4.8 to 5, with a vapour phase composition of 0.4 to 0.435. Structural analysis via X-ray diffraction showed a 6 % to 12 % solid antimony content. Photoluminescence exhibited a single broad peak for all samples, with extended band tails. Temperature and power dependant analysis of luminescence indicated a convolution of extrinsic band to tail and band to band recombinations. Hall measurements indicated the InAsSb was n-type material with an apparent measured maximum mobility at 120 K of 9.5 × 103 cm2/V.s. and a room temperature apparent mobility of 7.5 × 103 cm2/V.s. Analysis of hall results using a two-layer model calculated a true bulk mobility of the epilayer at room temperature with an increased value of 15.4 × 103 cm2/V.s. The two-layer model details the effects of the surface conduction. From photoconductivity measurements and further analysis a resultant effective lifetime, at room temperature, was found to be on same order of magnitude as that of InAs materials. Application of a single Einstein oscillator extrapolated 0 K energy gaps for two samples of solid Sb contents of 6 % and 12 %, of 354 meV and 332 meV, respectively. Finally consideration was given to the growth of aluminium containing compounds, particularly AlGaSb. Multiple phases were observed under scanning electron microscope showing growth of GaSb regions surrounded by amorphous solid aluminium and/or aluminium oxide phases. The failure of the aluminium to incorporate into the desired crystal structure is speculated to be due to impure precursor introducing oxygen into the films. Additionally, the effectiveness of the gallium precursor compared to the aluminium precursor in helping the removal of the methyl groups at the growth surface, could also promote a preference for GaSb growth.
- Full Text:
- Date Issued: 2020
Development of Tio 2 nanostructure arrays for photonic extraction of hydrogen gas
- Authors: Suliali, Nyasha Joseph
- Date: 2020
- Subjects: Nanostructures , Nanostructured materials Hydrogen
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/49314 , vital:41620
- Description: Amid the energy crisis of the 21st century, renewable energy is a thriving field of study, light harvesting materials being a central theme due to the abundance of solar energy. Nanostructured TiO2 is the most studied photocatalysis material, since the discovery of its energy harvesting properties by Fujishima and Honda in 1972. Environmentally friendly products such as hydrogen fuel, can be produced using TiO2 due to its non-toxicity, chemical stability and photocatalytic activity. The surprising aspect of this important material is that it can be prepared using cost-effective methods such as hydrothermal synthesis, solution gelation and anodic oxidation. This research focused on the three key elements required to develop TiO2 photoelectrodes i.e. the deposition of Ti films on transparent substrates, a thorough analysis of the chemistry of the anodic oxidation process and the development of the TiO2 thin films. Glass substrates that have Ti films are the base component for TiO2 photoelectrode production. Ti films with thicknesses up to 4 μm, were developed on commercial F-doped SnO2 (FTO) glass substrates using high-power impulse magnetron sputtering (HiPIMS). The sputter deposition experiments were performed in the 1 to 8 kW range at a substrate temperature of 500 °C and Ar pressure of 400 mPa. At higher powers, thicker films were deposited, resulting in increased intensity of Xray-diffraction peaks. However, on comparing the XRD patterns, the (001) peak outgrew the rest regardless of thickness of the film. The deposition process therefore favoured orientation of most of the α-Ti phase crystallites with the [001] axis perpendicular to the substrate surface. Surface roughness results were interesting, showing a non-linear dependence of the surface roughness on HiPIMS pulse energy in the 1 to 8 kW range. The surface roughness is highest at the starting deposition power of 1450 W and reduces to a minimum at 4500 W. From this minimum, it increases to its second highest value at 7900 W. From this data, the parameters required to produce Ti films of lowest surface roughness, for deposition on commercial Technistro® FTO glass, were deduced at the inflection point, where the deposition power was 4500 W. The surface roughness obtained is a critical result for the anodic production of quality TiO2 photoelectrodes, which if high, leads to uneven etching, thus irregular and inefficient photoelectrodes. Direct current magnetron sputtering was also carried out in the 1 to 5 kW range to obtain ratios of power-normalised growth rates of the Ti films. This investigation provided the Ti films on FTO glass, the transparent, conductive substrates which were used to develop TiO2 photoelectrodes. To elucidate the chemistry of anodic oxidation of Ti, a mathematical model of the anodic current density, which had not been reported at the time of its publication, was developed. The technique, a highlight of this research, is a predictive numerical computation of the instantaneous quantities of species that participate in the anodization process. From eleven chemical reactions, 14 first order ordinary differential equations were compiled using the principles of chemical reaction kinetics. The pattern, transient behaviour and response to anodization parameters of the current density signal, were successfully predicted. Strong agreement between the model and measurements was demonstrated in seven experiments. The results confirm that the current density signal is a numerical integral of the kinetics of redox reactions of water. The bulk of this research was on the development of TiO2 nanotubular arrays on Ti foil substrates and Ti films on FTO glass. TiO2 films with well-defined tubular structures were synthesised. The films were developed in anhydrous, polar organic hosts with water and etching agents in the range of 10 V to 70 V. The control of geometrical properties of the tubes such as the length, pore diameter, wall thickness, tube separation and number of nanotubes per unit area was demonstrated. Anatase only and mixed anatase-rutile phase compositions were obtained at different annealing temperatures. Nanotubes with diameters as small as 20 nm and thickness as high as 29 μm were produced. Apart from an increase in nanotube thickness, a decrease in distance between nanotubes grown in diethylene glycol was observed at longer anodization times. Studies of the effects of anodization parameters on the current density measured, morphological and crystallographic properties of the nanotube films were conducted in three main investigations. The first was the study of the effect of anodization parameters on current density. Besides the obvious increase of current density with anodic voltage, the first steady state of the growth process was found to depend on the NH4F concentration. The second investigation focused on the effect of accelerated growth of TiO2 nanotubular films. In the study, 9 μm-thick nanotube films were synthesised at twice the growth rate of a 9 μm-thick control sample. The array obtained by accelerated growth had distinguishable nanotubes, however, the morphological quality was reduced. The third investigation demonstrates the control of the number of nanotubes per unit area. By varying the etchant content, the anodic voltage and the viscosity of the electrolyte host, various distributions were obtained. The research ends with a photoelectrochemical application: measurement on photocurrents generated in a two-electrode setup. The photocurrent densities measured in the off and on conditions were 30 nA/cm2 and 2.57 μA/cm2, respectively, demonstrating photoactivity of the developed films.
- Full Text:
- Date Issued: 2020
- Authors: Suliali, Nyasha Joseph
- Date: 2020
- Subjects: Nanostructures , Nanostructured materials Hydrogen
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/49314 , vital:41620
- Description: Amid the energy crisis of the 21st century, renewable energy is a thriving field of study, light harvesting materials being a central theme due to the abundance of solar energy. Nanostructured TiO2 is the most studied photocatalysis material, since the discovery of its energy harvesting properties by Fujishima and Honda in 1972. Environmentally friendly products such as hydrogen fuel, can be produced using TiO2 due to its non-toxicity, chemical stability and photocatalytic activity. The surprising aspect of this important material is that it can be prepared using cost-effective methods such as hydrothermal synthesis, solution gelation and anodic oxidation. This research focused on the three key elements required to develop TiO2 photoelectrodes i.e. the deposition of Ti films on transparent substrates, a thorough analysis of the chemistry of the anodic oxidation process and the development of the TiO2 thin films. Glass substrates that have Ti films are the base component for TiO2 photoelectrode production. Ti films with thicknesses up to 4 μm, were developed on commercial F-doped SnO2 (FTO) glass substrates using high-power impulse magnetron sputtering (HiPIMS). The sputter deposition experiments were performed in the 1 to 8 kW range at a substrate temperature of 500 °C and Ar pressure of 400 mPa. At higher powers, thicker films were deposited, resulting in increased intensity of Xray-diffraction peaks. However, on comparing the XRD patterns, the (001) peak outgrew the rest regardless of thickness of the film. The deposition process therefore favoured orientation of most of the α-Ti phase crystallites with the [001] axis perpendicular to the substrate surface. Surface roughness results were interesting, showing a non-linear dependence of the surface roughness on HiPIMS pulse energy in the 1 to 8 kW range. The surface roughness is highest at the starting deposition power of 1450 W and reduces to a minimum at 4500 W. From this minimum, it increases to its second highest value at 7900 W. From this data, the parameters required to produce Ti films of lowest surface roughness, for deposition on commercial Technistro® FTO glass, were deduced at the inflection point, where the deposition power was 4500 W. The surface roughness obtained is a critical result for the anodic production of quality TiO2 photoelectrodes, which if high, leads to uneven etching, thus irregular and inefficient photoelectrodes. Direct current magnetron sputtering was also carried out in the 1 to 5 kW range to obtain ratios of power-normalised growth rates of the Ti films. This investigation provided the Ti films on FTO glass, the transparent, conductive substrates which were used to develop TiO2 photoelectrodes. To elucidate the chemistry of anodic oxidation of Ti, a mathematical model of the anodic current density, which had not been reported at the time of its publication, was developed. The technique, a highlight of this research, is a predictive numerical computation of the instantaneous quantities of species that participate in the anodization process. From eleven chemical reactions, 14 first order ordinary differential equations were compiled using the principles of chemical reaction kinetics. The pattern, transient behaviour and response to anodization parameters of the current density signal, were successfully predicted. Strong agreement between the model and measurements was demonstrated in seven experiments. The results confirm that the current density signal is a numerical integral of the kinetics of redox reactions of water. The bulk of this research was on the development of TiO2 nanotubular arrays on Ti foil substrates and Ti films on FTO glass. TiO2 films with well-defined tubular structures were synthesised. The films were developed in anhydrous, polar organic hosts with water and etching agents in the range of 10 V to 70 V. The control of geometrical properties of the tubes such as the length, pore diameter, wall thickness, tube separation and number of nanotubes per unit area was demonstrated. Anatase only and mixed anatase-rutile phase compositions were obtained at different annealing temperatures. Nanotubes with diameters as small as 20 nm and thickness as high as 29 μm were produced. Apart from an increase in nanotube thickness, a decrease in distance between nanotubes grown in diethylene glycol was observed at longer anodization times. Studies of the effects of anodization parameters on the current density measured, morphological and crystallographic properties of the nanotube films were conducted in three main investigations. The first was the study of the effect of anodization parameters on current density. Besides the obvious increase of current density with anodic voltage, the first steady state of the growth process was found to depend on the NH4F concentration. The second investigation focused on the effect of accelerated growth of TiO2 nanotubular films. In the study, 9 μm-thick nanotube films were synthesised at twice the growth rate of a 9 μm-thick control sample. The array obtained by accelerated growth had distinguishable nanotubes, however, the morphological quality was reduced. The third investigation demonstrates the control of the number of nanotubes per unit area. By varying the etchant content, the anodic voltage and the viscosity of the electrolyte host, various distributions were obtained. The research ends with a photoelectrochemical application: measurement on photocurrents generated in a two-electrode setup. The photocurrent densities measured in the off and on conditions were 30 nA/cm2 and 2.57 μA/cm2, respectively, demonstrating photoactivity of the developed films.
- Full Text:
- Date Issued: 2020
Foraging in a dynamic environment: movement and stable isotope ecology of marine top predators breeding at the Prince Edward Archipelago
- Authors: Carpenter-Kling, Tegan
- Date: 2020
- Subjects: Marine ecology , Stable isotope tracers , Estuarine ecology
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/49049 , vital:41596
- Description: Marine ecosystems are experiencing rapid changes due to climate change. The associated temporal and spatial changes in resource distribution impacts on the foraging behaviour of marine top predators. If these changes negatively affect the ability of marine predators to forage efficiently, there may be dire consequences for their populations. However, evidence of foraging plasticity during adverse conditions, or generalist foraging behaviour, can allow inference about the resilience of species to environmental change and provide essential knowledge towards effective and proactive conservation measures. I examined plasticity in the trophic ecology of 12 marine predator species breeding on Marion Island, southern Indian Ocean, over three years (2015 – 2018), a period characterized by pronounced environmental variability. Firstly, I correlated behavioural states inferred along the GPS tracks of incubating wandering, grey-headed, sooty and light-mantled albatrosses to environmental variables that are indicative of biologically productive oceanographic features. Secondly, I analysed δ13C and δ15N blood values in 12 marine predator species (the afore-mentioned albatrosses as well as king, gentoo, macaroni and eastern rockhopper penguins, northern and southern giant petrels and Antarctic and sub-Antarctic fur seals) over two seasons: summer and autumn. My results revealed that the foraging behaviour of all the species is, to some degree, either plastic (temporally variable isotopic niche) or general (large isotopic niche which remained similar over time), except for the king penguin (small isotopic niche which remained similar over time), a known foraging specialist. Furthermore, despite their dynamic foraging behaviour, resource partitioning among the predators was maintained over time. Due to the ease and relatively low cost of collecting tissues for stable isotope analysis it has the potential to be a powerful tool to monitor the trophic ecology of marine predators. I thus used my simultaneously collected dataset of GPS tracks with the stable isotope blood compositions to investigate some of the assumptions underlying the inferences made from marine predator δ13C and δ15N blood values. I reconstructed species- and guild- specific δ13C and δ15N isoscapes for eight seabird species. Following this, I coupled individual-based movement models of northern giant petrels with global marine isotope models to explore the sensitivity of tissue δ13C values to a range of extrinsic (environmental) and intrinsic (behavioural, physiological) drivers. My results demonstrate the strong influence of reference isoscapes on the inference of stable isotope compositions of marine predators. Furthermore, I show that caution should be used when using non-species-specific or temporally inaccurate isoscapes. I furthermore demonstrate that biological interactions, such as competition for food resources, either past or present, as well as spatio-temporal distribution of food patches strongly influence the foraging behaviour of marine predators. These findings highlight the importance of integrating biological interactions in species distribution models which are used to predict possible distributional shifts of marine predators in the context of global changes. My thesis further developed previously available methods and presents a novel approach to investigate sources of variance in the stable isotopic composition of animals’ tissues.
- Full Text:
- Date Issued: 2020
- Authors: Carpenter-Kling, Tegan
- Date: 2020
- Subjects: Marine ecology , Stable isotope tracers , Estuarine ecology
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/49049 , vital:41596
- Description: Marine ecosystems are experiencing rapid changes due to climate change. The associated temporal and spatial changes in resource distribution impacts on the foraging behaviour of marine top predators. If these changes negatively affect the ability of marine predators to forage efficiently, there may be dire consequences for their populations. However, evidence of foraging plasticity during adverse conditions, or generalist foraging behaviour, can allow inference about the resilience of species to environmental change and provide essential knowledge towards effective and proactive conservation measures. I examined plasticity in the trophic ecology of 12 marine predator species breeding on Marion Island, southern Indian Ocean, over three years (2015 – 2018), a period characterized by pronounced environmental variability. Firstly, I correlated behavioural states inferred along the GPS tracks of incubating wandering, grey-headed, sooty and light-mantled albatrosses to environmental variables that are indicative of biologically productive oceanographic features. Secondly, I analysed δ13C and δ15N blood values in 12 marine predator species (the afore-mentioned albatrosses as well as king, gentoo, macaroni and eastern rockhopper penguins, northern and southern giant petrels and Antarctic and sub-Antarctic fur seals) over two seasons: summer and autumn. My results revealed that the foraging behaviour of all the species is, to some degree, either plastic (temporally variable isotopic niche) or general (large isotopic niche which remained similar over time), except for the king penguin (small isotopic niche which remained similar over time), a known foraging specialist. Furthermore, despite their dynamic foraging behaviour, resource partitioning among the predators was maintained over time. Due to the ease and relatively low cost of collecting tissues for stable isotope analysis it has the potential to be a powerful tool to monitor the trophic ecology of marine predators. I thus used my simultaneously collected dataset of GPS tracks with the stable isotope blood compositions to investigate some of the assumptions underlying the inferences made from marine predator δ13C and δ15N blood values. I reconstructed species- and guild- specific δ13C and δ15N isoscapes for eight seabird species. Following this, I coupled individual-based movement models of northern giant petrels with global marine isotope models to explore the sensitivity of tissue δ13C values to a range of extrinsic (environmental) and intrinsic (behavioural, physiological) drivers. My results demonstrate the strong influence of reference isoscapes on the inference of stable isotope compositions of marine predators. Furthermore, I show that caution should be used when using non-species-specific or temporally inaccurate isoscapes. I furthermore demonstrate that biological interactions, such as competition for food resources, either past or present, as well as spatio-temporal distribution of food patches strongly influence the foraging behaviour of marine predators. These findings highlight the importance of integrating biological interactions in species distribution models which are used to predict possible distributional shifts of marine predators in the context of global changes. My thesis further developed previously available methods and presents a novel approach to investigate sources of variance in the stable isotopic composition of animals’ tissues.
- Full Text:
- Date Issued: 2020
Ocean dynamics of the shelf and bays of the eastern Agulhas bank: a process-oriented numerical modelling study
- Authors: Bailey, Dylan Francis
- Date: 2020
- Subjects: Oceanography -- Research , Marine biology , Marine meteorology , Hydrography
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/49038 , vital:41595
- Description: There is a general need for a broader base of information on the finer detail of ocean processes in the Eastern Agulhas Bank region. In this thesis, the Regional Ocean Modelling System (ROMS) was applied to resolve high resolution (~1km) bay and shelf scale ocean processes along the south eastern coastline of South Africa using an offline multi-nested approach. Experimental and reference ocean model runs were used with in situ instrument mooring data to investigate how currents and temperature in the bays and the adjacent shelf are affected by the bathymetry, Agulhas Current (AC), regional winds, and coastal trapped waves (CTW). Focus was given to temperature, circulation, and coastal trapped waves, given the abundance of hourly in situ mooring data for these variables, making them suitable for comparison with the ocean model. SST variability trends were seasonally and spatially predictable, being greatest in the summer months and smallest in the winter months. A region of high SST variability extended from Port Alfred toward the south-west, indicating entrainment of nearshore upwelled water and instabilities in the inshore front of the AC. Shelf bottom temperature variability had a more spatially diverse seasonal trend, with coastal variability the highest (lowest) and offshore variability the lowest (highest) during the summer (winter) months. A distinct asymmetry in mean seasonal shelf bottom temperatures evident to the south of Algoa Bay and St Francis Bay was shown to be dependent on the profile of the shelf margin in relation to the inshore front of the AC using an experimental model run. Bay-scale anti-cyclonic circulation patterns dominate the bays throughout the year, with regions of strong mean surface currents occurring south of Cape Recife and Cape St Francis. General circulation patterns over the shelf are strongly governed by regional wind regimes, with the strongest flows being south-westward occurring during north-easterly wind events. Shelf bottom flows typically follow the expected Ekman-veering response, apart from near coastal flows that are topographically steered. A region of consistently strong south-westward shelf bottom flow extends from the east of Port Alfred to the shelf margin south of Bird Island. This flow is predominantly driven by the AC, although its velocity is strongly governed by regional winds. Circulation patterns during powerful eastward moving wind systems are strongly governed by the passage of associated CTW. Comparisons with an experimental model run where surface winds were removed revealed that the shelf wave was responsible for 61% of water volume displacement during a CTW event, and that the coastal shear turbulence caused by this displacement has a direct effect on the mixed layer depth. The links made between the complex shelf and bay ocean processes indicate that subtle changes to the ocean state will have far reaching effects on the regional ocean environment and its biota. Given the challenges of climate change, this is of paramount importance if we are to quantify these effects and mitigate their impacts, particularly with respect to the socio-economic dependency of those who rely on the ocean for its living resources.
- Full Text:
- Date Issued: 2020
- Authors: Bailey, Dylan Francis
- Date: 2020
- Subjects: Oceanography -- Research , Marine biology , Marine meteorology , Hydrography
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/49038 , vital:41595
- Description: There is a general need for a broader base of information on the finer detail of ocean processes in the Eastern Agulhas Bank region. In this thesis, the Regional Ocean Modelling System (ROMS) was applied to resolve high resolution (~1km) bay and shelf scale ocean processes along the south eastern coastline of South Africa using an offline multi-nested approach. Experimental and reference ocean model runs were used with in situ instrument mooring data to investigate how currents and temperature in the bays and the adjacent shelf are affected by the bathymetry, Agulhas Current (AC), regional winds, and coastal trapped waves (CTW). Focus was given to temperature, circulation, and coastal trapped waves, given the abundance of hourly in situ mooring data for these variables, making them suitable for comparison with the ocean model. SST variability trends were seasonally and spatially predictable, being greatest in the summer months and smallest in the winter months. A region of high SST variability extended from Port Alfred toward the south-west, indicating entrainment of nearshore upwelled water and instabilities in the inshore front of the AC. Shelf bottom temperature variability had a more spatially diverse seasonal trend, with coastal variability the highest (lowest) and offshore variability the lowest (highest) during the summer (winter) months. A distinct asymmetry in mean seasonal shelf bottom temperatures evident to the south of Algoa Bay and St Francis Bay was shown to be dependent on the profile of the shelf margin in relation to the inshore front of the AC using an experimental model run. Bay-scale anti-cyclonic circulation patterns dominate the bays throughout the year, with regions of strong mean surface currents occurring south of Cape Recife and Cape St Francis. General circulation patterns over the shelf are strongly governed by regional wind regimes, with the strongest flows being south-westward occurring during north-easterly wind events. Shelf bottom flows typically follow the expected Ekman-veering response, apart from near coastal flows that are topographically steered. A region of consistently strong south-westward shelf bottom flow extends from the east of Port Alfred to the shelf margin south of Bird Island. This flow is predominantly driven by the AC, although its velocity is strongly governed by regional winds. Circulation patterns during powerful eastward moving wind systems are strongly governed by the passage of associated CTW. Comparisons with an experimental model run where surface winds were removed revealed that the shelf wave was responsible for 61% of water volume displacement during a CTW event, and that the coastal shear turbulence caused by this displacement has a direct effect on the mixed layer depth. The links made between the complex shelf and bay ocean processes indicate that subtle changes to the ocean state will have far reaching effects on the regional ocean environment and its biota. Given the challenges of climate change, this is of paramount importance if we are to quantify these effects and mitigate their impacts, particularly with respect to the socio-economic dependency of those who rely on the ocean for its living resources.
- Full Text:
- Date Issued: 2020
Plankton dynamics in response to algal blooms in a highly eutrophic estuary in South Africa
- Authors: Smit, Taryn
- Date: 2020
- Subjects: Algal blooms -- South Africa , Estuaries -- South Africa
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/49303 , vital:41619
- Description: The global rise in the incidence of harmful algal blooms in estuaries is a major concern given their importance as a nursery area for a variety of invertebrates and fishes. On the warm-temperate, south-east coast of South Africa, the Sundays Estuary experiences persistent, seasonal HABs due to stable eutrophic conditions. Spring blooms of the dinoflagellate, Heterosigma akashiwo, were recorded for several years in the estuary. This Raphidophyte is known for its consequences in fish and shellfish aquaculture and exhibits harmful properties such as mucilage and reactive oxygen species production. Subsequently, the dissertation investigated larval fish assemblage dynamics, larval prey and predator dynamics and body condition in association with H. akashiwo blooms. HABs mainly altered the water quality within the mesohaline zone in terms of microalgal biomass and dissolved oxygen. Hypereutrophic blooms of H. akashiwo reached 24 996 cells∙mL-1 during the study period. Supersaturated dissolved oxygen concentrations (up to ~13 mg∙L-1 at night and ~21 mg∙L-1 during the day) were recorded during hypereutrophic blooms (> 80 μg∙L-1, ≥ 2781 H. akashiwo cells∙mL-1). Dissolved oxygen differed significantly by bloom state, with the highest dissolved oxygen concentrations. Larval fish density, diversity and community structure variability were lower during hypereutrophic blooms compared to standard blooms and bloom absence, albeit statistically insignificant. Estuarine resident larval density was negatively correlated with dissolved oxygen concentration (P < 0.001). Dominant prey copepod of G. aestuaria larvae, P. hessei, was less abundant and variable during hypereutrophic blooms (mean: 664, range: 4-4301 individuals per m3) compared to standard blooms (mean: 950, range: 2-5237 individuals per m3) and bloom absence (mean: 1182, range: 23-7259 individuals per m3). Although HABs did not appear to drive a mismatch between prey and G. aestuaria larvae, early juvenile larvae made only a small contribution to the assemblage (0.2-1.8%). In terms of predator effects, yolk sac larvae were negatively related to R. terranatalis (P < 0.05). A significant positive relationship was found between prey, P. hessei and G. aestuaria preflexion (P < 0.001), flexion (P < 0.001) and postflexion (P < 0.05) larvae. In a world first, the effects of HABs were investigated on a larval fish species, Gilchristella aestuaria, using biochemical body condition analyses. During severe bloom conditions, G. aestuaria larvae were smaller and the assemblage consisted of mainly early developmental stages (stats). Body condition (RNA per dry weight) was poorest in the presence of hypereutrophic blooms (P < 0.001). A bell-curved relationship between RNA/DW and dissolved oxygen concentration was also produced in a 3-day time-lagged GAM (P < 0.001). In conclusion, hypereutrophic bloom conditions appeared to trigger adult G. aestuaria spawning in the upper mesohaline zone of the estuary in response to heightened primary productivity which occurs during the early bloom stage. However, reductions in copepod abundance (P. hessei) and unfavourable oxygen concentrations interfered with grow-out of G. aestuaria larvae. This research showed that HABs can alter nursery properties to the detriment of G. aestuaria larvae can have serious knock-on effects up the food chain. The loss in abundance of adult, juvenile and larval G. aestuaria, a key forage fish in South African estuaries, has major implications for estuarine food webs, especially in reduced prey availability to juvenile fishes and birds. There is still much to learn about the effects of HABs on the physiological and behavioural aspects on plankton communities worldwide.
- Full Text:
- Date Issued: 2020
- Authors: Smit, Taryn
- Date: 2020
- Subjects: Algal blooms -- South Africa , Estuaries -- South Africa
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/49303 , vital:41619
- Description: The global rise in the incidence of harmful algal blooms in estuaries is a major concern given their importance as a nursery area for a variety of invertebrates and fishes. On the warm-temperate, south-east coast of South Africa, the Sundays Estuary experiences persistent, seasonal HABs due to stable eutrophic conditions. Spring blooms of the dinoflagellate, Heterosigma akashiwo, were recorded for several years in the estuary. This Raphidophyte is known for its consequences in fish and shellfish aquaculture and exhibits harmful properties such as mucilage and reactive oxygen species production. Subsequently, the dissertation investigated larval fish assemblage dynamics, larval prey and predator dynamics and body condition in association with H. akashiwo blooms. HABs mainly altered the water quality within the mesohaline zone in terms of microalgal biomass and dissolved oxygen. Hypereutrophic blooms of H. akashiwo reached 24 996 cells∙mL-1 during the study period. Supersaturated dissolved oxygen concentrations (up to ~13 mg∙L-1 at night and ~21 mg∙L-1 during the day) were recorded during hypereutrophic blooms (> 80 μg∙L-1, ≥ 2781 H. akashiwo cells∙mL-1). Dissolved oxygen differed significantly by bloom state, with the highest dissolved oxygen concentrations. Larval fish density, diversity and community structure variability were lower during hypereutrophic blooms compared to standard blooms and bloom absence, albeit statistically insignificant. Estuarine resident larval density was negatively correlated with dissolved oxygen concentration (P < 0.001). Dominant prey copepod of G. aestuaria larvae, P. hessei, was less abundant and variable during hypereutrophic blooms (mean: 664, range: 4-4301 individuals per m3) compared to standard blooms (mean: 950, range: 2-5237 individuals per m3) and bloom absence (mean: 1182, range: 23-7259 individuals per m3). Although HABs did not appear to drive a mismatch between prey and G. aestuaria larvae, early juvenile larvae made only a small contribution to the assemblage (0.2-1.8%). In terms of predator effects, yolk sac larvae were negatively related to R. terranatalis (P < 0.05). A significant positive relationship was found between prey, P. hessei and G. aestuaria preflexion (P < 0.001), flexion (P < 0.001) and postflexion (P < 0.05) larvae. In a world first, the effects of HABs were investigated on a larval fish species, Gilchristella aestuaria, using biochemical body condition analyses. During severe bloom conditions, G. aestuaria larvae were smaller and the assemblage consisted of mainly early developmental stages (stats). Body condition (RNA per dry weight) was poorest in the presence of hypereutrophic blooms (P < 0.001). A bell-curved relationship between RNA/DW and dissolved oxygen concentration was also produced in a 3-day time-lagged GAM (P < 0.001). In conclusion, hypereutrophic bloom conditions appeared to trigger adult G. aestuaria spawning in the upper mesohaline zone of the estuary in response to heightened primary productivity which occurs during the early bloom stage. However, reductions in copepod abundance (P. hessei) and unfavourable oxygen concentrations interfered with grow-out of G. aestuaria larvae. This research showed that HABs can alter nursery properties to the detriment of G. aestuaria larvae can have serious knock-on effects up the food chain. The loss in abundance of adult, juvenile and larval G. aestuaria, a key forage fish in South African estuaries, has major implications for estuarine food webs, especially in reduced prey availability to juvenile fishes and birds. There is still much to learn about the effects of HABs on the physiological and behavioural aspects on plankton communities worldwide.
- Full Text:
- Date Issued: 2020
The potential of Raman spectroscopy in distinguishing between wool and mohair fibres
- Authors: Notayi, Mzwamadoda
- Date: 2020
- Subjects: Textile fabrics , Textile fibers -- Mechanical properties Wool -- Dissertations Mohair -- Dissertations
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/49248 , vital:41614
- Description: The possible application of the FT Raman, Raman micro-spectroscopy and ATR-FTIR micro-spectroscopy, have been investigated for distinguishing between wool and mohair. Highly identical Raman and FTIR spectra were obtained from the two fibre types, indicating that indeed they share similar basic molecular structural chemistry. The analysis of the amide I through curve fitting of wool and mohair FT Raman spectra showed that the protein and polypeptide secondary structure exists mainly in the α-helical structural conformation with smaller proportions of β-pleated sheet and β-Turns. These proportions, however, could not be used to distinguish between wool and mohair, due to the significant overlap observed between the two fibres. This study also determined the disulphide contents for possibly distinguishing between wool and mohair fibres, with the average and standard deviation values of 0.20±0.04 and 0.17±0.03 for wool and mohair, respectively, being found. Despite the mean values being found to differ statistically significant (p<0.05), a considerable overlap was observed, posing a doubt in the possible application of the method for distinguishing between the two fibres and blend composition analysis of the two fibres. The application of ratiometric analysis, based on the relative peak heights of certain FT Raman bands, showed that a combination of ratios A (I2932/I1450) and D (I508/I1450) could hold great potential in distinguishing between wool and mohair fibre samples. The individual values of ratios A and D varied a great deal from one mohair sample to the other and even more from one wool sample to another, with the individual values for ratio A ranging from 2.71-3.68 and 2.35-3.08 for wool and mohair, respectively, while ratio D ranged from 0.18-0.32 and 0.17-0.22 for wool and mohair, respectively. An important observation from this study is that if, for an unknown sample, if individual values of ratios A and D exceed 3.1 and 0.22, respectively, are found then the sample is most likely to be either a pure wool or blend of wool and mohair, whereas if all the values fall below the two threshold values, then the unknown sample can be declared a pure mohair sample. A Raman spectral database or library of approximately 100 high quality Raman average spectra of wool and mohair fibres has been established for the Bruker 80V FTIR/Raman spectrophotometer at the Nelson Mandela University (NMU). Although this has not been fully validated due to the unforeseen frequent breakdown encountered with the FT Raman system, at this stage, it has been realized that verification of unknown materials is highly possible. A great need for the development of a classification model based on multivariate or chemometrics has been realized. An ATR-FTIR LUMOS micro-spectroscopic system was also investigated for the possible application in distinguishing between wool and mohair single fibres. The amide I/II band ratios were determined for both wool and mohair fibres to distinguish between the two fibre types. The mean and standard deviation values of 1.20±0.02 and 1.21±0.01 for mohair and wool, respectively, were found and were shown not to differ statistically significant (p˃0.05). The secondary structure analysis showed that the content of the α-helical secondary structure might be different between the two fibre types, with a great overlap of individual values, however, being observed between the two fibre types (wool and mohair), raising concerns in the possible application of the α-helical content for distinguishing the two fibres.
- Full Text:
- Date Issued: 2020
- Authors: Notayi, Mzwamadoda
- Date: 2020
- Subjects: Textile fabrics , Textile fibers -- Mechanical properties Wool -- Dissertations Mohair -- Dissertations
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/49248 , vital:41614
- Description: The possible application of the FT Raman, Raman micro-spectroscopy and ATR-FTIR micro-spectroscopy, have been investigated for distinguishing between wool and mohair. Highly identical Raman and FTIR spectra were obtained from the two fibre types, indicating that indeed they share similar basic molecular structural chemistry. The analysis of the amide I through curve fitting of wool and mohair FT Raman spectra showed that the protein and polypeptide secondary structure exists mainly in the α-helical structural conformation with smaller proportions of β-pleated sheet and β-Turns. These proportions, however, could not be used to distinguish between wool and mohair, due to the significant overlap observed between the two fibres. This study also determined the disulphide contents for possibly distinguishing between wool and mohair fibres, with the average and standard deviation values of 0.20±0.04 and 0.17±0.03 for wool and mohair, respectively, being found. Despite the mean values being found to differ statistically significant (p<0.05), a considerable overlap was observed, posing a doubt in the possible application of the method for distinguishing between the two fibres and blend composition analysis of the two fibres. The application of ratiometric analysis, based on the relative peak heights of certain FT Raman bands, showed that a combination of ratios A (I2932/I1450) and D (I508/I1450) could hold great potential in distinguishing between wool and mohair fibre samples. The individual values of ratios A and D varied a great deal from one mohair sample to the other and even more from one wool sample to another, with the individual values for ratio A ranging from 2.71-3.68 and 2.35-3.08 for wool and mohair, respectively, while ratio D ranged from 0.18-0.32 and 0.17-0.22 for wool and mohair, respectively. An important observation from this study is that if, for an unknown sample, if individual values of ratios A and D exceed 3.1 and 0.22, respectively, are found then the sample is most likely to be either a pure wool or blend of wool and mohair, whereas if all the values fall below the two threshold values, then the unknown sample can be declared a pure mohair sample. A Raman spectral database or library of approximately 100 high quality Raman average spectra of wool and mohair fibres has been established for the Bruker 80V FTIR/Raman spectrophotometer at the Nelson Mandela University (NMU). Although this has not been fully validated due to the unforeseen frequent breakdown encountered with the FT Raman system, at this stage, it has been realized that verification of unknown materials is highly possible. A great need for the development of a classification model based on multivariate or chemometrics has been realized. An ATR-FTIR LUMOS micro-spectroscopic system was also investigated for the possible application in distinguishing between wool and mohair single fibres. The amide I/II band ratios were determined for both wool and mohair fibres to distinguish between the two fibre types. The mean and standard deviation values of 1.20±0.02 and 1.21±0.01 for mohair and wool, respectively, were found and were shown not to differ statistically significant (p˃0.05). The secondary structure analysis showed that the content of the α-helical secondary structure might be different between the two fibre types, with a great overlap of individual values, however, being observed between the two fibre types (wool and mohair), raising concerns in the possible application of the α-helical content for distinguishing the two fibres.
- Full Text:
- Date Issued: 2020
Water balance processes in indigenous and introduced spaces three production system in the Southern Cape region of South Africa
- Authors: Mapeto, Tatenda
- Date: 2020
- Subjects: Hydrology -- Research -- South Africa , Groundwater -- Research -- South Africa Trees -- Breeding Sustainable forestry
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/49325 , vital:41621
- Description: Forests and trees are important modulators of water flow and their position in the landscape impacts the spatial and temporal availability of ground water and surface water resources. As such, the global extent of natural forests is linked to the availability of water. In the circumstances where tree production systems have been introduced and are managed in landscapes previously populated with other forms of vegetation, it is necessary to understand their hydrological impact so as to ensure that the benefits derived from these tree systems do not outweigh the costs of water availability for sustaining human and environmental water requirements.
- Full Text: false
- Date Issued: 2020
- Authors: Mapeto, Tatenda
- Date: 2020
- Subjects: Hydrology -- Research -- South Africa , Groundwater -- Research -- South Africa Trees -- Breeding Sustainable forestry
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/49325 , vital:41621
- Description: Forests and trees are important modulators of water flow and their position in the landscape impacts the spatial and temporal availability of ground water and surface water resources. As such, the global extent of natural forests is linked to the availability of water. In the circumstances where tree production systems have been introduced and are managed in landscapes previously populated with other forms of vegetation, it is necessary to understand their hydrological impact so as to ensure that the benefits derived from these tree systems do not outweigh the costs of water availability for sustaining human and environmental water requirements.
- Full Text: false
- Date Issued: 2020
A framework to measure human behaviour whilst reading
- Salehzadeh, Seyed Amirsaleh, Greyling, Jean
- Authors: Salehzadeh, Seyed Amirsaleh , Greyling, Jean
- Date: 2019
- Subjects: Computational intelligence , Machine learning Artificial intelligence Neural networks (Computer science)
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/43578 , vital:36921
- Description: The brain is the most complex object in the known universe that gives a sense of being to humans and characterises human behaviour. Building models of brain functions is perhaps the most fascinating scientific challenge in the 21st century. Reading is a significant cognitive process in the human brain that plays a critical role in the vital process of learning and in performing some daily activities. The study of human behaviour during reading has been an area of interest for researchers in different fields of science. This thesis is based upon providing a novel framework, called ARSAT (Assisting Researchers in the Selection of Appropriate Technologies), that measures the behaviour of humans when reading text. The ARSAT framework aims at assisting researchers in the selection and application of appropriate technologies to measure the behaviour of a person who is reading text. The ARSAT framework will assist to researchers who investigate the reading process and find it difficult to select appropriate theories, metrics, data collection methods and data analytics techniques. The ARSAT framework enhances the ability of its users to select appropriate metrics indicating the effective factors on the characterisation of different aspects of human behaviour during the reading process. As will be shown in this research study, human behaviour is characterised by a complicated interplay of action, cognition and emotion. The ARSAT framework also facilitates selecting appropriate sensory technologies that can be used to monitor and collect data for the metrics. Moreover, this research study will introduce BehaveNet, a novel Deep Learning modelling approach, which can be used for training Deep Learning models of human behaviour from the sensory data collected. In this thesis, a comprehensive literature study is presented that was conducted to acquire adequate knowledge for designing the ARSAT framework. In order to identify the contributing factors that affect the reading process, an overview of some existing theories of the reading process is provided. Furthermore, a number of sensory technologies and techniques that can be applied to monitoring the changes in the metrics indicating the factors are also demonstrated. Only, the technologies that are commercially available on the market are recommended by the ARSAT framework. A variety of Machine Learning techniques were also investigated when designing the BehaveNet. The BehaveNet takes advantage of the complementarity of Convolutional Neural Networks, Long Short-Term Memory networks and Deep Neural Networks. The design of a Human Behaviour Monitoring System (HBMS), by utilising the ARSAT framework for recognising three attention-seeking activities of humans, is also presented in this research study. Reading printed text, as well as speaking out loudly and watching a programme on TV were proposed as activities that a person unintentionally may shift his/her attention from reading into distractions. Between sensory devices recommended by the ARSAT framework, the Muse headband which is an Electroencephalography (EEG) and head motion-sensing wearable device, was selected to track the forehead EEG and a person’s head movements. The EEG and 3-axes accelerometer data were recorded from eight participants when they read printed text, as well as the time they performed two other activities. An imbalanced dataset consisting over 1.2 million rows of noisy data was created and used to build a model of the activities (60% training and 20% validating data) and evaluating the model (20% of the data). The efficiency of the framework is demonstrated by comparing the performance of the models built by utilising the BehaveNet, with the models built by utilising a number of competing Deep Learning models for raw EEG and accelerometer data, that have attained state-of-the-art performance. The classification results are evaluated by some metrics including the classification accuracy, F1 score, confusion matrix, Receiver Operating Characteristic curve, and Area under Curve (AUC) score. By considering the results, the BehaveNet contributed to the body of knowledge as an approach for measuring human behaviour by using sensory devices. In comparison with the performance of the other models, the models built by utilising the BehaveNet, attained better performance when classifying data of two EEG channels (Accuracy = 95%; AUC=0.99; F1 = 0.95), data of a single EEG channel (Accuracy = 85%; AUC=0.96; F1 = 0.83), accelerometer data (Accuracy = 81%; AUC = 0.9; F1 = 0.76) and all of the data in the dataset (Accuracy = 97%; AUC = 0.99; F1 = 0.96). The dataset and the source code of this project are also published on the Internet to help the science community. The Muse headband is also shown to be an economical and standard wearable device that can be successfully used in behavioural research.
- Full Text:
- Date Issued: 2019
- Authors: Salehzadeh, Seyed Amirsaleh , Greyling, Jean
- Date: 2019
- Subjects: Computational intelligence , Machine learning Artificial intelligence Neural networks (Computer science)
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/43578 , vital:36921
- Description: The brain is the most complex object in the known universe that gives a sense of being to humans and characterises human behaviour. Building models of brain functions is perhaps the most fascinating scientific challenge in the 21st century. Reading is a significant cognitive process in the human brain that plays a critical role in the vital process of learning and in performing some daily activities. The study of human behaviour during reading has been an area of interest for researchers in different fields of science. This thesis is based upon providing a novel framework, called ARSAT (Assisting Researchers in the Selection of Appropriate Technologies), that measures the behaviour of humans when reading text. The ARSAT framework aims at assisting researchers in the selection and application of appropriate technologies to measure the behaviour of a person who is reading text. The ARSAT framework will assist to researchers who investigate the reading process and find it difficult to select appropriate theories, metrics, data collection methods and data analytics techniques. The ARSAT framework enhances the ability of its users to select appropriate metrics indicating the effective factors on the characterisation of different aspects of human behaviour during the reading process. As will be shown in this research study, human behaviour is characterised by a complicated interplay of action, cognition and emotion. The ARSAT framework also facilitates selecting appropriate sensory technologies that can be used to monitor and collect data for the metrics. Moreover, this research study will introduce BehaveNet, a novel Deep Learning modelling approach, which can be used for training Deep Learning models of human behaviour from the sensory data collected. In this thesis, a comprehensive literature study is presented that was conducted to acquire adequate knowledge for designing the ARSAT framework. In order to identify the contributing factors that affect the reading process, an overview of some existing theories of the reading process is provided. Furthermore, a number of sensory technologies and techniques that can be applied to monitoring the changes in the metrics indicating the factors are also demonstrated. Only, the technologies that are commercially available on the market are recommended by the ARSAT framework. A variety of Machine Learning techniques were also investigated when designing the BehaveNet. The BehaveNet takes advantage of the complementarity of Convolutional Neural Networks, Long Short-Term Memory networks and Deep Neural Networks. The design of a Human Behaviour Monitoring System (HBMS), by utilising the ARSAT framework for recognising three attention-seeking activities of humans, is also presented in this research study. Reading printed text, as well as speaking out loudly and watching a programme on TV were proposed as activities that a person unintentionally may shift his/her attention from reading into distractions. Between sensory devices recommended by the ARSAT framework, the Muse headband which is an Electroencephalography (EEG) and head motion-sensing wearable device, was selected to track the forehead EEG and a person’s head movements. The EEG and 3-axes accelerometer data were recorded from eight participants when they read printed text, as well as the time they performed two other activities. An imbalanced dataset consisting over 1.2 million rows of noisy data was created and used to build a model of the activities (60% training and 20% validating data) and evaluating the model (20% of the data). The efficiency of the framework is demonstrated by comparing the performance of the models built by utilising the BehaveNet, with the models built by utilising a number of competing Deep Learning models for raw EEG and accelerometer data, that have attained state-of-the-art performance. The classification results are evaluated by some metrics including the classification accuracy, F1 score, confusion matrix, Receiver Operating Characteristic curve, and Area under Curve (AUC) score. By considering the results, the BehaveNet contributed to the body of knowledge as an approach for measuring human behaviour by using sensory devices. In comparison with the performance of the other models, the models built by utilising the BehaveNet, attained better performance when classifying data of two EEG channels (Accuracy = 95%; AUC=0.99; F1 = 0.95), data of a single EEG channel (Accuracy = 85%; AUC=0.96; F1 = 0.83), accelerometer data (Accuracy = 81%; AUC = 0.9; F1 = 0.76) and all of the data in the dataset (Accuracy = 97%; AUC = 0.99; F1 = 0.96). The dataset and the source code of this project are also published on the Internet to help the science community. The Muse headband is also shown to be an economical and standard wearable device that can be successfully used in behavioural research.
- Full Text:
- Date Issued: 2019
A multi-factor model for range estimation in electric vehicles
- Authors: Smuts, Martin Bradley
- Date: 2019
- Subjects: Electric vehicles , Hybrid electric vehicles Energy consumption Machine learning Information technology -- Management
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/43589 , vital:36926
- Description: Electric vehicles (EVs) are well-known for their challenges related to trip planning and energy consumption estimation. Range anxiety is currently a barrier to the adoption of EVs. One of the issues influencing range anxiety is the inaccuracy of the remaining driving range (RDR) estimate in on-board displays. RDR displays are important as they can help drivers with trip planning. The RDR is a parameter that changes under environmental and behavioural conditions. Several factors (for example, weather, and traffic) can influence the energy consumption of an EV that are not considered during the RDR estimation in traditional on-board computers or third-party applications, such as navigation or mapping applications. The need for accurate RDR estimation is growing, since this can reduce the range anxiety of drivers. One way of overcoming range anxiety is to provide trip planning applications that provide accurate estimations of the RDR, based on various factors, and which adapt to the users’ driving behaviour. Existing models used for estimating the RDR are often simplified, and do not consider all the factors that can influence it. Collecting data for each factor also presents several challenges. Powerful computing resources are required to collect, transform, and analyse the disparate datasets that are required for each factor. The aim of this research was to design a Multi-factor Model for range estimation in EVs. Five main factors that influence the energy consumption of EVs were identified from literature, namely, Route and Terrain, Driving Behaviour, Weather and Environment, Vehicle Modelling, and Battery Modelling. These factors were used throughout this research to guide the data collection and analysis processes. A Multi-factor Model was proposed based on four main components that collect, process, analyse, and visualise data from available data sources to produce estimates relating to trip planning. A proof-of-concept RDR system was developed and evaluated in field experiments, to demonstrate that the Multi-factor Model addresses the main aim of this research. The experiments were performed to collect data for each of the five factors, and to analyse their impact on energy consumption. Several machine learning techniques were used, and evaluated, for accuracy in estimating the energy consumption, from which the RDR can be derived, for a specified trip. A case study was conducted with an electric mobility programme (uYilo) in Port Elizabeth, South Africa (SA). The case study was used to investigate whether the available resources at uYilo were sufficient to provide data for each of the five factors. Several challenges were noted during the data collection. These were shortages of software applications, a lack of quality data, technical interoperability and data access between the data collection instruments and systems. Data access was a problem in some cases, since proprietary systems restrict access to external developers. The theoretical contribution of this research is a list of factors that influence RDR and a classification of machine learning techniques that can be used to estimate the RDR. The practical contributions of this research include a database of EV trips, proof-of-concept RDR estimation system, and a deployed machine learning model that can be accessed by researchers and EV practitioners. Four research papers were published and presented at local and international conferences. In addition, one conference paper was published in an accredited journal: NextComp 2017 (Appendix C), Conference Paper, Pointe aux Piments (Mauritius); SATNAC 2017 (Appendix F), Conference Paper, Barcelona (Spain); GITMA 2018 (Appendix B), Conference Paper, Mexico City (Mexico); SATNAC 2018 (Appendix G), Conference Paper, George (South Africa), and IFIP World Computer Congress 2018 (Appendix E), Journal Article.
- Full Text:
- Date Issued: 2019
- Authors: Smuts, Martin Bradley
- Date: 2019
- Subjects: Electric vehicles , Hybrid electric vehicles Energy consumption Machine learning Information technology -- Management
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/43589 , vital:36926
- Description: Electric vehicles (EVs) are well-known for their challenges related to trip planning and energy consumption estimation. Range anxiety is currently a barrier to the adoption of EVs. One of the issues influencing range anxiety is the inaccuracy of the remaining driving range (RDR) estimate in on-board displays. RDR displays are important as they can help drivers with trip planning. The RDR is a parameter that changes under environmental and behavioural conditions. Several factors (for example, weather, and traffic) can influence the energy consumption of an EV that are not considered during the RDR estimation in traditional on-board computers or third-party applications, such as navigation or mapping applications. The need for accurate RDR estimation is growing, since this can reduce the range anxiety of drivers. One way of overcoming range anxiety is to provide trip planning applications that provide accurate estimations of the RDR, based on various factors, and which adapt to the users’ driving behaviour. Existing models used for estimating the RDR are often simplified, and do not consider all the factors that can influence it. Collecting data for each factor also presents several challenges. Powerful computing resources are required to collect, transform, and analyse the disparate datasets that are required for each factor. The aim of this research was to design a Multi-factor Model for range estimation in EVs. Five main factors that influence the energy consumption of EVs were identified from literature, namely, Route and Terrain, Driving Behaviour, Weather and Environment, Vehicle Modelling, and Battery Modelling. These factors were used throughout this research to guide the data collection and analysis processes. A Multi-factor Model was proposed based on four main components that collect, process, analyse, and visualise data from available data sources to produce estimates relating to trip planning. A proof-of-concept RDR system was developed and evaluated in field experiments, to demonstrate that the Multi-factor Model addresses the main aim of this research. The experiments were performed to collect data for each of the five factors, and to analyse their impact on energy consumption. Several machine learning techniques were used, and evaluated, for accuracy in estimating the energy consumption, from which the RDR can be derived, for a specified trip. A case study was conducted with an electric mobility programme (uYilo) in Port Elizabeth, South Africa (SA). The case study was used to investigate whether the available resources at uYilo were sufficient to provide data for each of the five factors. Several challenges were noted during the data collection. These were shortages of software applications, a lack of quality data, technical interoperability and data access between the data collection instruments and systems. Data access was a problem in some cases, since proprietary systems restrict access to external developers. The theoretical contribution of this research is a list of factors that influence RDR and a classification of machine learning techniques that can be used to estimate the RDR. The practical contributions of this research include a database of EV trips, proof-of-concept RDR estimation system, and a deployed machine learning model that can be accessed by researchers and EV practitioners. Four research papers were published and presented at local and international conferences. In addition, one conference paper was published in an accredited journal: NextComp 2017 (Appendix C), Conference Paper, Pointe aux Piments (Mauritius); SATNAC 2017 (Appendix F), Conference Paper, Barcelona (Spain); GITMA 2018 (Appendix B), Conference Paper, Mexico City (Mexico); SATNAC 2018 (Appendix G), Conference Paper, George (South Africa), and IFIP World Computer Congress 2018 (Appendix E), Journal Article.
- Full Text:
- Date Issued: 2019
An investigation of plant conservation strategies employed in Makhanda, South Africa: an educational exploration
- Authors: Cimi, Pumlani Viwe
- Date: 2019
- Subjects: Plant conservation , Rare plants Biodiversity conservation
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/37979 , vital:34276
- Description: The school curriculum in South Africa gives educators the freedom to design and organise learning experiences according to their local circumstances and availability of resources. In that context, this study advocates conservation education through use of plants on school grounds and visits to local herbaria and botanical gardens as part of the school curriculum. Trees growing on street verges in town suburbs are also a resource that could be using for teaching. School learners from three schools representing different areas of Makhanda (Grahamstown) participated in plant-related activities and were surveyed to determine their prior knowledge about plants and plant conservation. Learners answered a series of questions in writing using a funnel sequence and the inverted funnel sequence in questionnaire design. This motivated respondents to co-operate and fully complete a questionnaire. In addition, the non-participant observation technique was used to capture behavioural reactions to the activities in order to supplement the data generated through questionnaires and interviews. This proved to be a purposeful, systematic and a selective way of watching and listening to an interaction phenomenon as it takes place. Interviews and questionnaires revealed that most Makhanda residents (especially in rural and township areas) use traditional medicine and medicinal plants to treat many diseases. They also use plants for cultural activities. Traditional remedies are practised among the rural and township communities because of ease of availability, convenience, and also due to social, psychological and cultural reasons. Medicinal plants have been increasingly recognized for their role not only for health care, but also for improving the economic status. Community members and street vendors were also questioned using semi-structured interviews. Open-ended questions were used successfully to assess the person’s knowledge, attitudes, opinions, beliefs and feelings. This type of interview ensures that a specific question does not lose its purpose. These interviews were used to determine how community members see, interpret and relate to nature. In the interviews, I also assessed the real needs of the local communities because there is a tendency to think that our power, knowledge and resources can give us the right to predict what communities really need without actually asking them. When our imposed initiatives fail, we often accuse these communities of a lack of interest in our issue of interest. Documents on plants and their uses kept in the Selmar Schönland Herbarium were compared with the information given by community members and street vendors. Xhosa plant names and ethnobotanical information given by the community members and street vendors was used to augment existing information on plant use. A further aim of this study was to investigate the potential use of plant resources provided by the street trees and gardens for education. Makhanda streets were sampled and, of the 1 435 plants that were countered from 17 streets, only 15 indigenous species were represented, while 20 alien species were recorded. These trees can be used for lessons on both alien and indigenous plants. Notable differences in tree density and species richness were evident across suburbs with the highest density and richness found in the more affluent suburbs and poor representation of trees in the township. This will affect learner perceptions and viewpoints. The results from the analysed data revealed that use of the school grounds, botanical gardens, the local herbarium and town streets are effective in plant conservation education in schools and communities. These types of resources could be used for environmental education for future generations in South Africa and all over the world.
- Full Text:
- Date Issued: 2019
- Authors: Cimi, Pumlani Viwe
- Date: 2019
- Subjects: Plant conservation , Rare plants Biodiversity conservation
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/37979 , vital:34276
- Description: The school curriculum in South Africa gives educators the freedom to design and organise learning experiences according to their local circumstances and availability of resources. In that context, this study advocates conservation education through use of plants on school grounds and visits to local herbaria and botanical gardens as part of the school curriculum. Trees growing on street verges in town suburbs are also a resource that could be using for teaching. School learners from three schools representing different areas of Makhanda (Grahamstown) participated in plant-related activities and were surveyed to determine their prior knowledge about plants and plant conservation. Learners answered a series of questions in writing using a funnel sequence and the inverted funnel sequence in questionnaire design. This motivated respondents to co-operate and fully complete a questionnaire. In addition, the non-participant observation technique was used to capture behavioural reactions to the activities in order to supplement the data generated through questionnaires and interviews. This proved to be a purposeful, systematic and a selective way of watching and listening to an interaction phenomenon as it takes place. Interviews and questionnaires revealed that most Makhanda residents (especially in rural and township areas) use traditional medicine and medicinal plants to treat many diseases. They also use plants for cultural activities. Traditional remedies are practised among the rural and township communities because of ease of availability, convenience, and also due to social, psychological and cultural reasons. Medicinal plants have been increasingly recognized for their role not only for health care, but also for improving the economic status. Community members and street vendors were also questioned using semi-structured interviews. Open-ended questions were used successfully to assess the person’s knowledge, attitudes, opinions, beliefs and feelings. This type of interview ensures that a specific question does not lose its purpose. These interviews were used to determine how community members see, interpret and relate to nature. In the interviews, I also assessed the real needs of the local communities because there is a tendency to think that our power, knowledge and resources can give us the right to predict what communities really need without actually asking them. When our imposed initiatives fail, we often accuse these communities of a lack of interest in our issue of interest. Documents on plants and their uses kept in the Selmar Schönland Herbarium were compared with the information given by community members and street vendors. Xhosa plant names and ethnobotanical information given by the community members and street vendors was used to augment existing information on plant use. A further aim of this study was to investigate the potential use of plant resources provided by the street trees and gardens for education. Makhanda streets were sampled and, of the 1 435 plants that were countered from 17 streets, only 15 indigenous species were represented, while 20 alien species were recorded. These trees can be used for lessons on both alien and indigenous plants. Notable differences in tree density and species richness were evident across suburbs with the highest density and richness found in the more affluent suburbs and poor representation of trees in the township. This will affect learner perceptions and viewpoints. The results from the analysed data revealed that use of the school grounds, botanical gardens, the local herbarium and town streets are effective in plant conservation education in schools and communities. These types of resources could be used for environmental education for future generations in South Africa and all over the world.
- Full Text:
- Date Issued: 2019
Behavioural flexibility in an endangered seabird during current changes
- Authors: Traisnel, Gwendoline
- Date: 2019
- Subjects: Penguins -- South Africa , Penguins -- Behavior Sea birds -- Behavior Sea birds -- Ecology
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/43800 , vital:37048
- Description: Penguins spend a large part of their life cycle at sea and are amongst the most threatened seabirds as they suffer high mortality and have low reproductive success. Across the world, many populations are currently declining despite the constant development of conservation actions. The lack of information at the individual level reduces the understanding of individual variability and its potential role in the population dynamics of these vulnerable species. For example, long-lived species are characterised by substantial individual heterogeneity in their contribution to the overall population dynamic that should be understood when considering the implemention of conservation policies. This thesis explored some of the primary mechanisms behind inter-individual differences in behaviour and breeding success in the endangered African penguin Spheniscus demersus. On Bird island (Algoa Bay), nest defence behaviours were investigated on adults brooding chicks 1-3 weeks-old and chicks were measured at regular intervals (every 5 days) to assess their growth rate, a proxy for breeding success in this species. All adults were sexed and implanted with a passive integrated transponder which was used for individual identification. In this species, males are usually larger than females and have a higher survival rates both as juveniles and as adults. A subsample of them was equipped with tracking devices (GPSs and Time Depth Recorders), sometimes over consecutive foraging trips and across years. Finally, mate and nest fidelity were investigated to understand the drivers and consequences of this trait in that population. Breeding success was linked to nest defence behaviours, with bolder birds generally showing lower breeding success. This relation could not, however, be explained by differences in foraging strategies. Males were generally bolder than females when defending the nests, but their foraging behaviour did not vary with their personality. They generally had lower foraging effort than females. By contrast, bolder females performed more sinuous path, more wiggles and travelled larger vertical distances than shyer ones. Overall, females were more flexible than males over consecutive trips, suggesting that they probably adjust their behaviours to the offspring needs in this species. However, no sex difference in behavioural flexibility between sexes was visible across years, indicating that environmental variability may equally impact females and males’ behaviour over such time-scale. Generally, consistency was time-scale dependent as penguins increased their foraging flexibility across years, potentially adjusting to the variability of the environment. Interestingly, foraging consistency over consecutive trips increased during years of poor environmental conditions, and individuals with consistent foraging strategies were more successful than more flexible individuals when resources were extremely scarce. Finally, while nest and mate fidelity were high in this species, individuals that changed nest/partners, improved their breeding outcome. Particularly, after low breeding success females were more likely to change nest site and thereby partner to possibly improve their fitness. The potential biased adult sex ratio towards males in this population may facilitate changes in females which may have more partners available to re-mate than males. The present thesis reveals the existence of individual differences in behaviour and breeding success which relate to personality in penguins. These findings highlight the importance to integrate individual variability to predict future population dynamics in the context of global changes to understand the resilience and vulnerability of the species.
- Full Text:
- Date Issued: 2019
- Authors: Traisnel, Gwendoline
- Date: 2019
- Subjects: Penguins -- South Africa , Penguins -- Behavior Sea birds -- Behavior Sea birds -- Ecology
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/43800 , vital:37048
- Description: Penguins spend a large part of their life cycle at sea and are amongst the most threatened seabirds as they suffer high mortality and have low reproductive success. Across the world, many populations are currently declining despite the constant development of conservation actions. The lack of information at the individual level reduces the understanding of individual variability and its potential role in the population dynamics of these vulnerable species. For example, long-lived species are characterised by substantial individual heterogeneity in their contribution to the overall population dynamic that should be understood when considering the implemention of conservation policies. This thesis explored some of the primary mechanisms behind inter-individual differences in behaviour and breeding success in the endangered African penguin Spheniscus demersus. On Bird island (Algoa Bay), nest defence behaviours were investigated on adults brooding chicks 1-3 weeks-old and chicks were measured at regular intervals (every 5 days) to assess their growth rate, a proxy for breeding success in this species. All adults were sexed and implanted with a passive integrated transponder which was used for individual identification. In this species, males are usually larger than females and have a higher survival rates both as juveniles and as adults. A subsample of them was equipped with tracking devices (GPSs and Time Depth Recorders), sometimes over consecutive foraging trips and across years. Finally, mate and nest fidelity were investigated to understand the drivers and consequences of this trait in that population. Breeding success was linked to nest defence behaviours, with bolder birds generally showing lower breeding success. This relation could not, however, be explained by differences in foraging strategies. Males were generally bolder than females when defending the nests, but their foraging behaviour did not vary with their personality. They generally had lower foraging effort than females. By contrast, bolder females performed more sinuous path, more wiggles and travelled larger vertical distances than shyer ones. Overall, females were more flexible than males over consecutive trips, suggesting that they probably adjust their behaviours to the offspring needs in this species. However, no sex difference in behavioural flexibility between sexes was visible across years, indicating that environmental variability may equally impact females and males’ behaviour over such time-scale. Generally, consistency was time-scale dependent as penguins increased their foraging flexibility across years, potentially adjusting to the variability of the environment. Interestingly, foraging consistency over consecutive trips increased during years of poor environmental conditions, and individuals with consistent foraging strategies were more successful than more flexible individuals when resources were extremely scarce. Finally, while nest and mate fidelity were high in this species, individuals that changed nest/partners, improved their breeding outcome. Particularly, after low breeding success females were more likely to change nest site and thereby partner to possibly improve their fitness. The potential biased adult sex ratio towards males in this population may facilitate changes in females which may have more partners available to re-mate than males. The present thesis reveals the existence of individual differences in behaviour and breeding success which relate to personality in penguins. These findings highlight the importance to integrate individual variability to predict future population dynamics in the context of global changes to understand the resilience and vulnerability of the species.
- Full Text:
- Date Issued: 2019
Design of metal ion-selective reagents for recovery of precious metals
- Authors: Moleko-Boyce, Pulleng
- Date: 2019
- Subjects: Bioinorganic chemistry , Metal complexes Speciation (Chemistry)
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/42510 , vital:36664
- Description: The study is divided into two sections; namely, (1) the design of rhodium(III) specific chelating ligands (tridentate bis-benzimidazole derivatives), and (2) the development of iridium(IV)-specific quaternary diammonium cations with electron donating and electron withdrawing groups. Bis-benzimidazole chelating ligands used were bis((1H-benzimidazol-2-yl)methyl)amine (NNN1), bis((1H-benzimidazol-2-yl)ethyl)amine (NNN2), bis((1H-benzimidazol-2-yl)methyl)sulfide (NSN1) and bis((1H-benzimidazol-2-yl)ethyl)sulfide (NSN2). Quaternary diammonium cations used were tetramethylbenzyl-1,10-diammonium chloride (QuatDMDAMeBnz), tetrabenzyl-1,10-diammonium chloride (QuatDMDABnz), tetratrifluoromethylbenzyl-1,10-diammonium chloride (QuatDMDACF3Bnz) and tetranitrobenzyl-1,10-diammonium chloride (QuatDMDANO2Bnz). For both studies, polyvinylbenzylchloride (PVBC) nanofibers were used as support material. The PVBC nanofibers which were functionalised with bis-benzimidazole derivatives and quaternary diammonium cations, respectively, were investigated for the selectivity for Rh(III) over Ir(III), Pt(II), Pd(II) and Ni(II), and for separation of Ir(IV) from Rh(III), respectively. The sorbent materials were characterised by FTIR, SEM, BET surface area, TGA, EDS and elemental analysis, and the results showed that the functionalization of the sorbent materials was successful.The efficiency of bis-benzimidazole derivatives and quaternary diammonium cations, respectively, were investigated in a column study under dynamic flow adsorption conditions. The adsorption kinetics and isotherms were investigated under batch conditions and fitted on pseudo-first-order and pseudo-second-order model, and Freundlich and Langmuir isotherm, respectively. It was observed that the bis-benzimidazole derivatives showed uptake of [RhCl3(H2O)3], and the loading capacities were observed in the following order; NSN1 (181.06 mg/g) > NSN2 (148.55 mg/g) > NNN1 (131.88 mg/g) > NNN2 (75.87 mg/g). The bis-benzimidazole derivatives preference for metal ions was further investigated with a multi-element solution containing Rh(III), Ir(III), Pt(II), Pd(II) and Ni(II). The bis-benzimidazole derivatives showed the following order of loading capacity: NSN1 (47.28 mg/g) > NSN2 (23.89 mg/g) > NNN1 (17.47 mg/g) > NNN2 (14.91 mg/g) for Rh(III); NSN2 (10.64 mg/g) > NNN2 (6.84 mg/g) > NSN1 (5.74 mg/g) > NNN1 (5.02 mg/g) for Ir(III); NNN2 (33.96 mg/g) > NSN1 (30.95 mg/g) > NSN2 (19.95 mg/g) > NNN1 (14.92 mg/g) for Pt(II); NNN1 (47.94 mg/g) > NNN2 (28.90 mg/g) > NSN1 (16.22 mg/g) > NSN2 (15.83 mg/g) for Pd(II). Bis-benzimidazole derivatives showed no uptake of nickel(II) under these conditions. It was observed the ligand-selectivity order for Rh(III) was similar in both single-element and multi-element studies. This order showed that the bis-benzimidazoles containing a sulfur atom showed a high preference for rhodium(III) compared to Pt(II) which had a high preference for NNN2 as well as Pd(II) which had a high preference for NNN1. Ir(III) generally had a lower preference for the ligands presumably due to its higher kinetic inertness compared with Rh(III). Column sorption of [IrCl6]2- and [RhCl5(H2O)]2- on nanofibers functionalized with diammonium cations was carried out and the loading capacities of [IrCl6]2- were obtained. [RhCl5(H2O)]2- was not adsorbed by the sorbent materials while [IrCl6]2- was loaded onto the column. The loading capacities of [IrCl6]2- with the quaternary diammonium sorbent materials increased in the order of F-QuatDMDAMeBnz (60.29 mg/g) < F-QuatDMDABnz (67.61 mg/g) < F-QuatDMDACF3Bnz (107.59 mg/g) < F-QuatDMDANO2Bnz (140.47 mg/g). The loading capacity for Ir(IV) with quaternary diammonium cationic nanofibers increased with an increase in the electron-withdrawing nature of the quaternizing group. The charge delocalizing ability of the nitrobenzyl group resulted in the best interaction of the diammonium cation with [IrCl6]2-. Batch equilibrium studies were carried out to assess the efficiency of bis-benzimidazole chelating derivatives as adsorbents using a multi-metal solution (Rh(III), Ir(III), Pt(II), Pd(II) and Ni(II)) in 0.5 M HCl. The efficiency of the quaternary diammonium cations was tested using a binary metal solution (Ir(IV) and Rh(III)) in 6 M HCl. The isothermal batch adsorption studies of a multi-metal solution with bis-benzimidazoles derivatives fitted the Langmuir isotherm model which confirmed monolayer adsorption onto a homogeneous surface. The Langmuir isotherm parameter (qe (mg/g)), using functionalized nanofibers, showed the order of NNN2 (128.21 mg/g) > NSN1 (99.01 mg/g) > NSN2 (91.74 mg/g) > NNN1 (84.03 mg/g) for Pt(II); NNN1 (66.23 mg/g) > NNN2 (5.89 mg/g) > NSN1 (1.40 mg/g) > NSN2 (0.59 mg/g) for Pd(II); NSN2 (10.64 mg/g) > NNN2 (6.84 mg/g) > NSN1 (5.74 mg/g) > NNN1 (5.02 mg/g) for Ir(III); NSN1 (140.85 mg/g) > NSN2 (109.89 mg/g) > NNN1 (104.17 mg/g) > NNN2 (91.74 mg/g) for Rh(III). The pseudo-first-order kinetics model was found to be the best fit to describe the adsorption kinetics of all metal ions onto all the sorbent materials. K1 (min-1) value in pseudo-first-order kinetics showed the same order of adsorption as observed in the Langmuir isotherms. The isothermal batch adsorption studies of [IrCl6]2- and [RhCl5(H2O)]2- with quaternary diammonium cations fitted the Freundlich isotherm model and confirmed to be effective for multiple-layered adsorption onto a heterogeneous surface. The Freundlich isotherm parameter (kf (mg/g)) using functionalized quaternary diammonium cationic nanofibers increased in the order of F-QuatDMDANO2Bnz (794.33 mg/g) > F-QuatDMDACF3Bnz (185.35 mg/g) > F-QuatDMDABnz (156.32 mg/g) > F-QuatDMDAMeBnz (112.46 mg/g) for Ir(IV) uptake. F-QuatDMDANO2Bnz resin showed the highest adsorption than that of F-QuatDMDAMeBnz, F-QuatDMDABnz and F-QuatDMDACF3Bnz and this order is similar to what was observed in column studies. The quaternary diammonium cations were shown to have the highest adsorption capacity for Ir(IV) compared with Rh(III). The adsorption of Rh(III) was also observed to increase in the order of F-QuatDMDANO2Bnz (177.83 mg/g) > F-QuatDMDACF3Bnz (40.37 mg/g) > F-QuatDMDABnz (36.98 mg/g) > F-QuatDMDAMeBnz (12.71 mg/g). The pseudo-second-order kinetic model was found to be the best fit to describe the adsorption kinetics of both metal ions onto all the sorbent materials. K2 (g.mg-1min-1) value in pseudo-second-order kinetics showed the same order of adsorption as observed in the Freundlich isotherms. The adsorption studies showed adsorption takes place via chemisorption process. This thesis presents PGMs and iridium-specific materials that could be applied in solutions of secondary PGMs sources containing rhodium, platinum and palladium with bis-benzimidazoles as well as in feed solutions from ore processing with diammonium cations for iridium recovery.
- Full Text:
- Date Issued: 2019
- Authors: Moleko-Boyce, Pulleng
- Date: 2019
- Subjects: Bioinorganic chemistry , Metal complexes Speciation (Chemistry)
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/42510 , vital:36664
- Description: The study is divided into two sections; namely, (1) the design of rhodium(III) specific chelating ligands (tridentate bis-benzimidazole derivatives), and (2) the development of iridium(IV)-specific quaternary diammonium cations with electron donating and electron withdrawing groups. Bis-benzimidazole chelating ligands used were bis((1H-benzimidazol-2-yl)methyl)amine (NNN1), bis((1H-benzimidazol-2-yl)ethyl)amine (NNN2), bis((1H-benzimidazol-2-yl)methyl)sulfide (NSN1) and bis((1H-benzimidazol-2-yl)ethyl)sulfide (NSN2). Quaternary diammonium cations used were tetramethylbenzyl-1,10-diammonium chloride (QuatDMDAMeBnz), tetrabenzyl-1,10-diammonium chloride (QuatDMDABnz), tetratrifluoromethylbenzyl-1,10-diammonium chloride (QuatDMDACF3Bnz) and tetranitrobenzyl-1,10-diammonium chloride (QuatDMDANO2Bnz). For both studies, polyvinylbenzylchloride (PVBC) nanofibers were used as support material. The PVBC nanofibers which were functionalised with bis-benzimidazole derivatives and quaternary diammonium cations, respectively, were investigated for the selectivity for Rh(III) over Ir(III), Pt(II), Pd(II) and Ni(II), and for separation of Ir(IV) from Rh(III), respectively. The sorbent materials were characterised by FTIR, SEM, BET surface area, TGA, EDS and elemental analysis, and the results showed that the functionalization of the sorbent materials was successful.The efficiency of bis-benzimidazole derivatives and quaternary diammonium cations, respectively, were investigated in a column study under dynamic flow adsorption conditions. The adsorption kinetics and isotherms were investigated under batch conditions and fitted on pseudo-first-order and pseudo-second-order model, and Freundlich and Langmuir isotherm, respectively. It was observed that the bis-benzimidazole derivatives showed uptake of [RhCl3(H2O)3], and the loading capacities were observed in the following order; NSN1 (181.06 mg/g) > NSN2 (148.55 mg/g) > NNN1 (131.88 mg/g) > NNN2 (75.87 mg/g). The bis-benzimidazole derivatives preference for metal ions was further investigated with a multi-element solution containing Rh(III), Ir(III), Pt(II), Pd(II) and Ni(II). The bis-benzimidazole derivatives showed the following order of loading capacity: NSN1 (47.28 mg/g) > NSN2 (23.89 mg/g) > NNN1 (17.47 mg/g) > NNN2 (14.91 mg/g) for Rh(III); NSN2 (10.64 mg/g) > NNN2 (6.84 mg/g) > NSN1 (5.74 mg/g) > NNN1 (5.02 mg/g) for Ir(III); NNN2 (33.96 mg/g) > NSN1 (30.95 mg/g) > NSN2 (19.95 mg/g) > NNN1 (14.92 mg/g) for Pt(II); NNN1 (47.94 mg/g) > NNN2 (28.90 mg/g) > NSN1 (16.22 mg/g) > NSN2 (15.83 mg/g) for Pd(II). Bis-benzimidazole derivatives showed no uptake of nickel(II) under these conditions. It was observed the ligand-selectivity order for Rh(III) was similar in both single-element and multi-element studies. This order showed that the bis-benzimidazoles containing a sulfur atom showed a high preference for rhodium(III) compared to Pt(II) which had a high preference for NNN2 as well as Pd(II) which had a high preference for NNN1. Ir(III) generally had a lower preference for the ligands presumably due to its higher kinetic inertness compared with Rh(III). Column sorption of [IrCl6]2- and [RhCl5(H2O)]2- on nanofibers functionalized with diammonium cations was carried out and the loading capacities of [IrCl6]2- were obtained. [RhCl5(H2O)]2- was not adsorbed by the sorbent materials while [IrCl6]2- was loaded onto the column. The loading capacities of [IrCl6]2- with the quaternary diammonium sorbent materials increased in the order of F-QuatDMDAMeBnz (60.29 mg/g) < F-QuatDMDABnz (67.61 mg/g) < F-QuatDMDACF3Bnz (107.59 mg/g) < F-QuatDMDANO2Bnz (140.47 mg/g). The loading capacity for Ir(IV) with quaternary diammonium cationic nanofibers increased with an increase in the electron-withdrawing nature of the quaternizing group. The charge delocalizing ability of the nitrobenzyl group resulted in the best interaction of the diammonium cation with [IrCl6]2-. Batch equilibrium studies were carried out to assess the efficiency of bis-benzimidazole chelating derivatives as adsorbents using a multi-metal solution (Rh(III), Ir(III), Pt(II), Pd(II) and Ni(II)) in 0.5 M HCl. The efficiency of the quaternary diammonium cations was tested using a binary metal solution (Ir(IV) and Rh(III)) in 6 M HCl. The isothermal batch adsorption studies of a multi-metal solution with bis-benzimidazoles derivatives fitted the Langmuir isotherm model which confirmed monolayer adsorption onto a homogeneous surface. The Langmuir isotherm parameter (qe (mg/g)), using functionalized nanofibers, showed the order of NNN2 (128.21 mg/g) > NSN1 (99.01 mg/g) > NSN2 (91.74 mg/g) > NNN1 (84.03 mg/g) for Pt(II); NNN1 (66.23 mg/g) > NNN2 (5.89 mg/g) > NSN1 (1.40 mg/g) > NSN2 (0.59 mg/g) for Pd(II); NSN2 (10.64 mg/g) > NNN2 (6.84 mg/g) > NSN1 (5.74 mg/g) > NNN1 (5.02 mg/g) for Ir(III); NSN1 (140.85 mg/g) > NSN2 (109.89 mg/g) > NNN1 (104.17 mg/g) > NNN2 (91.74 mg/g) for Rh(III). The pseudo-first-order kinetics model was found to be the best fit to describe the adsorption kinetics of all metal ions onto all the sorbent materials. K1 (min-1) value in pseudo-first-order kinetics showed the same order of adsorption as observed in the Langmuir isotherms. The isothermal batch adsorption studies of [IrCl6]2- and [RhCl5(H2O)]2- with quaternary diammonium cations fitted the Freundlich isotherm model and confirmed to be effective for multiple-layered adsorption onto a heterogeneous surface. The Freundlich isotherm parameter (kf (mg/g)) using functionalized quaternary diammonium cationic nanofibers increased in the order of F-QuatDMDANO2Bnz (794.33 mg/g) > F-QuatDMDACF3Bnz (185.35 mg/g) > F-QuatDMDABnz (156.32 mg/g) > F-QuatDMDAMeBnz (112.46 mg/g) for Ir(IV) uptake. F-QuatDMDANO2Bnz resin showed the highest adsorption than that of F-QuatDMDAMeBnz, F-QuatDMDABnz and F-QuatDMDACF3Bnz and this order is similar to what was observed in column studies. The quaternary diammonium cations were shown to have the highest adsorption capacity for Ir(IV) compared with Rh(III). The adsorption of Rh(III) was also observed to increase in the order of F-QuatDMDANO2Bnz (177.83 mg/g) > F-QuatDMDACF3Bnz (40.37 mg/g) > F-QuatDMDABnz (36.98 mg/g) > F-QuatDMDAMeBnz (12.71 mg/g). The pseudo-second-order kinetic model was found to be the best fit to describe the adsorption kinetics of both metal ions onto all the sorbent materials. K2 (g.mg-1min-1) value in pseudo-second-order kinetics showed the same order of adsorption as observed in the Freundlich isotherms. The adsorption studies showed adsorption takes place via chemisorption process. This thesis presents PGMs and iridium-specific materials that could be applied in solutions of secondary PGMs sources containing rhodium, platinum and palladium with bis-benzimidazoles as well as in feed solutions from ore processing with diammonium cations for iridium recovery.
- Full Text:
- Date Issued: 2019
Development of spunlaced nonwoven filters from PAN, PPS and PI fibres for industrial use
- Authors: Maduna, Lebo
- Date: 2019
- Subjects: Textile fibers -- Mechanical properties , Nonwoven fabrics --Technological innovations Textile fibers, Synthetic -- Mechanical properties
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/32610 , vital:32266
- Description: The aim of this work was to produce filter fabrics suitable for high temperature and other conditions encountered in coal power plants using the spunlacing manufacturing technique. Polyacrylonitrile (PAN), polyphenylene sulfide (PPS) and polyimide (PI) fibres being combined suitable for this purpose, each at three aerial density of (440, 500 and 560 g/m2) and each at three water jet pressures of (60, 80 and 90 bars) were produced. The effect of changing these on the fabric air permeability, tensile strength and filtration properties were investigated and optimized. For this purpose, the Box-Behnken experimental design (BBD) was considered the best and therefore used. The PPS/PI fabrics were found to consistently have the highest air permeability, tensile strength and filtration performance, followed by the PPS fabrics and the PAN fabrics. As could be expected, an increase in area weight and water jet pressure generally resulted in a decrease in air permeability irrespective of the fabric types, this being due to the increased number of fibres and therefore greater fibre surface area and entanglement respectively. For fabric tensile strength, the effects of varying fabric area weight and water jet pressure were not so straight forward. For cross direction (CD) an increase in water jet pressure resulted in an increase tensile strength whereas in the machine direction (MD) there was a decrease when the water jet pressure was increased. The filtration efficiency and dust holding capacity of the fabrics were found to be around 96% and 211g/m2, respectively, with the exception of the PPS (500 g/m2 80 bar) fabrics. Increasing the amount of dust being fed to the fabric during each loading cycle, it caused an increase in pressure drop, as the fabrics became increasingly clogged forming a cake layer.
- Full Text:
- Date Issued: 2019
- Authors: Maduna, Lebo
- Date: 2019
- Subjects: Textile fibers -- Mechanical properties , Nonwoven fabrics --Technological innovations Textile fibers, Synthetic -- Mechanical properties
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/32610 , vital:32266
- Description: The aim of this work was to produce filter fabrics suitable for high temperature and other conditions encountered in coal power plants using the spunlacing manufacturing technique. Polyacrylonitrile (PAN), polyphenylene sulfide (PPS) and polyimide (PI) fibres being combined suitable for this purpose, each at three aerial density of (440, 500 and 560 g/m2) and each at three water jet pressures of (60, 80 and 90 bars) were produced. The effect of changing these on the fabric air permeability, tensile strength and filtration properties were investigated and optimized. For this purpose, the Box-Behnken experimental design (BBD) was considered the best and therefore used. The PPS/PI fabrics were found to consistently have the highest air permeability, tensile strength and filtration performance, followed by the PPS fabrics and the PAN fabrics. As could be expected, an increase in area weight and water jet pressure generally resulted in a decrease in air permeability irrespective of the fabric types, this being due to the increased number of fibres and therefore greater fibre surface area and entanglement respectively. For fabric tensile strength, the effects of varying fabric area weight and water jet pressure were not so straight forward. For cross direction (CD) an increase in water jet pressure resulted in an increase tensile strength whereas in the machine direction (MD) there was a decrease when the water jet pressure was increased. The filtration efficiency and dust holding capacity of the fabrics were found to be around 96% and 211g/m2, respectively, with the exception of the PPS (500 g/m2 80 bar) fabrics. Increasing the amount of dust being fed to the fabric during each loading cycle, it caused an increase in pressure drop, as the fabrics became increasingly clogged forming a cake layer.
- Full Text:
- Date Issued: 2019
Mechanisms and modes of β-N-methylamino-lalanine neurotoxicity: the basis for designing therapies
- Authors: Van Onselen, Rianita
- Date: 2019
- Subjects: Cyanobacteria , Amino acids -- Toxicology , Neurotoxic agents
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/32971 , vital:32483
- Description: Since the discovery of the non-canonical amino acid β-N-methylamino-L-alanine (BMAA) and the demonstration of its acute neurotoxicity in chicks and rats, it has been postulated that BMAA might contribute to the development of neurodegenerative diseases worldwide due to its presence in numerous aquatic and terrestrial food webs. This hypothesized link was widely criticized because of the inability to reproduce symptoms in a BMAA-exposed animal model that resembled the symptoms observed in humans, and for the inability to achieve significant levels of toxicity in in vitro models via the postulated mechanisms of toxicity. The most widely described mechanism of BMAA toxicity was excitotoxicity by over-excitation of ionotropic and/or metabotropic glutamate receptors following activation by BMAA. However, the excitotoxic potency of BMAA is much lower than those of other known excitotoxins and it was not known whether BMAA could accumulate in significant concentrations in synapses to cause the said excitotoxicity. Therefore, uptake of BMAA into synaptic vesicles from where it can be released into synapses in high concentrations, was investigated and it was found that, unlike the uptake that was observed for glutamate, BMAA was not taken up into synaptic vesicles. This discovery suggests that BMAA is not released into synapses via synaptic vesicles and that excitotoxicity is an unlikely mechanism of BMAA toxicity in mammalian systems. Misincorporation of BMAA into proteins in the place of L-serine was suggested to be an important mechanism of BMAA toxicity that could lead to protein misfolding and the subsequent protein aggregates that are typically found in the central nervous system (CNS) of neurodegenerative disease patients. However, previous studies in prokaryotes and in a rat pheochromocytoma PC12 cell line showed that misincorporation of BMAA does not occur to any significant extent. However, these studies were criticized for not using human-derived model systems to show that misincorporation does not occur, and it was argued that due to differences in mitochondrial protein synthesis mechanisms, misincorporation of BMAA into human proteins could not be ruled out as a possible mechanism of toxicity. Therefore, misincorporation of BMAA was investigated in a number of human-derived non-neuronal cell lines and directly compared to the misincorporation of other known amino acid analogues. No evidence of misincorporation of BMAA into these cell lines was obtained and therefore it was concluded that misincorporation of BMAA into proteins does not occur in human-derived cell models. Although misincorporation of BMAA into proteins was refuted as a mechanism of toxicity, the strong interactions between BMAA and proteins that require extensive purification procedures to remove the associated BMAA, could not be discounted as a possible contributor to the toxicity of BMAA. Cell-free interactions between BMAA and enzymes, which resulted in reduced activity, were described previously but the nature of these interactions was never determined. Therefore, the direct interactions between BMAA and a range of commercial proteins and melanin (that is known to also have a strong affinity for BMAA) were investigated in an attempt to describe the nature of these interactions. It was discovered that BMAA has a high affinity for hydroxyl groups, and that if these hydroxyl groups in the form of hydroxyl containing amino acid residues occurred in important regulatory or active sites of proteins, BMAA reduced the enzyme activity. Catalase was subsequently selected as an important enzyme required for the maintenance of the delicate reactive oxygen species (ROS) balance in the CNS, to test the effect of BMAA on the activity of the enzyme. BMAA inhibited a human commercial extract of catalase in a cell free system, and this inhibition appeared to be non-competitive in nature. Subsequently, catalase in an extract from a human cell line was also shown to be inhibited by BMAA and it was concluded that this BMAA induced inhibition of catalase could be an important contributor to the toxicity of BMAA in in vivo systems. The affinity of BMAA for hydroxyl groups, especially the reactive L-tyrosine side chain hydroxyl, was recognized as a possible mechanism that can be utilized to protect against the toxicity of BMAA. It was subsequently shown that excess concentrations of L-serine and L-tyrosine could protect against the BMAA-induced enzyme inhibition and improper folding of proteins in a cell-free system. By administering an equimolar concentration of either L-phenylalanine (the soluble precursor of L-tyrosine) or L-serine an hour before administration of BMAA in a rat model, the BMAA-induced neurotoxicity was greatly reduced, especially by treatment with L-phenylalanine, which resulted in a decrease of between 60-70% in the observed neuropathologies. It was recognized that the protection offered by L-phenylalanine was greater than would be expected if protection was by virtue of direct hydroxyl binding alone and it was subsequently hypothesized that the conversion of L-phenylalanine to dopamine could have contributed to the observed protection. Subsequently, the possible protection offered by dopamine, administered as L-DOPA, against BMAA neurotoxicity was investigated in the same neonatal rat model and compared to the protection offered by L-tyrosine. It was discovered that dopamine protected against the BMAAinduced neuronal cell losses in the hippocampus, striatum and spinal cord but it was not as efficient as L-tyrosine in protection against the BMAA-induced proteinopathies, suggesting two distinct mechanisms of BMAA toxicity, one of which is a depletion of dopamine, which had not been previously described. Finally, the nature of the BMAA-induced dopamine depletion was investigated by administering BMAA in combination with other dopaminergic modifiers viz. apomorphine (a D1/D2 receptor agonist), a dopamine transporter inhibitor (GBR12783) and reserpine (a vesicular monoamine transporter -VMAT2- inhibitor) to the neonatal rat model in an attempt to describe how BMAA functions as a dopaminergic toxin. Based on these results it was concluded that BMAA inhibits uptake of dopamine into synaptic vesicles by inhibiting VMAT2-mediated uptake of dopamine, which causes neuronal loss in the hippocampus, striatum and substantia nigra pars compacta, and that the BMAA-induced inhibition of catalase contributes significantly to the toxicity of BMAA by causing an accumulation of hydrogen peroxide in the hippocampus, striatum and spinal cord, which results in extensive neuronal damage in these areas. This work was the first to thoroughly investigate the mechanisms that explain the observed pathologies caused by BMAA in an in vivo model, and was the first to suggest that BMAA can reduce the dopamine in the CNS by inhibiting VMAT2-mediated uptake of dopamine into synaptic vesicles, and increase damage by reactive oxygen species by inhibiting catalase. BMAA is therefore a multimechanistic and multimodal.
- Full Text:
- Date Issued: 2019
- Authors: Van Onselen, Rianita
- Date: 2019
- Subjects: Cyanobacteria , Amino acids -- Toxicology , Neurotoxic agents
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/32971 , vital:32483
- Description: Since the discovery of the non-canonical amino acid β-N-methylamino-L-alanine (BMAA) and the demonstration of its acute neurotoxicity in chicks and rats, it has been postulated that BMAA might contribute to the development of neurodegenerative diseases worldwide due to its presence in numerous aquatic and terrestrial food webs. This hypothesized link was widely criticized because of the inability to reproduce symptoms in a BMAA-exposed animal model that resembled the symptoms observed in humans, and for the inability to achieve significant levels of toxicity in in vitro models via the postulated mechanisms of toxicity. The most widely described mechanism of BMAA toxicity was excitotoxicity by over-excitation of ionotropic and/or metabotropic glutamate receptors following activation by BMAA. However, the excitotoxic potency of BMAA is much lower than those of other known excitotoxins and it was not known whether BMAA could accumulate in significant concentrations in synapses to cause the said excitotoxicity. Therefore, uptake of BMAA into synaptic vesicles from where it can be released into synapses in high concentrations, was investigated and it was found that, unlike the uptake that was observed for glutamate, BMAA was not taken up into synaptic vesicles. This discovery suggests that BMAA is not released into synapses via synaptic vesicles and that excitotoxicity is an unlikely mechanism of BMAA toxicity in mammalian systems. Misincorporation of BMAA into proteins in the place of L-serine was suggested to be an important mechanism of BMAA toxicity that could lead to protein misfolding and the subsequent protein aggregates that are typically found in the central nervous system (CNS) of neurodegenerative disease patients. However, previous studies in prokaryotes and in a rat pheochromocytoma PC12 cell line showed that misincorporation of BMAA does not occur to any significant extent. However, these studies were criticized for not using human-derived model systems to show that misincorporation does not occur, and it was argued that due to differences in mitochondrial protein synthesis mechanisms, misincorporation of BMAA into human proteins could not be ruled out as a possible mechanism of toxicity. Therefore, misincorporation of BMAA was investigated in a number of human-derived non-neuronal cell lines and directly compared to the misincorporation of other known amino acid analogues. No evidence of misincorporation of BMAA into these cell lines was obtained and therefore it was concluded that misincorporation of BMAA into proteins does not occur in human-derived cell models. Although misincorporation of BMAA into proteins was refuted as a mechanism of toxicity, the strong interactions between BMAA and proteins that require extensive purification procedures to remove the associated BMAA, could not be discounted as a possible contributor to the toxicity of BMAA. Cell-free interactions between BMAA and enzymes, which resulted in reduced activity, were described previously but the nature of these interactions was never determined. Therefore, the direct interactions between BMAA and a range of commercial proteins and melanin (that is known to also have a strong affinity for BMAA) were investigated in an attempt to describe the nature of these interactions. It was discovered that BMAA has a high affinity for hydroxyl groups, and that if these hydroxyl groups in the form of hydroxyl containing amino acid residues occurred in important regulatory or active sites of proteins, BMAA reduced the enzyme activity. Catalase was subsequently selected as an important enzyme required for the maintenance of the delicate reactive oxygen species (ROS) balance in the CNS, to test the effect of BMAA on the activity of the enzyme. BMAA inhibited a human commercial extract of catalase in a cell free system, and this inhibition appeared to be non-competitive in nature. Subsequently, catalase in an extract from a human cell line was also shown to be inhibited by BMAA and it was concluded that this BMAA induced inhibition of catalase could be an important contributor to the toxicity of BMAA in in vivo systems. The affinity of BMAA for hydroxyl groups, especially the reactive L-tyrosine side chain hydroxyl, was recognized as a possible mechanism that can be utilized to protect against the toxicity of BMAA. It was subsequently shown that excess concentrations of L-serine and L-tyrosine could protect against the BMAA-induced enzyme inhibition and improper folding of proteins in a cell-free system. By administering an equimolar concentration of either L-phenylalanine (the soluble precursor of L-tyrosine) or L-serine an hour before administration of BMAA in a rat model, the BMAA-induced neurotoxicity was greatly reduced, especially by treatment with L-phenylalanine, which resulted in a decrease of between 60-70% in the observed neuropathologies. It was recognized that the protection offered by L-phenylalanine was greater than would be expected if protection was by virtue of direct hydroxyl binding alone and it was subsequently hypothesized that the conversion of L-phenylalanine to dopamine could have contributed to the observed protection. Subsequently, the possible protection offered by dopamine, administered as L-DOPA, against BMAA neurotoxicity was investigated in the same neonatal rat model and compared to the protection offered by L-tyrosine. It was discovered that dopamine protected against the BMAAinduced neuronal cell losses in the hippocampus, striatum and spinal cord but it was not as efficient as L-tyrosine in protection against the BMAA-induced proteinopathies, suggesting two distinct mechanisms of BMAA toxicity, one of which is a depletion of dopamine, which had not been previously described. Finally, the nature of the BMAA-induced dopamine depletion was investigated by administering BMAA in combination with other dopaminergic modifiers viz. apomorphine (a D1/D2 receptor agonist), a dopamine transporter inhibitor (GBR12783) and reserpine (a vesicular monoamine transporter -VMAT2- inhibitor) to the neonatal rat model in an attempt to describe how BMAA functions as a dopaminergic toxin. Based on these results it was concluded that BMAA inhibits uptake of dopamine into synaptic vesicles by inhibiting VMAT2-mediated uptake of dopamine, which causes neuronal loss in the hippocampus, striatum and substantia nigra pars compacta, and that the BMAA-induced inhibition of catalase contributes significantly to the toxicity of BMAA by causing an accumulation of hydrogen peroxide in the hippocampus, striatum and spinal cord, which results in extensive neuronal damage in these areas. This work was the first to thoroughly investigate the mechanisms that explain the observed pathologies caused by BMAA in an in vivo model, and was the first to suggest that BMAA can reduce the dopamine in the CNS by inhibiting VMAT2-mediated uptake of dopamine into synaptic vesicles, and increase damage by reactive oxygen species by inhibiting catalase. BMAA is therefore a multimechanistic and multimodal.
- Full Text:
- Date Issued: 2019
Microalgae as indicators of environmental change in the St Lucia estuarine system
- Authors: Nunes, Monique
- Date: 2019
- Subjects: Microalgae -- South Africa -- St Lucia estuarine
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/31165 , vital:31333
- Description: The St Lucia system represents 56% of the estuarine habitat area in South Africa. This dynamic estuarine lake responds to a complex interaction of climatic and anthropogenic drivers. There has been a long history of human interventions and the extended drought (2002-2012) led to major changes. In 1952, the inlet shared between the estuarine system and the Mfolozi River was artificially separated, which significantly reduced freshwater inflow. The systems were separated because of the silt load from the expanding agricultural activities in the Mfolozi River catchment. The drought prompted a critical review of re-establishing the historic connection, and in July 2012, the beach spillway was excavated marking the start of restoring connectivity. Increased rainfall coupled with the partial re-connection with the agriculturally-impacted Mfolozi River shifted the St Lucia estuarine system to a new wet phase. This study investigated the microalgal dynamics of the estuarine system as it transitioned to this new wet phase. Results showed that the phytoplankton and microphytobenthos (MPB) biomass was low (< 5 ug l-1 and < 60 mg m-2 respectively) throughout the system due to nutrient and light availability, water residence time and sediment type. The increased freshwater inflow from the Mfolozi River was limited to the Narrows and available nutrients promoted the growth of cyanobacteria and dinophytes in the water column. Of interest was the presence of the potentially harmful algal species Prorocentrum minimum. A weekly study investigated the relationship between freshwater inflow, phytoplankton succession and the introduction and persistence of harmful microalgal species in the lower reaches. The increased freshwater inflow received from the Mfolozi River has led to a system in a constant state of flux. The changes in salinity, light availability and water residence time, shifted the phytoplankton functional group structure to one where chlorophytes and euglenoids were dominant. The persistence of the nutrient tolerant euglenoids highlighted the influence of the nutrient-enriched freshwater inflow on the phytoplankton functional groups. However, the relative abundance of all functional groups remained low due to the rapid change in environmental conditions. The next step was to test the use of an artificial substrate as a monitoring method to detect nutrient induced change. Findings from the 28-day study showed that epilithic diatom communities growing on glass slides were good indicators of nutrient enrichment. The epilithic algal biomass increased significantly from 1.5 mg m-2 (Day 14) to 6 mg m-2 (Day 21), following peak river inflow. In contrast, the phytoplankton and MPB showed no discernible increase in biomass. Epilithic diatom community diversity, richness and evenness scores declined due to the increasing relative abundance of a single nutrient tolerant species, Cocconeis placentula var. euglypta. The use of diatoms as indicators of change depends on the accuracy of identification to species level. This can be a slow process requiring a high level of taxonomic expertise. Considering that molecular phylogenetic methods are readily used to produce a reliable taxonomic list, the suitability of DNA metabarcoding for diatom monitoring in the St Lucia Estuary was tested. Comparisons between the molecular and morphological method indicated that the diatom taxonomic composition was similar at 9 to 27% for genus level and 2 to 9% at species level. The low taxonomic similarity was related to the large number of unclassified DNA representative sequences included in the molecular inventory due to the incomplete DNA library database. However, the available operational taxonomic units (OTUs) from the molecular method did provide an effective alternative to determine the relevant community diversity estimates required for the application of the South African Estuarine Health Index (EHI). This research has highlighted the dynamic nature of the St Lucia estuarine system and the effectiveness of the microalgal communities to differentiate between the cyclical climatic phases and anthropogenic stressors such as nutrient enrichment.
- Full Text:
- Date Issued: 2019
- Authors: Nunes, Monique
- Date: 2019
- Subjects: Microalgae -- South Africa -- St Lucia estuarine
- Language: English
- Type: Thesis , Doctoral , DPhil
- Identifier: http://hdl.handle.net/10948/31165 , vital:31333
- Description: The St Lucia system represents 56% of the estuarine habitat area in South Africa. This dynamic estuarine lake responds to a complex interaction of climatic and anthropogenic drivers. There has been a long history of human interventions and the extended drought (2002-2012) led to major changes. In 1952, the inlet shared between the estuarine system and the Mfolozi River was artificially separated, which significantly reduced freshwater inflow. The systems were separated because of the silt load from the expanding agricultural activities in the Mfolozi River catchment. The drought prompted a critical review of re-establishing the historic connection, and in July 2012, the beach spillway was excavated marking the start of restoring connectivity. Increased rainfall coupled with the partial re-connection with the agriculturally-impacted Mfolozi River shifted the St Lucia estuarine system to a new wet phase. This study investigated the microalgal dynamics of the estuarine system as it transitioned to this new wet phase. Results showed that the phytoplankton and microphytobenthos (MPB) biomass was low (< 5 ug l-1 and < 60 mg m-2 respectively) throughout the system due to nutrient and light availability, water residence time and sediment type. The increased freshwater inflow from the Mfolozi River was limited to the Narrows and available nutrients promoted the growth of cyanobacteria and dinophytes in the water column. Of interest was the presence of the potentially harmful algal species Prorocentrum minimum. A weekly study investigated the relationship between freshwater inflow, phytoplankton succession and the introduction and persistence of harmful microalgal species in the lower reaches. The increased freshwater inflow received from the Mfolozi River has led to a system in a constant state of flux. The changes in salinity, light availability and water residence time, shifted the phytoplankton functional group structure to one where chlorophytes and euglenoids were dominant. The persistence of the nutrient tolerant euglenoids highlighted the influence of the nutrient-enriched freshwater inflow on the phytoplankton functional groups. However, the relative abundance of all functional groups remained low due to the rapid change in environmental conditions. The next step was to test the use of an artificial substrate as a monitoring method to detect nutrient induced change. Findings from the 28-day study showed that epilithic diatom communities growing on glass slides were good indicators of nutrient enrichment. The epilithic algal biomass increased significantly from 1.5 mg m-2 (Day 14) to 6 mg m-2 (Day 21), following peak river inflow. In contrast, the phytoplankton and MPB showed no discernible increase in biomass. Epilithic diatom community diversity, richness and evenness scores declined due to the increasing relative abundance of a single nutrient tolerant species, Cocconeis placentula var. euglypta. The use of diatoms as indicators of change depends on the accuracy of identification to species level. This can be a slow process requiring a high level of taxonomic expertise. Considering that molecular phylogenetic methods are readily used to produce a reliable taxonomic list, the suitability of DNA metabarcoding for diatom monitoring in the St Lucia Estuary was tested. Comparisons between the molecular and morphological method indicated that the diatom taxonomic composition was similar at 9 to 27% for genus level and 2 to 9% at species level. The low taxonomic similarity was related to the large number of unclassified DNA representative sequences included in the molecular inventory due to the incomplete DNA library database. However, the available operational taxonomic units (OTUs) from the molecular method did provide an effective alternative to determine the relevant community diversity estimates required for the application of the South African Estuarine Health Index (EHI). This research has highlighted the dynamic nature of the St Lucia estuarine system and the effectiveness of the microalgal communities to differentiate between the cyclical climatic phases and anthropogenic stressors such as nutrient enrichment.
- Full Text:
- Date Issued: 2019