Probing the binding sites of novel 2-substituted imidazole-4,5-dicarboxylic acids : Towards new imidazole-based drugs
- Authors: Clark, Candyce Raine
- Date: 2021
- Subjects: Heterocyclic compounds Coordination compounds
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/50848 , vital:42957
- Description: A series of benzimidazole and imidazole-4,5-dicarboxylic acids were prepared using known and adapted protocols. Novel ligands synthesised in this study include 2-tert-butyl-benzimidazole, 2-tert-butyl-imidazole-4,5-dicarboxylic acid, an intermediate in the synthesis of 1,2- bis(2-benzimidazyl)-1,2,-dihydroxyethane and 2-(2-hydroxyphenyl)-benzimidazole with a salicylic acid co-crystallisate. All ligands prepared were reacted with metal salt solutions in order to prepare metal coordination compounds. Complexes with Cu(II), Ag(I), Mn(II), Ni(II) and Cd(II) ions and 2-phenyl-imidazole-4,5-dicarboxylic acid were prepared, as well as Cu(II) and Mn(II) with 2-trifluoro-methyl-imidazole-4,5-dicarboxylic acid and Cu(II) ions with 2-(2- hydroxyphenyl)-benzimidazole. The ligands and complexes were characterised by means of single-crystal X-ray diffraction studies, NMR analysis, UV spectrophotometric analysis and IR spectroscopy. DFT studies were carried out for selected ligands in order to provide theoretical values for comparison. Antimicrobial studies were performed on some of the ligands and complexes and these results show promise for the future development of new imidazole-based drugs.
- Full Text:
- Date Issued: 2021
- Authors: Clark, Candyce Raine
- Date: 2021
- Subjects: Heterocyclic compounds Coordination compounds
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/50848 , vital:42957
- Description: A series of benzimidazole and imidazole-4,5-dicarboxylic acids were prepared using known and adapted protocols. Novel ligands synthesised in this study include 2-tert-butyl-benzimidazole, 2-tert-butyl-imidazole-4,5-dicarboxylic acid, an intermediate in the synthesis of 1,2- bis(2-benzimidazyl)-1,2,-dihydroxyethane and 2-(2-hydroxyphenyl)-benzimidazole with a salicylic acid co-crystallisate. All ligands prepared were reacted with metal salt solutions in order to prepare metal coordination compounds. Complexes with Cu(II), Ag(I), Mn(II), Ni(II) and Cd(II) ions and 2-phenyl-imidazole-4,5-dicarboxylic acid were prepared, as well as Cu(II) and Mn(II) with 2-trifluoro-methyl-imidazole-4,5-dicarboxylic acid and Cu(II) ions with 2-(2- hydroxyphenyl)-benzimidazole. The ligands and complexes were characterised by means of single-crystal X-ray diffraction studies, NMR analysis, UV spectrophotometric analysis and IR spectroscopy. DFT studies were carried out for selected ligands in order to provide theoretical values for comparison. Antimicrobial studies were performed on some of the ligands and complexes and these results show promise for the future development of new imidazole-based drugs.
- Full Text:
- Date Issued: 2021
Applied phylogeography : mapping the genetic resource of Honeybush across the Cape Floristic Region
- Authors: Galuszynski, Nicholas C
- Date: 2020
- Subjects: Phylogeography -- South Africa
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/46771 , vital:39655
- Description: Aim: Honeybush tea is a herbal infusion made from members of the Cape Floristic Region (CFR) endemic Cyclopia Vent. Consting of 21 extant species, six are anthropogenically redistributed across the CFR for Honeybush cultivation. However, with no information regarding the distribution and levels of genetic diversity among wild Honeybush populations, anthropogenic translocation of cultivated genotypes may risk disrupting natural genetic diversity patterns. In this thesis, an applied phylogeographic approach is used to provide baseline insights into the spatial structuring of Honeybush genetic diversity. Location: The Cape Floristic Region (CFR), located along the southern Cape of South Africa. Methods: Spatial structuring of evolutionary lineages within Cyclopia are explored at a genus level using data generated from the combination of Anchored Hybrid Enrichment library preparation and high through put sequencing. A High Resolution Melt analysis (HRM) toolkit is then developed to screen haplotype variation across three non-coding chloroplast loci. This HRM toolkit coupled with haplotype confirmation Sanger sequencing is then applied to describe the spatial structuring of genetic diversity in wild populations for two Honeybush species (C. intermedia and C. subternata) and to compare genetic diversity among wild and cultivated populations of three Honeybush species (C. intermedia, C. subternata, and C. longifolia). A reanalysis of published phylogeographic data focusing on CFR plant taxa is used to test if phylogeographic breaks occur across the boundaries among adjacent Centers of Endemism (CoEs). Finally, the low-copy nuclear data set generated via AHE and high throughput sequencing is summarized. Results: Phylogeographic structuring was detected at the inter- and intraspecific levels in Cyclopia. The genus level analysis revealed a strong phylogenetic split between western CFR and eastern CFR endemic species in the nuclear and chloroplast genomes. At the intraspecific level, HRM proved to be a high throughput and accurate tool for haplotype detection, revealing phylogeographic structuring of genetic diversity in C.intermediaand C.subternata. This genetic diversity and structuring was not detected in cultivated Honeybush populations, which appear to have originated from a small number of founding individuals and have experienced a genetic bottleneck. Spatial structuring of genetic lineages is common among CFR plant taxa and the reanalysis of existing data found a significant trend for phylogeographic breaks tooccuracrosstheboundariesamongadjacentCoEs. Finally, over200000baseswere sequenced across 445 low-copy nuclear loci for 14 Cyclopia species. Conclusions: This thesis provides important baseline information on the spatial distribution of wild Honeybush genetic diversity. Phylogeographic structuring is present in wild Honeybush populations. This however is not accounted for in cultivated populations, which may place wild genetic diversity at risk of being disrupted if geneflow among wild and cultivated populations occurs. The results from this thesis should therefore be used to develop precautionary guidelines for the anthropogenic redistribution of Honeybush genetic material for cultivation. Furthermore, the novel insights into regional patterns of phylogeohraphic structuring of plant populations should encourage more efficient and hypothesis driven sampling designs in future phylogeographic work in the CFR. The high throughput sequence data generated in this thesis should transform Cyclopia to a model organism for the study of molecular evolution in the CFR.
- Full Text:
- Date Issued: 2020
- Authors: Galuszynski, Nicholas C
- Date: 2020
- Subjects: Phylogeography -- South Africa
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/46771 , vital:39655
- Description: Aim: Honeybush tea is a herbal infusion made from members of the Cape Floristic Region (CFR) endemic Cyclopia Vent. Consting of 21 extant species, six are anthropogenically redistributed across the CFR for Honeybush cultivation. However, with no information regarding the distribution and levels of genetic diversity among wild Honeybush populations, anthropogenic translocation of cultivated genotypes may risk disrupting natural genetic diversity patterns. In this thesis, an applied phylogeographic approach is used to provide baseline insights into the spatial structuring of Honeybush genetic diversity. Location: The Cape Floristic Region (CFR), located along the southern Cape of South Africa. Methods: Spatial structuring of evolutionary lineages within Cyclopia are explored at a genus level using data generated from the combination of Anchored Hybrid Enrichment library preparation and high through put sequencing. A High Resolution Melt analysis (HRM) toolkit is then developed to screen haplotype variation across three non-coding chloroplast loci. This HRM toolkit coupled with haplotype confirmation Sanger sequencing is then applied to describe the spatial structuring of genetic diversity in wild populations for two Honeybush species (C. intermedia and C. subternata) and to compare genetic diversity among wild and cultivated populations of three Honeybush species (C. intermedia, C. subternata, and C. longifolia). A reanalysis of published phylogeographic data focusing on CFR plant taxa is used to test if phylogeographic breaks occur across the boundaries among adjacent Centers of Endemism (CoEs). Finally, the low-copy nuclear data set generated via AHE and high throughput sequencing is summarized. Results: Phylogeographic structuring was detected at the inter- and intraspecific levels in Cyclopia. The genus level analysis revealed a strong phylogenetic split between western CFR and eastern CFR endemic species in the nuclear and chloroplast genomes. At the intraspecific level, HRM proved to be a high throughput and accurate tool for haplotype detection, revealing phylogeographic structuring of genetic diversity in C.intermediaand C.subternata. This genetic diversity and structuring was not detected in cultivated Honeybush populations, which appear to have originated from a small number of founding individuals and have experienced a genetic bottleneck. Spatial structuring of genetic lineages is common among CFR plant taxa and the reanalysis of existing data found a significant trend for phylogeographic breaks tooccuracrosstheboundariesamongadjacentCoEs. Finally, over200000baseswere sequenced across 445 low-copy nuclear loci for 14 Cyclopia species. Conclusions: This thesis provides important baseline information on the spatial distribution of wild Honeybush genetic diversity. Phylogeographic structuring is present in wild Honeybush populations. This however is not accounted for in cultivated populations, which may place wild genetic diversity at risk of being disrupted if geneflow among wild and cultivated populations occurs. The results from this thesis should therefore be used to develop precautionary guidelines for the anthropogenic redistribution of Honeybush genetic material for cultivation. Furthermore, the novel insights into regional patterns of phylogeohraphic structuring of plant populations should encourage more efficient and hypothesis driven sampling designs in future phylogeographic work in the CFR. The high throughput sequence data generated in this thesis should transform Cyclopia to a model organism for the study of molecular evolution in the CFR.
- Full Text:
- Date Issued: 2020
Comprehensive characterization of the antidiabetic potential of selected plants and macrofungi from Africa using an in vitro target-directed screening platform and cellomics
- Authors: Pringle, Nadine Alex
- Date: 2020
- Subjects: Hypoglycemic agents
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/46750 , vital:39653
- Description: Several synthetic antidiabetic drugs have been developed to date, however, most are accompanied by adverse side-effects while remaining expensive and largely inaccessible to the vast majority of those who need it. To provide enough scientific evidence to support the inclusion of more affordable African antidiabetic medicinal plants and macrofungi into healthcare programs, this study sought out to develop a comprehensive in vitro antidiabetic target-directed screening platform incorporating high content screening and analysis/ cellomics. To test the success of this model, the potential antidiabetic mechanisms of five plants (Aspalathus linearis, Brachylaena discolor, Carpobrotus deliciosus, Sutherlandia frutescens and Tarchonanthus camphoratus) and two macrofungal species (Ganoderma lucidum and Hericium erinaceus) were explored. The screening model consisted of approximately 22 assays exploring the antidiabetic effects of selected aqueous and ethanolic extracts in five well-characterised antidiabetic targets: the intestine, liver, skeletal muscle, adipose tissue/ obesity and pancreatic β-cells. These targets were further categorised and scored under three mechanistic classes/ therapeutic targets (postprandial hyperglycaemia; insulin resistance and inflammation; pancreatic β-cell function) to elucidate their potential mechanisms of action and select appropriate animal models for future studies. Almost any normal or diabetic rodent model would be suitable to explore the antidiabetic potential of extracts such as A. linearis, B. discolor ethanol, C. deliciosus ethanol or T. camphoratus which obtained high cumulative scores under postprandial hyperglycaemia while high fat diet and genetic models of obesity appear more suited towards extracts such as H. erinaceus aqueous that obtained their highest cumulative score under insulin resistance. In general, a combination of rodent models ranging from non-obese models to models of obesity and β-cell destruction presenting symptoms from all three mechanistic classes should be considered due to the pleiotropic nature of the tested extracts, however, establishing appropriate experimental designs is crucial. To demonstrate the versatility of the screening platform and emphasise the importance of in vitro screening pertaining to diabetic complications, a more detailed biochemical investigation into the potential therapeutic benefits of A. linearis in the treatment of diabetic wounds was conducted. Several properties supporting the therapeutic potential of rooibos were highlighted with the green and fermented extracts presenting distinctly different characteristics. The pro-inflammatory nature of fermented rooibos may have therapeutic value for wounds characterised with a delayed initial inflammatory phase, such as early diabetic wounds while the green extract appears more suited to wounds burdened with excessive inflammation as it attenuated COX-2 levels and effectively protected fibroblasts against oxidative stress. To date, this appears to be the most comprehensive antidiabetic screening platform in existence and consequently provides the only feasible solution that will enable natural product antidiabetic research to progress to the point where natural products can be commercialised and incorporated into meaningful healthcare programs. Future research should be focussed towards further expanding this model by incorporating additional targets, more sophisticated cell culturing techniques, multiplexed high content screening assays and carrying out combination treatments that explore the antidiabetic effects of two or more crude extracts to establish whether they are capable of acting in a synergistic manner.
- Full Text:
- Date Issued: 2020
- Authors: Pringle, Nadine Alex
- Date: 2020
- Subjects: Hypoglycemic agents
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/46750 , vital:39653
- Description: Several synthetic antidiabetic drugs have been developed to date, however, most are accompanied by adverse side-effects while remaining expensive and largely inaccessible to the vast majority of those who need it. To provide enough scientific evidence to support the inclusion of more affordable African antidiabetic medicinal plants and macrofungi into healthcare programs, this study sought out to develop a comprehensive in vitro antidiabetic target-directed screening platform incorporating high content screening and analysis/ cellomics. To test the success of this model, the potential antidiabetic mechanisms of five plants (Aspalathus linearis, Brachylaena discolor, Carpobrotus deliciosus, Sutherlandia frutescens and Tarchonanthus camphoratus) and two macrofungal species (Ganoderma lucidum and Hericium erinaceus) were explored. The screening model consisted of approximately 22 assays exploring the antidiabetic effects of selected aqueous and ethanolic extracts in five well-characterised antidiabetic targets: the intestine, liver, skeletal muscle, adipose tissue/ obesity and pancreatic β-cells. These targets were further categorised and scored under three mechanistic classes/ therapeutic targets (postprandial hyperglycaemia; insulin resistance and inflammation; pancreatic β-cell function) to elucidate their potential mechanisms of action and select appropriate animal models for future studies. Almost any normal or diabetic rodent model would be suitable to explore the antidiabetic potential of extracts such as A. linearis, B. discolor ethanol, C. deliciosus ethanol or T. camphoratus which obtained high cumulative scores under postprandial hyperglycaemia while high fat diet and genetic models of obesity appear more suited towards extracts such as H. erinaceus aqueous that obtained their highest cumulative score under insulin resistance. In general, a combination of rodent models ranging from non-obese models to models of obesity and β-cell destruction presenting symptoms from all three mechanistic classes should be considered due to the pleiotropic nature of the tested extracts, however, establishing appropriate experimental designs is crucial. To demonstrate the versatility of the screening platform and emphasise the importance of in vitro screening pertaining to diabetic complications, a more detailed biochemical investigation into the potential therapeutic benefits of A. linearis in the treatment of diabetic wounds was conducted. Several properties supporting the therapeutic potential of rooibos were highlighted with the green and fermented extracts presenting distinctly different characteristics. The pro-inflammatory nature of fermented rooibos may have therapeutic value for wounds characterised with a delayed initial inflammatory phase, such as early diabetic wounds while the green extract appears more suited to wounds burdened with excessive inflammation as it attenuated COX-2 levels and effectively protected fibroblasts against oxidative stress. To date, this appears to be the most comprehensive antidiabetic screening platform in existence and consequently provides the only feasible solution that will enable natural product antidiabetic research to progress to the point where natural products can be commercialised and incorporated into meaningful healthcare programs. Future research should be focussed towards further expanding this model by incorporating additional targets, more sophisticated cell culturing techniques, multiplexed high content screening assays and carrying out combination treatments that explore the antidiabetic effects of two or more crude extracts to establish whether they are capable of acting in a synergistic manner.
- Full Text:
- Date Issued: 2020
Conservation biology of an apex predator in the Anthropocene : poaching, pastoralism and lions in multi-use landscapes, South-Eastern Africa
- Authors: Everatt, Kristoffer T
- Date: 2020
- Subjects: Conservation biology -- Africa, Southern
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/48290 , vital:40839
- Description: Many of the world’s apex predators are experiencing catastrophic declines as a result of competition with humans. Understanding the mechanisms and ecosystem impacts of apex predator declines is a fundamental ecological question crucial to conserving the Earth’s biodiversity and functioning ecosystems. In this thesis I used the African lion Panthera leo, as a model species to investigate the impacts of anthropogenic pressures on the conservation and ecology of an apex predator. Specifically, I investigated the relative and cumulative influences of pastoralism and poaching on lion occurrence, population connectivity, ecological role, space use, prey selection and viability across a 73 000 km2 multi-use landscape in southern Africa. Using landscape occupancy spoor surveys, I tested the hypotheses that lions were most limited by either interference or by exploitative competition with humans and identified thresholds of lion tolerance to human activities. My results showed that lions occupied only a fraction of the landscape and were limited by a combination of interference and exploitative competition with humans. Interference competition with pastoralism however was the biggest driver limiting lion occupancy, creating a clear disturbance threshold for lions cumulating in their near complete loss from the landscape. I employed call-up surveys, pride monitoring and mortality analysis to investigate the numerical impacts of anthropogenic pressures on the viability of a lion sub-population. I found that persecution by pastoralists was the greatest source of lion mortalities across the landscape. However, the targeted poaching of lions for body parts had emerged as the greatest threat to lions in a nominally protected National Park where I documented a steep population decline and collapse of lion prides. I used GPS tracking and diet analysis of lions at the human-wildlands interface to test if lions foraged optimally or were constrained by competition with humans. I fount that individual lions appeared to select for prey and habitat optimally, while also showing some level of risk avoidance towards anthropogenic pressures. I then considered landscape resistances to test if sink habitats may provide connectivity between sources or act as ecological traps. I found that potential connectivity for lions between the region’s two source populations was limited by a loss of habitat and prey. Furthermore, the impacts of by-catch in snares risked transforming the few remaining potential conservation corridors into ecological traps. Finally, I examined interactions between lions and syntopic mesopredators across gradients of anthropogenic pressures to test if the functional role of lions was affected by human pressures. I found that lions showed limited suppression of mesopredators, however, anthropogenic pressures increased lion’s interactions with syntopic predators. As an apex predator, lions have evolved limited capacity to mitigate top-down competitive pressures, however, like many of the world’s apex predators, they are becoming increasingly limited by anthropogenic pressures. This study provides a predictive understanding of an apex predator’s ecological responses to top-down anthropogenic pressures which can be applied globally to the question of conservation in the Anthropocene.
- Full Text:
- Date Issued: 2020
- Authors: Everatt, Kristoffer T
- Date: 2020
- Subjects: Conservation biology -- Africa, Southern
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/48290 , vital:40839
- Description: Many of the world’s apex predators are experiencing catastrophic declines as a result of competition with humans. Understanding the mechanisms and ecosystem impacts of apex predator declines is a fundamental ecological question crucial to conserving the Earth’s biodiversity and functioning ecosystems. In this thesis I used the African lion Panthera leo, as a model species to investigate the impacts of anthropogenic pressures on the conservation and ecology of an apex predator. Specifically, I investigated the relative and cumulative influences of pastoralism and poaching on lion occurrence, population connectivity, ecological role, space use, prey selection and viability across a 73 000 km2 multi-use landscape in southern Africa. Using landscape occupancy spoor surveys, I tested the hypotheses that lions were most limited by either interference or by exploitative competition with humans and identified thresholds of lion tolerance to human activities. My results showed that lions occupied only a fraction of the landscape and were limited by a combination of interference and exploitative competition with humans. Interference competition with pastoralism however was the biggest driver limiting lion occupancy, creating a clear disturbance threshold for lions cumulating in their near complete loss from the landscape. I employed call-up surveys, pride monitoring and mortality analysis to investigate the numerical impacts of anthropogenic pressures on the viability of a lion sub-population. I found that persecution by pastoralists was the greatest source of lion mortalities across the landscape. However, the targeted poaching of lions for body parts had emerged as the greatest threat to lions in a nominally protected National Park where I documented a steep population decline and collapse of lion prides. I used GPS tracking and diet analysis of lions at the human-wildlands interface to test if lions foraged optimally or were constrained by competition with humans. I fount that individual lions appeared to select for prey and habitat optimally, while also showing some level of risk avoidance towards anthropogenic pressures. I then considered landscape resistances to test if sink habitats may provide connectivity between sources or act as ecological traps. I found that potential connectivity for lions between the region’s two source populations was limited by a loss of habitat and prey. Furthermore, the impacts of by-catch in snares risked transforming the few remaining potential conservation corridors into ecological traps. Finally, I examined interactions between lions and syntopic mesopredators across gradients of anthropogenic pressures to test if the functional role of lions was affected by human pressures. I found that lions showed limited suppression of mesopredators, however, anthropogenic pressures increased lion’s interactions with syntopic predators. As an apex predator, lions have evolved limited capacity to mitigate top-down competitive pressures, however, like many of the world’s apex predators, they are becoming increasingly limited by anthropogenic pressures. This study provides a predictive understanding of an apex predator’s ecological responses to top-down anthropogenic pressures which can be applied globally to the question of conservation in the Anthropocene.
- Full Text:
- Date Issued: 2020
Creating Citizen-Science for Groundwater Monitoring prior to potential Shale Gas Development in Cradock (South-Eastern Karoo, South Africa)
- Authors: Dhliwayo, Nyaradzo
- Date: 2020
- Subjects: Natural gas -- Prospecting -- Environmental aspects -- South Africa
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/48224 , vital:40526
- Description: Citizen Science is an integral tool for community engagement in scientific project design, implementation, data interpretation and reporting, in the quest to promote local capacity development as well as scientific knowledge. Based on the relevance of public engagement in aspects of the natural environment and associated scientific issues associated with the risks and opportunities of potential Shale Gas Development (SGD), a Citizen-Science (CS) study was designed in the Eastern Karoo region of South Africa linked to a Shale Gas Baseline Study initiated by the Africa Earth Observatory Network (AEON). As the pilot area of the baseline study, the town of Cradock was chosen to be the study area for this research; and CS was used to facilitate the identification of the Cradock community’s abilities to monitor the effects of potential SGD in this region. The development of new knowledge, skills and support, as well as a deeper understanding of the community’s role in Citizen Science studies, was also facilitated by this process. This study demonstrates the feasibility of integrating Citizen Science into existing community structures in Africa, thereby encouraging community engagement in the developmental programmes through participatory methods. Using a collaborative Action-Research Approach, consistent community roundtable meetings and key informant interviews served as invaluable platforms for the establishment of a 7-member community working group, that played the liaison role between the researcher and the Cradock community in the CS implementation process. Coupled with this was the recruitment and the training of eight young citizens in conducting a hydro-census and groundwater sampling for six boreholes within the identified Cradock commonage farms, as well as two boreholes on the farm of an identified emergent farmer. The engagements between the researcher, AEON scientists, the community working group and the community at large, enabled the successful implementation of CS training in two groundwater monitoring aspects and the testing of eleven water quality parameters. The training process was combined with the design and the development of a customised ‘Xoras’ Online Application, which was used to capture and share the hydro-census data recorded. Experiential learning in hydro-census and groundwater sampling resulted subsequently in an increased understanding and awareness of these aspects (Figure A). Even if SGD does not materialise in the South-Eastern Karoo, CS training will enable communities in the Shale-Gas Development precincts to participate in local decision-making forums on ground water, health, or on any related regional development projects. It is anticipated that the adoption of CS will ix promote future community engagements, especially about water across this water-scarce region, allowing for greater community-voice representation in resource-policy decisions related to potential Shale-Gas and related natural resource industries in the Karoo.
- Full Text:
- Date Issued: 2020
- Authors: Dhliwayo, Nyaradzo
- Date: 2020
- Subjects: Natural gas -- Prospecting -- Environmental aspects -- South Africa
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/48224 , vital:40526
- Description: Citizen Science is an integral tool for community engagement in scientific project design, implementation, data interpretation and reporting, in the quest to promote local capacity development as well as scientific knowledge. Based on the relevance of public engagement in aspects of the natural environment and associated scientific issues associated with the risks and opportunities of potential Shale Gas Development (SGD), a Citizen-Science (CS) study was designed in the Eastern Karoo region of South Africa linked to a Shale Gas Baseline Study initiated by the Africa Earth Observatory Network (AEON). As the pilot area of the baseline study, the town of Cradock was chosen to be the study area for this research; and CS was used to facilitate the identification of the Cradock community’s abilities to monitor the effects of potential SGD in this region. The development of new knowledge, skills and support, as well as a deeper understanding of the community’s role in Citizen Science studies, was also facilitated by this process. This study demonstrates the feasibility of integrating Citizen Science into existing community structures in Africa, thereby encouraging community engagement in the developmental programmes through participatory methods. Using a collaborative Action-Research Approach, consistent community roundtable meetings and key informant interviews served as invaluable platforms for the establishment of a 7-member community working group, that played the liaison role between the researcher and the Cradock community in the CS implementation process. Coupled with this was the recruitment and the training of eight young citizens in conducting a hydro-census and groundwater sampling for six boreholes within the identified Cradock commonage farms, as well as two boreholes on the farm of an identified emergent farmer. The engagements between the researcher, AEON scientists, the community working group and the community at large, enabled the successful implementation of CS training in two groundwater monitoring aspects and the testing of eleven water quality parameters. The training process was combined with the design and the development of a customised ‘Xoras’ Online Application, which was used to capture and share the hydro-census data recorded. Experiential learning in hydro-census and groundwater sampling resulted subsequently in an increased understanding and awareness of these aspects (Figure A). Even if SGD does not materialise in the South-Eastern Karoo, CS training will enable communities in the Shale-Gas Development precincts to participate in local decision-making forums on ground water, health, or on any related regional development projects. It is anticipated that the adoption of CS will ix promote future community engagements, especially about water across this water-scarce region, allowing for greater community-voice representation in resource-policy decisions related to potential Shale-Gas and related natural resource industries in the Karoo.
- Full Text:
- Date Issued: 2020
Development of a UV-tolerant strain of the South African isolate of Cryptophlebia leucotreta granulovirus for use as an enhanced biopesticide for Thaumatotibia leucotreta control on citrus
- Authors: Mwanza, Patrick
- Date: 2020
- Subjects: Baculoviruses -- South Africa , Ultraviolet astronomy
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/48533 , vital:40885
- Description: Baculoviruses are pathogenic to insects in the orders Diptera, Hymenoptera, and Lepidoptera. As a result of this natural relationship with insects they provide an environmentally friendly method to combat crop and forest pests. As such, a number of baculoviruses have been formulated into biopesticides. The use of baculovirus biopesticides is gaining popularity as the use of chemical pesticides has come under stringent regulatory conditions imposed by governments and continental blocks such as the European Union. Baculoviruses have a narrow host range and therefore do not harm non-pests or humans who consume the crops. One such baculovirus is Cryptophlebia leucotreta granulovirus (CrleGV), which is pathogenic to the citrus pest Thaumatotibia leucotreta, commonly referred to as the false codling moth (FCM). CrleGV has an occlusion body (OB) that encloses a single virion. Several CrleGV biopesticides have been registered in South Africa for use on citrus, avocadoes, macadamias, grapes and other crops by two commercial producers, River Bioscience (SA) and Andermatt (Switzerland). These biopesticides are used as part of the FCM integrated pest management (IPM) programme, a multifacetted approach to controlling FCM. However, baculoviruses are susceptible to the ultraviolet (UV) radiation component of sunlight and lose their activity within hours to a few days, after exposure to UV. Several substances have been tested as UV protectants to improve the persistence of baculovirus biopesticides in the field. These include optical brighteners, UV absorbers and anti-oxidants. While very promising in the laboratory, UV-protectants have not been as successful in the field. A few published reports have reported, that UV-tolerant baculoviruses could be isolated from a population by repeatedly exposing and re-exposing the virus to UV irradiation with a propagation step in insect host fourth or fifth instars between each exposure cycle. In this study, the South African isolate of Cryptophlebia leucotreta (CrleGV-SA) was exposed to UV irradiation for 5 exposure cycles in a Q-Sun Xe-3 HC test chamber (Qlab, USA) with parameters set to mimic a typical summer day in the Sundays River Valley, Eastern Cape Province, in South Africa. In between exposures the virus survivors were allowed to multiply in FCM fifth instars. Surface dose bioassays were also conducted to determine the LC50 of the virus after each exposure cycle. Samples from exposure cycle 1 and cycle 5 (UV-tolerant) irradiated for 72 h were prepared for Next Generation Sequencing (NGS) of DNA. The resultant sequence data were analysed using the Geneious R11 software (New Zealand) and compared with the unexposed CrleGV-SA sequence. In-silico restriction enzyme analsysis (REN) with several enzymes was also carried on both the cycle 1 and cycle 5 exposed samples and the resulting digestion patterns were compared with the original CrleGV-SA digestion patterns. The same samples were also analysed by transmission electron microscopy (TEM) and Attenuated Reflectance Fourier Transform Infrared Spectroscopy (ATR-FTIR) to evaluate the effect of UV irradiation on the structure of the CrleGV-SA OB. In addition, three UV protectants, lignin sulphate (Sappi, SA), BREAK-THRU®OE446 (OE446) (Evonik Industries, Germany) and Uvinul Easy (BASF, Germany) were prepared with CrleGV-SA to give final protectant concentrations of 0.09 %, 0.9 % and 9 %. The protectant-virus suspensions were exposed to UV for 24 h in the Q-Sun test chamber and bioassays conducted to determine the protective effect of each protectant concentration. The most successful protectants were then combined with the UV-tolerant CrleGV-SA and exposed to UV for 24 h in the Q-Sun test chamber and surface dose bioassays conducted afterwards. Samples exposed to UV in cycle 5 had lower LC50 values compared to samples in the early cycles. With each re-exposure cycle the LC50 values moved closer to that of the unexposed control. The LC50 of virus samples decreased from 2.89 x 108 OBs/ml after 24 h UV-exposure in cycle 1 to 2.16 x 105 OBs/ml after the same duration of exposure in cycle 5; and from 2.11 x 109 OBs/ml in cycle 1 after 72 h UV-exposure to 1.73 x 106 OBs/ml after the same duration of exposure. This represented a 1338-fold difference and a 1220-fold difference, respectively. When the UV-tolerant samples were sequenced seven SNPs were identified in cycle 1, which were thought to help establish UV tolerance, while a further seven SNPs were identified in cycle 5 samples; these were thought to further establish and maintain the UV-tolerance. Additionally, REN analysis with EcoR1 for both test samples yielded digestion patterns that were different from those of the original CrleGV-SA. TEM data showed that UV damages the virion as well as the crystalline structure of the OB. This is the first time visual evidence for UV damage to baculoviruses has been published. Comparison of cycle 1 and cycle 5 UV exposed OBs revealed that the cycle 5 OBs were significantly larger than the cycle 1 OBs (P<0.05). In addition, several peaks in the fingerprint region were shown to have either appeared or disapeered from the ATR-FTIR spectra after UV irradiation. However, there was no difference in the spectra of the Cycle 1 and Cycle 5 virus samples. The tests with potential UV-protectants revealed that the 0.9 % lignin, 9 % OE446 and 9 % Uvinul Easy were the most effective in protecting the virus from UV. However, there was no significant difference in their protection of UV tolerant CrleGV-SA and wild type CrleGV-SA. Going forward, it is recommended that the 0.9 % lignin, 9 % OE446 and 9 % Uvinul Easy combinations be explored further in future studies, particulary in the field. This study therefore forms an important foundation for the development of UV-tolerant baculovirus that will last longer in the field.
- Full Text:
- Date Issued: 2020
- Authors: Mwanza, Patrick
- Date: 2020
- Subjects: Baculoviruses -- South Africa , Ultraviolet astronomy
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/48533 , vital:40885
- Description: Baculoviruses are pathogenic to insects in the orders Diptera, Hymenoptera, and Lepidoptera. As a result of this natural relationship with insects they provide an environmentally friendly method to combat crop and forest pests. As such, a number of baculoviruses have been formulated into biopesticides. The use of baculovirus biopesticides is gaining popularity as the use of chemical pesticides has come under stringent regulatory conditions imposed by governments and continental blocks such as the European Union. Baculoviruses have a narrow host range and therefore do not harm non-pests or humans who consume the crops. One such baculovirus is Cryptophlebia leucotreta granulovirus (CrleGV), which is pathogenic to the citrus pest Thaumatotibia leucotreta, commonly referred to as the false codling moth (FCM). CrleGV has an occlusion body (OB) that encloses a single virion. Several CrleGV biopesticides have been registered in South Africa for use on citrus, avocadoes, macadamias, grapes and other crops by two commercial producers, River Bioscience (SA) and Andermatt (Switzerland). These biopesticides are used as part of the FCM integrated pest management (IPM) programme, a multifacetted approach to controlling FCM. However, baculoviruses are susceptible to the ultraviolet (UV) radiation component of sunlight and lose their activity within hours to a few days, after exposure to UV. Several substances have been tested as UV protectants to improve the persistence of baculovirus biopesticides in the field. These include optical brighteners, UV absorbers and anti-oxidants. While very promising in the laboratory, UV-protectants have not been as successful in the field. A few published reports have reported, that UV-tolerant baculoviruses could be isolated from a population by repeatedly exposing and re-exposing the virus to UV irradiation with a propagation step in insect host fourth or fifth instars between each exposure cycle. In this study, the South African isolate of Cryptophlebia leucotreta (CrleGV-SA) was exposed to UV irradiation for 5 exposure cycles in a Q-Sun Xe-3 HC test chamber (Qlab, USA) with parameters set to mimic a typical summer day in the Sundays River Valley, Eastern Cape Province, in South Africa. In between exposures the virus survivors were allowed to multiply in FCM fifth instars. Surface dose bioassays were also conducted to determine the LC50 of the virus after each exposure cycle. Samples from exposure cycle 1 and cycle 5 (UV-tolerant) irradiated for 72 h were prepared for Next Generation Sequencing (NGS) of DNA. The resultant sequence data were analysed using the Geneious R11 software (New Zealand) and compared with the unexposed CrleGV-SA sequence. In-silico restriction enzyme analsysis (REN) with several enzymes was also carried on both the cycle 1 and cycle 5 exposed samples and the resulting digestion patterns were compared with the original CrleGV-SA digestion patterns. The same samples were also analysed by transmission electron microscopy (TEM) and Attenuated Reflectance Fourier Transform Infrared Spectroscopy (ATR-FTIR) to evaluate the effect of UV irradiation on the structure of the CrleGV-SA OB. In addition, three UV protectants, lignin sulphate (Sappi, SA), BREAK-THRU®OE446 (OE446) (Evonik Industries, Germany) and Uvinul Easy (BASF, Germany) were prepared with CrleGV-SA to give final protectant concentrations of 0.09 %, 0.9 % and 9 %. The protectant-virus suspensions were exposed to UV for 24 h in the Q-Sun test chamber and bioassays conducted to determine the protective effect of each protectant concentration. The most successful protectants were then combined with the UV-tolerant CrleGV-SA and exposed to UV for 24 h in the Q-Sun test chamber and surface dose bioassays conducted afterwards. Samples exposed to UV in cycle 5 had lower LC50 values compared to samples in the early cycles. With each re-exposure cycle the LC50 values moved closer to that of the unexposed control. The LC50 of virus samples decreased from 2.89 x 108 OBs/ml after 24 h UV-exposure in cycle 1 to 2.16 x 105 OBs/ml after the same duration of exposure in cycle 5; and from 2.11 x 109 OBs/ml in cycle 1 after 72 h UV-exposure to 1.73 x 106 OBs/ml after the same duration of exposure. This represented a 1338-fold difference and a 1220-fold difference, respectively. When the UV-tolerant samples were sequenced seven SNPs were identified in cycle 1, which were thought to help establish UV tolerance, while a further seven SNPs were identified in cycle 5 samples; these were thought to further establish and maintain the UV-tolerance. Additionally, REN analysis with EcoR1 for both test samples yielded digestion patterns that were different from those of the original CrleGV-SA. TEM data showed that UV damages the virion as well as the crystalline structure of the OB. This is the first time visual evidence for UV damage to baculoviruses has been published. Comparison of cycle 1 and cycle 5 UV exposed OBs revealed that the cycle 5 OBs were significantly larger than the cycle 1 OBs (P<0.05). In addition, several peaks in the fingerprint region were shown to have either appeared or disapeered from the ATR-FTIR spectra after UV irradiation. However, there was no difference in the spectra of the Cycle 1 and Cycle 5 virus samples. The tests with potential UV-protectants revealed that the 0.9 % lignin, 9 % OE446 and 9 % Uvinul Easy were the most effective in protecting the virus from UV. However, there was no significant difference in their protection of UV tolerant CrleGV-SA and wild type CrleGV-SA. Going forward, it is recommended that the 0.9 % lignin, 9 % OE446 and 9 % Uvinul Easy combinations be explored further in future studies, particulary in the field. This study therefore forms an important foundation for the development of UV-tolerant baculovirus that will last longer in the field.
- Full Text:
- Date Issued: 2020
Development of bio-based xylan composites for food packaging applications
- Authors: Naidu, Darrel Sarvesh
- Date: 2020
- Subjects: Xylanases
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/48544 , vital:40886
- Description: Currently a large number of chemicals and plastics are produced from petroleum-based resources. However, due to the concerns surrounding the depletion of petroleum resources and growing carbon emissions, there is a desire to produce chemicals and plastics from renewable and carbon natural sources. Lignocellulosic biomass (biomass consisting of cellulose, lignin and hemicellulose) is the most common biomass on earth and is renewable and carbon neutral. Of the three main constituents of lignocellulosic biomass, hemicellulose is composed of a mixture of sugars which can be converted into chemicals and plastics. The most common form of hemicellulose found in nature is xylan. This study is aimed at extracting xylan from maize stalk waste residues and the development of xylan films with properties that are suitable for food packaging applications. Xylan was extracted from maize stalk waste residues using an alkaline pre-treatment method. The effects of bleaching conditions (time, temperature and bleach concentration) prior to alkaline treatment on the yield and purity of xylan extracted was studied using a Box-Behnken experimental design. It was observed that the experimental conditions tested had no effect on the yield of the xylan extracted but the bleach concentration had a significant effect on the purity of the xylan extracted. The samples with the lowest lignin content were found to be the samples treated with 3wt% bleach prior to alkaline pre-treatment. One of the main disadvantages of xylan is that it has poor film forming properties, a method of overcoming this is to combine it with another abundantly available biopolymer that has good film forming properties, such as alginate. The effect of xylan, alginate and glycerol content on the mechanical, thermal, moisture uptake and water barrier properties of the films were investigated. It was observed that with an increase in the alginate content there was an increase in the tensile strength and Young’s modulus of the films, whereas the water vapour permeability (WVP) of the films decreased. This was attributed to the greater cohesion between alginate polymer chains compared to the cohesion between xylan polymer chains. The xylanalginate films exhibited better optical and water sorption properties at higher xylan content. In order to improve the inherently poor mechanical and barrier properties of the xylan-alginate films, bentonite and halloysite were incorporated into the films. It was found that 5wt% incorporation of either bentonite or halloysite resulted in a 49% decrease of the WVP, which was attributed to water vapour impermeable nature of the silicate layers that make up both clays. The incorporation of the clays into the xylan-alginate matrix resulted in a significant
- Full Text:
- Date Issued: 2020
- Authors: Naidu, Darrel Sarvesh
- Date: 2020
- Subjects: Xylanases
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/48544 , vital:40886
- Description: Currently a large number of chemicals and plastics are produced from petroleum-based resources. However, due to the concerns surrounding the depletion of petroleum resources and growing carbon emissions, there is a desire to produce chemicals and plastics from renewable and carbon natural sources. Lignocellulosic biomass (biomass consisting of cellulose, lignin and hemicellulose) is the most common biomass on earth and is renewable and carbon neutral. Of the three main constituents of lignocellulosic biomass, hemicellulose is composed of a mixture of sugars which can be converted into chemicals and plastics. The most common form of hemicellulose found in nature is xylan. This study is aimed at extracting xylan from maize stalk waste residues and the development of xylan films with properties that are suitable for food packaging applications. Xylan was extracted from maize stalk waste residues using an alkaline pre-treatment method. The effects of bleaching conditions (time, temperature and bleach concentration) prior to alkaline treatment on the yield and purity of xylan extracted was studied using a Box-Behnken experimental design. It was observed that the experimental conditions tested had no effect on the yield of the xylan extracted but the bleach concentration had a significant effect on the purity of the xylan extracted. The samples with the lowest lignin content were found to be the samples treated with 3wt% bleach prior to alkaline pre-treatment. One of the main disadvantages of xylan is that it has poor film forming properties, a method of overcoming this is to combine it with another abundantly available biopolymer that has good film forming properties, such as alginate. The effect of xylan, alginate and glycerol content on the mechanical, thermal, moisture uptake and water barrier properties of the films were investigated. It was observed that with an increase in the alginate content there was an increase in the tensile strength and Young’s modulus of the films, whereas the water vapour permeability (WVP) of the films decreased. This was attributed to the greater cohesion between alginate polymer chains compared to the cohesion between xylan polymer chains. The xylanalginate films exhibited better optical and water sorption properties at higher xylan content. In order to improve the inherently poor mechanical and barrier properties of the xylan-alginate films, bentonite and halloysite were incorporated into the films. It was found that 5wt% incorporation of either bentonite or halloysite resulted in a 49% decrease of the WVP, which was attributed to water vapour impermeable nature of the silicate layers that make up both clays. The incorporation of the clays into the xylan-alginate matrix resulted in a significant
- Full Text:
- Date Issued: 2020
Genetic connectivity in sandy beach macroinfauna with contrasting life histories and zonation patterns
- Authors: Bezuidenhout, Karien
- Date: 2020
- Subjects: Fishes -- Genetics , Seashore ecology -- South Africa
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/45944 , vital:39380
- Description: Open-ocean sandy beaches are highly dynamic ecosystems that continuously respond to waves, tides and storms. These beaches support a suite of uniquely adapted fauna that exhibits a high degree of endemism. This diversity, which also includes some economically important resource species, contributes to the great socio-economic and ecological value of sandy beach systems, which provide ecosystem services such as nutrient recycling and water purification, fish nursery habitat (surf zone), and natural buffering against storms. Despite their broad global distribution comprising ~70 % of the world’s ice-free coastline, beaches are naturally fragmented by other intertidal marine habitats and extensive coastal development. Sea-level rise is now threatening beaches, particularly along developed coasts through coastal squeeze, with an increased risk of habitat loss and exacerbated fragmentation. Such habitat loss has important implications for population and species persistence, depending on the level of connectivity among populations. While sandy beach macroinfauna tend to be highly mobile across and along shore, they also show a strong bias towards direct, or aplanktonic, development; the latter could reduce opportunities for dispersal and sustained population connectivity. However, comparatively little is known about population connectivity on sandy beaches, limiting scientifically sound management of much of the world’s coastline. This study aims to address this knowledge gap by working towards a conceptual model quantifying population connectivity, and by investigating the specific roles of life history, across-shore zonation, and geographic range as putative indicators and possibly drivers of genetic differentiation (ΦST) and connectivity (gene flow) among beach macroinvertebrate populations. The main hypotheses are that 1) species with a pelagic larval stage will display lower genetic differentiation with higher rates of gene flow among populations than species with a benthic lecithotrophic larval stage or direct development; and 2) genetic differentiation among populations will decrease along an across-shore axis, from the supratidal to the low water mark. Two opposing hypotheses were tested concerning the relationship between genetic structure and distributional range: 3a) Species with a large range (measured as distance and number of latitudes) will show less genetic differentiation than those with a narrow range, because they are more able to disperse; or 3b) more widely distributed species may occur across more phylogenetic | v breaks, and thus show higher genetic differentiation than species restricted to a smaller ranges and single biogeographic provinces. First, the extent and likely drivers of population connectivity were investigated for a suite of temperate South African macroinfauna. Very low, non-significant genetic differentiation was demonstrated for the beach clam Donax serra, thus supporting the hypothesis that pelagic larval dispersers display low/no genetic differentiation among populations (Chapter 2). Statistical power of the COI locus was, however, low in this species, due to reduced genetic variability which could be attributed to an expansion from a small source population into extant habitats. The results highlighted the need for caution when interpreting a lack of genetic differentiation as evidence for population connectivity. Nevertheless, COI sequences verified that D. serra occupying the cool and warm-temperate regions of the South African coast, constitute a single species, despite morphological and behavioural differences between inhabitants of these two bioregions. D. sordidus, a congener of D. serra, and endemic to the warm-temperate (Agulhas) bioregion of South Africa, was investigated to test two opposing hypotheses (Chapter 3): 1) this pelagic larval disperser lack genetic differentiation with high rates of gene flow across its range, or alternatively 2) these clams are genetically structured across their range due to the influence of freshwater input (measured as distance to nearest estuary), and morphodynamics of the surf zone (represented by surf-zone width). Significant global ΦST was estimated for D. sordidus, with differentiation predominantly driven by haplotype frequencies at a single study site and estimates of gene flow among most sample locations generally high. Tentative evidence suggested that natural selection driven by differences in water salinity due to estuarine runoff and surf-zone width, might be maintaining two dominant haplotypes across the range of this species. Beach clams from the Southern Hemisphere, occupying different distributional ranges, were compared to test the hypothesis that genetic structure decreases with an increase in range. This hypothesis was poorly supported by the results; opposite (yet mostly non-significant) trends demonstrated for ΦST and haplotype diversity, thus providing some support for the alternative hypothesis that more phylogeographic breaks in broad-ranging species results in higher estimates of genetic structure in these species.
- Full Text:
- Date Issued: 2020
- Authors: Bezuidenhout, Karien
- Date: 2020
- Subjects: Fishes -- Genetics , Seashore ecology -- South Africa
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/45944 , vital:39380
- Description: Open-ocean sandy beaches are highly dynamic ecosystems that continuously respond to waves, tides and storms. These beaches support a suite of uniquely adapted fauna that exhibits a high degree of endemism. This diversity, which also includes some economically important resource species, contributes to the great socio-economic and ecological value of sandy beach systems, which provide ecosystem services such as nutrient recycling and water purification, fish nursery habitat (surf zone), and natural buffering against storms. Despite their broad global distribution comprising ~70 % of the world’s ice-free coastline, beaches are naturally fragmented by other intertidal marine habitats and extensive coastal development. Sea-level rise is now threatening beaches, particularly along developed coasts through coastal squeeze, with an increased risk of habitat loss and exacerbated fragmentation. Such habitat loss has important implications for population and species persistence, depending on the level of connectivity among populations. While sandy beach macroinfauna tend to be highly mobile across and along shore, they also show a strong bias towards direct, or aplanktonic, development; the latter could reduce opportunities for dispersal and sustained population connectivity. However, comparatively little is known about population connectivity on sandy beaches, limiting scientifically sound management of much of the world’s coastline. This study aims to address this knowledge gap by working towards a conceptual model quantifying population connectivity, and by investigating the specific roles of life history, across-shore zonation, and geographic range as putative indicators and possibly drivers of genetic differentiation (ΦST) and connectivity (gene flow) among beach macroinvertebrate populations. The main hypotheses are that 1) species with a pelagic larval stage will display lower genetic differentiation with higher rates of gene flow among populations than species with a benthic lecithotrophic larval stage or direct development; and 2) genetic differentiation among populations will decrease along an across-shore axis, from the supratidal to the low water mark. Two opposing hypotheses were tested concerning the relationship between genetic structure and distributional range: 3a) Species with a large range (measured as distance and number of latitudes) will show less genetic differentiation than those with a narrow range, because they are more able to disperse; or 3b) more widely distributed species may occur across more phylogenetic | v breaks, and thus show higher genetic differentiation than species restricted to a smaller ranges and single biogeographic provinces. First, the extent and likely drivers of population connectivity were investigated for a suite of temperate South African macroinfauna. Very low, non-significant genetic differentiation was demonstrated for the beach clam Donax serra, thus supporting the hypothesis that pelagic larval dispersers display low/no genetic differentiation among populations (Chapter 2). Statistical power of the COI locus was, however, low in this species, due to reduced genetic variability which could be attributed to an expansion from a small source population into extant habitats. The results highlighted the need for caution when interpreting a lack of genetic differentiation as evidence for population connectivity. Nevertheless, COI sequences verified that D. serra occupying the cool and warm-temperate regions of the South African coast, constitute a single species, despite morphological and behavioural differences between inhabitants of these two bioregions. D. sordidus, a congener of D. serra, and endemic to the warm-temperate (Agulhas) bioregion of South Africa, was investigated to test two opposing hypotheses (Chapter 3): 1) this pelagic larval disperser lack genetic differentiation with high rates of gene flow across its range, or alternatively 2) these clams are genetically structured across their range due to the influence of freshwater input (measured as distance to nearest estuary), and morphodynamics of the surf zone (represented by surf-zone width). Significant global ΦST was estimated for D. sordidus, with differentiation predominantly driven by haplotype frequencies at a single study site and estimates of gene flow among most sample locations generally high. Tentative evidence suggested that natural selection driven by differences in water salinity due to estuarine runoff and surf-zone width, might be maintaining two dominant haplotypes across the range of this species. Beach clams from the Southern Hemisphere, occupying different distributional ranges, were compared to test the hypothesis that genetic structure decreases with an increase in range. This hypothesis was poorly supported by the results; opposite (yet mostly non-significant) trends demonstrated for ΦST and haplotype diversity, thus providing some support for the alternative hypothesis that more phylogeographic breaks in broad-ranging species results in higher estimates of genetic structure in these species.
- Full Text:
- Date Issued: 2020
Ichthyofaunal and isotope changes along the rivere-stuarine continuum in a fluvially dominated Southern African coastal system
- Authors: Nashima, Festus Panduleni
- Date: 2020
- Subjects: Fishes -- Ecology
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/48555 , vital:40890
- Description: Catchment to coast and the continuum concept recognize linkages and make it imperative to take an ecosystem approach to catchment management, water allocation, fisheries management and environmental management as a whole. Fishes that utilise estuarine environments are influenced by physical and chemical conditions as well as biological interactions. Community structure, drivers and trophic linkages between species of different origin, utilising the low salinity waters along the riverestuarine continuum remain poorly understood in the lower Orange River and estuary. This body of work investigated the spatio-temporal dynamics structuring fishes in the lower Orange River Estuary Continuum (OREC) within the context of River Continuum (Vannote et al., 1980). Eight years (2004 - 2018) of seasonal sampling at 18 sites comprised of multi-species catches were analysed, including three years (2016 - 2018) exploring trophic linkages between the dominant species of marine, estuarine and freshwater fishes in the system. In total, 30 species belonging to 16 families were collected in the lower OREC. The Family Mugilidae dominated all catches. Fish species composition was comprised mostly of freshwater (47%), followed by the marine (33%) and estuarine species (20%). Numerically, the marine euryhaline Chelon richardsonii and solely estuarine Gilchristella aestuaria were the dominant species in the estuary region (site 1 - 11) with 95% and 4% contribution, respectively. Chelon richardsonii also dominated all catches (67%) upstream into the river (site 12 - 18), followed by freshwater Labeobarbus aeneus (13%) and Pseudocrenilabrus philander (8%). This highlights the point that besides the marine Chelon richardsonii, other species albeit marine, estuarine and freshwater are utilising unique areas within the OREC. Temporal differentiation in fish abundance (catch-per-unit-effort, CPUE), diversity, richness and evenness were higher in the high-flow season than the low-flow season. Mean CPUE was higher at the mouth of the estuary and declined upstream into the river, whilst species diversity (including richness and evenness) increased upstream. High catches but low species diversity observed close to the mouth of the estuary were attributed to the domination of Chelon richardsonii. However, species diversity increased upstream in the river due to the dominance of freshwater fishes coupled with several marine and estuarine species. This diversity patterns were inconsistent with the general Remane and revised model for estuaries, hence a new RiverEstuarine Continuum concept suitable for OREC was developed. With the exception of Chelon richardsonii, spatial variability in the length-frequency distribution of most fishes was observed. Physico-chemical drivers provided valuable information on adaptation limits of species movement and distribution. Fishes were structured along a salinity gradient and interaction between temperature and salinity influenced community structure and diversity patterns in the OREC. Fish species utilising the lower OREC are predominantly zoobenthivores but a few detritivores/planktivores dominate numerically. Trophic inferences from stable isotope analysis showed increased carbon isotope values (δ 13C) with salinity toward the mouth of the estuary, whilst nitrogen isotope values (δ 15N) increased upstream into the river. Ontogenetic shifts in the SI ratios of all five fish species (marine Chelon richardsonii; estuarine Gilchristella aestuaria; freshwater Labeobarbus aeneus, Pseudocrenilabrus philander and Mesobola brevianalis) were observed in the OREC. Variability in seasons and salinity gradient influenced the SI ratios of several fish species. Seasonal changes influenced the δ 13C ratios of all species, except for Labeobarbus aeneus and Pseudocrenilabrus philander whereas the δ 15N ratios of all species (except for Mesobola brevianalis) differed significantly between the high-flow and low-flow season. This can be due to their feeding on specific life-history stages, such as eggs, larvae, gametophytes, sporophytes or zoospores of prey. Stable isotope ratios revealed that Mesobola brevianalis fed at a higher trophic level than Chelon richardsonii and Gilchristella aestuaria. Mesobola brevianalis feed on planktonic crustacean and insects. Chelon richardsonii feed on a wide range of prey, with prey switching between planktonic and detritivore food resources, whilst Gilchristella aestuaria feed selectively on zooplankton but switched to filter feeding modes in turbid water. Generally, δ 15N SI ratios increase with growth, however, for Labeobarbus aeneus, it was significantly higher in small size length than in large size individuals, probably because most of the juveniles assessed were caught upstream where δ 15N increases. Juveniles of Labeobarbus aeneus mainly feed on zoobenthos whilst in larger fish phytoplankton becomes increasingly important, suggesting the possibility that zoobenthos was more 15N enriched than plant materials. The trophic niches of Mesobola brevianalis were isotopically distinguishable from Chelon richardsonii and Gilchristella aestuaria during the low-flow season, revealing that these species are probably accessing carbon sources from different food chains. During high-flow season, the isotopic niche overlapped among several species, indicating that these species are probably deriving carbon sources from the same or similar food chains. Results from Layman isotope-based metrics revealed that Chelon richardsonii, Pseudocrenilabrus philander and Labeobarbus aeneus occupied the widest isotopic niche in the lower OREC, suggesting that these species are utilising food sources which may be derived from a wide variety of carbon sources. Mesobola brevianalis had the narrowest isotopic niche, however this does not imply "a narrow range of food sources’’ for the species. Overall, the potential carbon sources for selected fish species revealed that they are supported by a variety of food sources (mainly phyto-and zooplanktonic) or their food items constituted different basal sources which enable them to coexist. This study represents the first, integrated investigation of trophic relations among fishes of different origin using δ 13C and δ15N SI analyses. The current study provides a broader understanding of fish ecology within the River-Estuarine Continuum in the lower Orange River and estuary. By filling research gaps identified in the lower OREC, this synthesis of work also supports and informs management, conservation and responsible utilisation of estuarine fishery resources, in particular within the OREC.
- Full Text:
- Date Issued: 2020
- Authors: Nashima, Festus Panduleni
- Date: 2020
- Subjects: Fishes -- Ecology
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/48555 , vital:40890
- Description: Catchment to coast and the continuum concept recognize linkages and make it imperative to take an ecosystem approach to catchment management, water allocation, fisheries management and environmental management as a whole. Fishes that utilise estuarine environments are influenced by physical and chemical conditions as well as biological interactions. Community structure, drivers and trophic linkages between species of different origin, utilising the low salinity waters along the riverestuarine continuum remain poorly understood in the lower Orange River and estuary. This body of work investigated the spatio-temporal dynamics structuring fishes in the lower Orange River Estuary Continuum (OREC) within the context of River Continuum (Vannote et al., 1980). Eight years (2004 - 2018) of seasonal sampling at 18 sites comprised of multi-species catches were analysed, including three years (2016 - 2018) exploring trophic linkages between the dominant species of marine, estuarine and freshwater fishes in the system. In total, 30 species belonging to 16 families were collected in the lower OREC. The Family Mugilidae dominated all catches. Fish species composition was comprised mostly of freshwater (47%), followed by the marine (33%) and estuarine species (20%). Numerically, the marine euryhaline Chelon richardsonii and solely estuarine Gilchristella aestuaria were the dominant species in the estuary region (site 1 - 11) with 95% and 4% contribution, respectively. Chelon richardsonii also dominated all catches (67%) upstream into the river (site 12 - 18), followed by freshwater Labeobarbus aeneus (13%) and Pseudocrenilabrus philander (8%). This highlights the point that besides the marine Chelon richardsonii, other species albeit marine, estuarine and freshwater are utilising unique areas within the OREC. Temporal differentiation in fish abundance (catch-per-unit-effort, CPUE), diversity, richness and evenness were higher in the high-flow season than the low-flow season. Mean CPUE was higher at the mouth of the estuary and declined upstream into the river, whilst species diversity (including richness and evenness) increased upstream. High catches but low species diversity observed close to the mouth of the estuary were attributed to the domination of Chelon richardsonii. However, species diversity increased upstream in the river due to the dominance of freshwater fishes coupled with several marine and estuarine species. This diversity patterns were inconsistent with the general Remane and revised model for estuaries, hence a new RiverEstuarine Continuum concept suitable for OREC was developed. With the exception of Chelon richardsonii, spatial variability in the length-frequency distribution of most fishes was observed. Physico-chemical drivers provided valuable information on adaptation limits of species movement and distribution. Fishes were structured along a salinity gradient and interaction between temperature and salinity influenced community structure and diversity patterns in the OREC. Fish species utilising the lower OREC are predominantly zoobenthivores but a few detritivores/planktivores dominate numerically. Trophic inferences from stable isotope analysis showed increased carbon isotope values (δ 13C) with salinity toward the mouth of the estuary, whilst nitrogen isotope values (δ 15N) increased upstream into the river. Ontogenetic shifts in the SI ratios of all five fish species (marine Chelon richardsonii; estuarine Gilchristella aestuaria; freshwater Labeobarbus aeneus, Pseudocrenilabrus philander and Mesobola brevianalis) were observed in the OREC. Variability in seasons and salinity gradient influenced the SI ratios of several fish species. Seasonal changes influenced the δ 13C ratios of all species, except for Labeobarbus aeneus and Pseudocrenilabrus philander whereas the δ 15N ratios of all species (except for Mesobola brevianalis) differed significantly between the high-flow and low-flow season. This can be due to their feeding on specific life-history stages, such as eggs, larvae, gametophytes, sporophytes or zoospores of prey. Stable isotope ratios revealed that Mesobola brevianalis fed at a higher trophic level than Chelon richardsonii and Gilchristella aestuaria. Mesobola brevianalis feed on planktonic crustacean and insects. Chelon richardsonii feed on a wide range of prey, with prey switching between planktonic and detritivore food resources, whilst Gilchristella aestuaria feed selectively on zooplankton but switched to filter feeding modes in turbid water. Generally, δ 15N SI ratios increase with growth, however, for Labeobarbus aeneus, it was significantly higher in small size length than in large size individuals, probably because most of the juveniles assessed were caught upstream where δ 15N increases. Juveniles of Labeobarbus aeneus mainly feed on zoobenthos whilst in larger fish phytoplankton becomes increasingly important, suggesting the possibility that zoobenthos was more 15N enriched than plant materials. The trophic niches of Mesobola brevianalis were isotopically distinguishable from Chelon richardsonii and Gilchristella aestuaria during the low-flow season, revealing that these species are probably accessing carbon sources from different food chains. During high-flow season, the isotopic niche overlapped among several species, indicating that these species are probably deriving carbon sources from the same or similar food chains. Results from Layman isotope-based metrics revealed that Chelon richardsonii, Pseudocrenilabrus philander and Labeobarbus aeneus occupied the widest isotopic niche in the lower OREC, suggesting that these species are utilising food sources which may be derived from a wide variety of carbon sources. Mesobola brevianalis had the narrowest isotopic niche, however this does not imply "a narrow range of food sources’’ for the species. Overall, the potential carbon sources for selected fish species revealed that they are supported by a variety of food sources (mainly phyto-and zooplanktonic) or their food items constituted different basal sources which enable them to coexist. This study represents the first, integrated investigation of trophic relations among fishes of different origin using δ 13C and δ15N SI analyses. The current study provides a broader understanding of fish ecology within the River-Estuarine Continuum in the lower Orange River and estuary. By filling research gaps identified in the lower OREC, this synthesis of work also supports and informs management, conservation and responsible utilisation of estuarine fishery resources, in particular within the OREC.
- Full Text:
- Date Issued: 2020
Investigating the fitness of sea turtles nesting in South Africa
- Le Gouvello du Timat, Diane Z M
- Authors: Le Gouvello du Timat, Diane Z M
- Date: 2020
- Subjects: Sea turtles -- South Africa
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/48445 , vital:40877
- Description: The South African rookeries of the South Western Indian Ocean (SWIO) loggerhead and leatherback regional management units have been continuously monitored for more than 5 decades, representing one of the longest-running sea turtle beach protection and monitoring programmes in the world. Although such conservation has been successful in protecting adults on land when they come to nest, it might not be enough to ensure population recovery if survival is not ensured across all life stages. Sea turtles are particularly vulnerable to direct threats and changing environmental conditions because they are long-lived, slow maturing, and have low survival rates at early life stages, making them conservation-dependent. These characteristics also render them difficult to conserve as a consequence of their wide geographic distributions, spanning multiple jurisdictions and legislative systems with different conservation targets, and a complex life cycle with changes in habitats and diets at different life stages. Sea turtle populations have been severely depleted as a result of historical over-exploitation and current threats including fisheries bycatch, direct harvest, habitat loss, pollution and pathogens, and climate change. In many cases however, population declines have been reversed by a variety of effective conservation measures such as beach protection programs, and marine protected areas, enabling successful population recovery around the world. Despite protection and apparent increase in abundance, indirect pressures can still affect turtle populations. Therefore, it is fundamental to evaluate the effectiveness of conservation measures (using fitness traits such as female and hatchling sizes) and examine underlying trends. The aim of this thesis was thus to evaluate the fitness of loggerhead and leatherback sea turtles nesting in South Africa. Long-term female size trends over time were evaluated and the populations were split between first-time nesters and experienced nesters. As sea turtle reproductive output (clutch size only) is correlated with female size and evidence is showing that in some populations individual female size is decreasing, the effects of maternal body size on reproductive output and hatchling fitness were investigated for both species. Crawling and swimming speeds were used as proxies of hatchling fitness, because these two traits positively influence their survival. An oceanic model was further used in combination with a particle tracking framework to estimate, for the first time, post-hatchling dispersal of loggerhead and leatherback turtles in the SWIO. I investigated whether the trend in female size for loggerhead turtles nesting in South Africa is a result of their foraging strategy to determine if there is a foraging dichotomy (neritic and oceanic). Despite an apparent increasing population, a significant 3.7 and 3.6 cm decrease in the size of both recruit and remingrant female loggerhead turtles, but a stable mean size of leatherbacks for both age groups was observed. I hypothesise that these contradictory effects are attributable to the very distinct foraging ecologies of the two turtle species and speculate that these variations in carapace length may be driven by environmental change. The reproductive strategy of both loggerhead and leatherback turtles nesting in South Africa supports the optimal egg size theory with larger turtles producing larger clutches but not larger eggs. Egg size was thus constant irrespective of maternal length. Hatchlings with longer carapaces and flippers had higher swimming speed, suggesting that larger hatchlings are fitter (potentially higher survival) than smaller ones by spending less time in high predation risk areas (i.e. beach and nearshore). The model revealed that dispersal trajectories of both loggerhead and leatherback sea turtle hatchlings were very similar and simulations indicated that initial active swimming (frenzy) as well as variability in oceanic conditions strongly influenced dispersal of virtual hatchlings. The trends observed in this study emphasize the importance of long-term monitoring, examining not only abundance but also individual size, to understand population dynamics, support recovery planning and prioritize future conservation practices. Furthermore, the findings of this study have important implications for overall population growth if smaller hatchlings of lower fitness are produced as a result of climate change and thus have decreased dispersal abilities and lower likelihood of survival.
- Full Text:
- Date Issued: 2020
- Authors: Le Gouvello du Timat, Diane Z M
- Date: 2020
- Subjects: Sea turtles -- South Africa
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/48445 , vital:40877
- Description: The South African rookeries of the South Western Indian Ocean (SWIO) loggerhead and leatherback regional management units have been continuously monitored for more than 5 decades, representing one of the longest-running sea turtle beach protection and monitoring programmes in the world. Although such conservation has been successful in protecting adults on land when they come to nest, it might not be enough to ensure population recovery if survival is not ensured across all life stages. Sea turtles are particularly vulnerable to direct threats and changing environmental conditions because they are long-lived, slow maturing, and have low survival rates at early life stages, making them conservation-dependent. These characteristics also render them difficult to conserve as a consequence of their wide geographic distributions, spanning multiple jurisdictions and legislative systems with different conservation targets, and a complex life cycle with changes in habitats and diets at different life stages. Sea turtle populations have been severely depleted as a result of historical over-exploitation and current threats including fisheries bycatch, direct harvest, habitat loss, pollution and pathogens, and climate change. In many cases however, population declines have been reversed by a variety of effective conservation measures such as beach protection programs, and marine protected areas, enabling successful population recovery around the world. Despite protection and apparent increase in abundance, indirect pressures can still affect turtle populations. Therefore, it is fundamental to evaluate the effectiveness of conservation measures (using fitness traits such as female and hatchling sizes) and examine underlying trends. The aim of this thesis was thus to evaluate the fitness of loggerhead and leatherback sea turtles nesting in South Africa. Long-term female size trends over time were evaluated and the populations were split between first-time nesters and experienced nesters. As sea turtle reproductive output (clutch size only) is correlated with female size and evidence is showing that in some populations individual female size is decreasing, the effects of maternal body size on reproductive output and hatchling fitness were investigated for both species. Crawling and swimming speeds were used as proxies of hatchling fitness, because these two traits positively influence their survival. An oceanic model was further used in combination with a particle tracking framework to estimate, for the first time, post-hatchling dispersal of loggerhead and leatherback turtles in the SWIO. I investigated whether the trend in female size for loggerhead turtles nesting in South Africa is a result of their foraging strategy to determine if there is a foraging dichotomy (neritic and oceanic). Despite an apparent increasing population, a significant 3.7 and 3.6 cm decrease in the size of both recruit and remingrant female loggerhead turtles, but a stable mean size of leatherbacks for both age groups was observed. I hypothesise that these contradictory effects are attributable to the very distinct foraging ecologies of the two turtle species and speculate that these variations in carapace length may be driven by environmental change. The reproductive strategy of both loggerhead and leatherback turtles nesting in South Africa supports the optimal egg size theory with larger turtles producing larger clutches but not larger eggs. Egg size was thus constant irrespective of maternal length. Hatchlings with longer carapaces and flippers had higher swimming speed, suggesting that larger hatchlings are fitter (potentially higher survival) than smaller ones by spending less time in high predation risk areas (i.e. beach and nearshore). The model revealed that dispersal trajectories of both loggerhead and leatherback sea turtle hatchlings were very similar and simulations indicated that initial active swimming (frenzy) as well as variability in oceanic conditions strongly influenced dispersal of virtual hatchlings. The trends observed in this study emphasize the importance of long-term monitoring, examining not only abundance but also individual size, to understand population dynamics, support recovery planning and prioritize future conservation practices. Furthermore, the findings of this study have important implications for overall population growth if smaller hatchlings of lower fitness are produced as a result of climate change and thus have decreased dispersal abilities and lower likelihood of survival.
- Full Text:
- Date Issued: 2020
Investigation of potential induced degradation as a performance limiting defect in photovoltaic modules
- Authors: Kwembur, Isaac Morko
- Date: 2020
- Subjects: Photovoltaic cells
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/48423 , vital:40875
- Description: Potential Induced Degradation (PID) impacts negatively on photovoltaic (PV) module durability because it significantly affects the output of PV modules and systems. Unless detected at infancy, PID progression can be catastrophic. This study involved systematic PID stressing of PV modules using a custom-built environmental chamber that can achieve suitable environmental conditions, viz., of the 35 °C ± 1 °C and relative humidity of 75 % ± 5 %. The first part of this work was to induce PID using three approaches: climate chamber testing, inducing PID using a conductive aluminium plate on the surface of the module without touching the frame and a localised PID induction on one cell in a module. The second part is to detect induced PID using Electroluminescence (EL) images taken at current corresponding to 10% Isc, EL histograms analysis and Voc ratio taken at 1000 W/m2 to 200 W/m2 . The third part is to study module regeneration after PID shunting degradation in two ways, viz., forced reverse polarization and natural recovery. The PID detection tools used in this work are well known module characterization techniques such as EL imaging, Infrared imaging, and light and dark current-voltage measurements. These characterisation tools are used in combination to detect defects such as optical losses, cracks, breakage, electric circuit degradation and PID. Under normal testing PID was detected and in some cases, modules were able to recover, while for advanced stage PID regeneration or PID reversal was difficult. This thesis focuses on PID detection at infancy using three approaches; EL imaging at current corresponding to 10% of Isc. Light and dark current – voltage measurements (L-IV & D-IV) and open circuit voltage (Voc) ratios at low irradiance. The early detection procedures are essential in reversing the degradation caused by PID which is reversible. The time taken to reverse the PID degradation will depend on the extent of the degradation. If detected early, it will take a short period of time to completely reverse lost power. Infrared thermography is a non-contact characteristic tool that can be deployed in large scale plants using drones to detect the presence of PID in PV plants. Module performance and device parameters extracted from the L-IV curves on a module before and after PID stress, such as Pmpp, Voc, Isc Fill Factor (FF), shunt resistance (Rsh) and series resistance (Rs) and ideality (n) are sensitive to PID shunting. Voc and Rsh drop significantly with the onset of PID, while Rs increases. The decrease in Voc and Rsh is due to heavy shunting on the module resulting in increased carrier recombination, while the increase in Rs is due to increased shunting paths leading to decreased photocurrent. When substantial degradation on a module occurs Pmpp, FF and n will drop and at very advanced stage of PID degradation Isc may drop excessively.
- Full Text:
- Date Issued: 2020
- Authors: Kwembur, Isaac Morko
- Date: 2020
- Subjects: Photovoltaic cells
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/48423 , vital:40875
- Description: Potential Induced Degradation (PID) impacts negatively on photovoltaic (PV) module durability because it significantly affects the output of PV modules and systems. Unless detected at infancy, PID progression can be catastrophic. This study involved systematic PID stressing of PV modules using a custom-built environmental chamber that can achieve suitable environmental conditions, viz., of the 35 °C ± 1 °C and relative humidity of 75 % ± 5 %. The first part of this work was to induce PID using three approaches: climate chamber testing, inducing PID using a conductive aluminium plate on the surface of the module without touching the frame and a localised PID induction on one cell in a module. The second part is to detect induced PID using Electroluminescence (EL) images taken at current corresponding to 10% Isc, EL histograms analysis and Voc ratio taken at 1000 W/m2 to 200 W/m2 . The third part is to study module regeneration after PID shunting degradation in two ways, viz., forced reverse polarization and natural recovery. The PID detection tools used in this work are well known module characterization techniques such as EL imaging, Infrared imaging, and light and dark current-voltage measurements. These characterisation tools are used in combination to detect defects such as optical losses, cracks, breakage, electric circuit degradation and PID. Under normal testing PID was detected and in some cases, modules were able to recover, while for advanced stage PID regeneration or PID reversal was difficult. This thesis focuses on PID detection at infancy using three approaches; EL imaging at current corresponding to 10% of Isc. Light and dark current – voltage measurements (L-IV & D-IV) and open circuit voltage (Voc) ratios at low irradiance. The early detection procedures are essential in reversing the degradation caused by PID which is reversible. The time taken to reverse the PID degradation will depend on the extent of the degradation. If detected early, it will take a short period of time to completely reverse lost power. Infrared thermography is a non-contact characteristic tool that can be deployed in large scale plants using drones to detect the presence of PID in PV plants. Module performance and device parameters extracted from the L-IV curves on a module before and after PID stress, such as Pmpp, Voc, Isc Fill Factor (FF), shunt resistance (Rsh) and series resistance (Rs) and ideality (n) are sensitive to PID shunting. Voc and Rsh drop significantly with the onset of PID, while Rs increases. The decrease in Voc and Rsh is due to heavy shunting on the module resulting in increased carrier recombination, while the increase in Rs is due to increased shunting paths leading to decreased photocurrent. When substantial degradation on a module occurs Pmpp, FF and n will drop and at very advanced stage of PID degradation Isc may drop excessively.
- Full Text:
- Date Issued: 2020
Patterns and Pattern Languages for Mobile Augmented Reality
- Authors: Munro, Donald
- Date: 2020
- Subjects: Augmented reality , Educational technology
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/48510 , vital:40883
- Description: Mixed Reality is a relatively new field in computer science which uses technology as a medium to provide modified or enhanced views of reality or to virtually generate a new reality. Augmented Reality is a branch of Mixed Reality which blends the real-world as viewed through a computer interface with virtual objects generated by a computer. The 21st century commodification of mobile devices with multi-core Central Processing Units, Graphics Processing Units, high definition displays and multiple sensors controlled by capable Operating Systems such as Android and iOS means that Mobile Augmented Reality applications have become increasingly feasible. Mobile Augmented Reality is a multi-disciplinary field requiring a synthesis of many technologies such as computer graphics, computer vision, machine learning and mobile device programming while also requiring theoretical knowledge of diverse fields such as Linear Algebra, Projective and Differential Geometry, Probability and Optimisation. This multi-disciplinary nature has led to a fragmentation of knowledge into various specialisations, making it difficult to integrate different solution components into a coherent architecture. Software design patterns provide a solution space of tried and tested best practices for a specified problem within a given context. The solution space is non-prescriptive and is described in terms of relationships between roles that can be assigned to software components. Architectural patterns are used to specify high level designs of complete systems, as opposed to domain or tactical level patterns that address specific lower level problem areas. Pattern Languages comprise multiple software patterns combining in multiple possible sequences to form a language with the individual patterns forming the language vocabulary while the valid sequences through the patterns define the grammar. Pattern Languages provide flexible generalised solutions within a particular domain that can be customised to solve problems of differing characteristics and levels of iii complexity within the domain. The specification of one or more Pattern Languages tailored to the Mobile Augmented Reality domain can therefore provide a generalised guide for the design and architecture of Mobile Augmented Reality applications from an architectural level down to the ”nuts-and-bolts” implementation level. While there is a large body of research into the technical specialisations pertaining to Mobile Augmented Reality, there is a dearth of up-to-date literature covering Mobile Augmented Reality design. This thesis fills this vacuum by: 1. Providing architectural patterns that provide the spine on which the design of Mobile Augmented Reality artefacts can be based; 2. Documenting existing patterns within the context of Mobile Augmented Reality; 3. Identifying new patterns specific to Mobile Augmented Reality; and 4. Combining the patterns into Pattern Languages for Detection & Tracking, Rendering & Interaction and Data Access for Mobile Augmented Reality. The resulting Pattern Languages support design at multiple levels of complexity from an object-oriented framework down to specific one-off Augmented Reality applications. The practical contribution of this thesis is the specification of architectural patterns and Pattern Language that provide a unified design approach for both the overall architecture and the detailed design of Mobile Augmented Reality artefacts. The theoretical contribution is a design theory for Mobile Augmented Reality gleaned from the extraction of patterns and creation of a pattern language or languages.
- Full Text:
- Date Issued: 2020
- Authors: Munro, Donald
- Date: 2020
- Subjects: Augmented reality , Educational technology
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/48510 , vital:40883
- Description: Mixed Reality is a relatively new field in computer science which uses technology as a medium to provide modified or enhanced views of reality or to virtually generate a new reality. Augmented Reality is a branch of Mixed Reality which blends the real-world as viewed through a computer interface with virtual objects generated by a computer. The 21st century commodification of mobile devices with multi-core Central Processing Units, Graphics Processing Units, high definition displays and multiple sensors controlled by capable Operating Systems such as Android and iOS means that Mobile Augmented Reality applications have become increasingly feasible. Mobile Augmented Reality is a multi-disciplinary field requiring a synthesis of many technologies such as computer graphics, computer vision, machine learning and mobile device programming while also requiring theoretical knowledge of diverse fields such as Linear Algebra, Projective and Differential Geometry, Probability and Optimisation. This multi-disciplinary nature has led to a fragmentation of knowledge into various specialisations, making it difficult to integrate different solution components into a coherent architecture. Software design patterns provide a solution space of tried and tested best practices for a specified problem within a given context. The solution space is non-prescriptive and is described in terms of relationships between roles that can be assigned to software components. Architectural patterns are used to specify high level designs of complete systems, as opposed to domain or tactical level patterns that address specific lower level problem areas. Pattern Languages comprise multiple software patterns combining in multiple possible sequences to form a language with the individual patterns forming the language vocabulary while the valid sequences through the patterns define the grammar. Pattern Languages provide flexible generalised solutions within a particular domain that can be customised to solve problems of differing characteristics and levels of iii complexity within the domain. The specification of one or more Pattern Languages tailored to the Mobile Augmented Reality domain can therefore provide a generalised guide for the design and architecture of Mobile Augmented Reality applications from an architectural level down to the ”nuts-and-bolts” implementation level. While there is a large body of research into the technical specialisations pertaining to Mobile Augmented Reality, there is a dearth of up-to-date literature covering Mobile Augmented Reality design. This thesis fills this vacuum by: 1. Providing architectural patterns that provide the spine on which the design of Mobile Augmented Reality artefacts can be based; 2. Documenting existing patterns within the context of Mobile Augmented Reality; 3. Identifying new patterns specific to Mobile Augmented Reality; and 4. Combining the patterns into Pattern Languages for Detection & Tracking, Rendering & Interaction and Data Access for Mobile Augmented Reality. The resulting Pattern Languages support design at multiple levels of complexity from an object-oriented framework down to specific one-off Augmented Reality applications. The practical contribution of this thesis is the specification of architectural patterns and Pattern Language that provide a unified design approach for both the overall architecture and the detailed design of Mobile Augmented Reality artefacts. The theoretical contribution is a design theory for Mobile Augmented Reality gleaned from the extraction of patterns and creation of a pattern language or languages.
- Full Text:
- Date Issued: 2020
Phytoplankton of the Southern Agulhas Large Marine Ecosystem (sACLME)
- Authors: Sonnekus, Martinus Jakobus
- Date: 2020
- Subjects: Botanical chemistry
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/47646 , vital:40261
- Description: The southern Agulhas Large Marine Ecosystem (sACLME) consists of the greater Agulhas Current system, including the areas influenced by the Agulhas Current, the Agulhas Shelf, Agulhas Bank, Agulhas Retroflection as well as the Agulhas Return Current. Four biogeographical regions were identified within the sACLME and the composition of the phytoplankton communities and the associated physico-chemical variables in each were investigated. Water temperature and nitrate concentration were found to be the main drivers of the sACLME phytoplankton community. A total of 215 phytoplankton taxa were identified that were separated into two groups: the Agulhas Current group and the Southwest Indian Ocean Ridge group. The structure of the phytoplankton community was consistently homogeneous throughout the sACLME. Nutrient concentrations of the water varied, but nitrate was the most abundant source of nitrogen, especially in regions that were subjected to the upwelling of cold nutrient-rich water. The phytoplankton communities were significantly influenced by the availability and stoichiometry of the macronutrients nitrogen; phosphorus and silicon. The low mean N:P ratio of 5.5 is an indication that the waters of the sACLME are generally nitrogen limited, as is typical of oceanic systems. There is a change in phytoplankton cell size and functional groups within in the various water masses that are specific to certain biogeographical regions within the Agulhas Current system. It is clear that frontal systems may form barriers between phytoplankton communities.
- Full Text:
- Date Issued: 2020
- Authors: Sonnekus, Martinus Jakobus
- Date: 2020
- Subjects: Botanical chemistry
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/47646 , vital:40261
- Description: The southern Agulhas Large Marine Ecosystem (sACLME) consists of the greater Agulhas Current system, including the areas influenced by the Agulhas Current, the Agulhas Shelf, Agulhas Bank, Agulhas Retroflection as well as the Agulhas Return Current. Four biogeographical regions were identified within the sACLME and the composition of the phytoplankton communities and the associated physico-chemical variables in each were investigated. Water temperature and nitrate concentration were found to be the main drivers of the sACLME phytoplankton community. A total of 215 phytoplankton taxa were identified that were separated into two groups: the Agulhas Current group and the Southwest Indian Ocean Ridge group. The structure of the phytoplankton community was consistently homogeneous throughout the sACLME. Nutrient concentrations of the water varied, but nitrate was the most abundant source of nitrogen, especially in regions that were subjected to the upwelling of cold nutrient-rich water. The phytoplankton communities were significantly influenced by the availability and stoichiometry of the macronutrients nitrogen; phosphorus and silicon. The low mean N:P ratio of 5.5 is an indication that the waters of the sACLME are generally nitrogen limited, as is typical of oceanic systems. There is a change in phytoplankton cell size and functional groups within in the various water masses that are specific to certain biogeographical regions within the Agulhas Current system. It is clear that frontal systems may form barriers between phytoplankton communities.
- Full Text:
- Date Issued: 2020
Synthesis, characterisation and antitumour activities of lanthanide complexes with hydrazones and carboxylic acid ligands
- Authors: Madanhire, Tatenda
- Date: 2020
- Subjects: Organic acids
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/48456 , vital:40878
- Description: The tridentate hydrazone ligands, (E)-N'-(2-hydroxybenzylidene)benzohydrazide (H2phen) and (E)-N'-(2-hydroxybenzylidene)nicotinohydrazide (H2Nic), were synthesised and complexed to Ln(III) acetates. The centrosymmetric, acetato-bridged dinuclear coordination compounds with the formulae, [La2(Hphen)2(OAc)4(H2O)2]·DMF·H2O (1), [Ln2(HNic)2(OAc)4(H2O)2]·DMF·H2O (Ln = La (2) and Nd (3)) and [Ln2(HNic)2(OAc)4(H2O)2]·DMF (Ln = Er (4) and Yb (5)) were isolated and characterised by elemental analyses, IR spectroscopy, UV-Vis spectroscopy, X-ray diffraction studies and SHAPE 2.1. The nine-coordinate complexes 1–3 crystallise in the triclinic space group P-1, with the metal centres having the distorted spherical capped square antiprism geometry (C4v), while the eight-coordinate Er(III) and Yb(III) complexes (monoclinic system, space group P21/c) display the geometry of distorted triangular dodecahedron (D2d). Geometry optimisation of the monoanionic forms of the hydrazone ligands (Hphen– and HNic– ) were performed using Density Functional Theory (DFT) with Becke’s three parameter hybrid method and correlation functional of Lee, Yang and Parr (B3LYP) with augcc-pVTZ basis set. Natural population analysis (NPA) and molecular electrostatic potential (MEP) maps indicated that the most preferred sites for electrophilic attack in the anionic ligands are the phenolate and carbonyl oxygens, and the azomethine nitrogens. The evaluation of the cytotoxic activity of the compounds on breast cancer (MCF-7), the endometrial carcinoma (HEC-1A) and the human monocytic (THP-1) cell lines using the 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl tetrazolium bromide (MTT) assay revealed that the hydrazone ligands and complexes 1–4 are partially cytotoxic against MCF-7 cells, while the Schiff bases and complexes 3–5 significantly inhibit cell growth in HEC-1A cells. The complexation reactions of Ce(III), Nd(III), Gd(III) and Er(III) with the chelating/ bridging monoanionic ligand N-(2,6-dimethylphenyl)oxamate (Hpma– ) in basic media were performed in view of the potential applications of oxamate derivatives as cytotoxic agents. The coordination compounds were characterised by different Abstract T. Madanhire Nelson Mandela University xxvi physico-chemical techniques: elemental analysis, conductivity measurements, IR, 1 H NMR and UV-Vis-NIR spectroscopy. The anionic Hpma– was obtained through conversion of the proligand ethyl (2,6-dimethylphenylcarbamoyl)formate (Hdmp). The reactions afforded lanthanide(III)–oxamate coordination polymers of formulae: {[Ln(Hpma)3(MeOH)(H2O)]∙2MeOH}n (Ln = Ce (1) and Nd (2)), {[Gd2(Hpma)6(MeOH)4]∙6MeOH}n (3), {[Er2(Hpma)6(MeOH)(H2O)3]∙2MeOH}n (4) and [Ln2Na2(Hpma)8(EtOH)(H2O)6]n (Ln = Nd (5) and Gd (6)). The polymeric complexes feature Ln-Hpma moieties bridged by μ2-η 1 :η 1 :η 1 Hpma– , giving onedimensional zig-zag chains of the –Ln–O–C–O–Ln– type. Atomic charge analysis and the MEP map of the Hpma– moiety done using the DFT/B3LYP method were found to be consistent with the chelating and bridging modes of the anionic ligand through all the oxygen atoms. The evaluation of the cytotoxic activities of the metal salts, the proligand and the novel lanthanide complexes on MCF-7, HEC-1A and THP-1 cell lines revealed that only the rare-earth metal salts [Ce(NO3)3∙6H2O] and [Nd(NO3)3∙6H2O] showed modest cytotoxicity against MCF-7 and HEC-1A cells, respectively.
- Full Text:
- Date Issued: 2020
- Authors: Madanhire, Tatenda
- Date: 2020
- Subjects: Organic acids
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/48456 , vital:40878
- Description: The tridentate hydrazone ligands, (E)-N'-(2-hydroxybenzylidene)benzohydrazide (H2phen) and (E)-N'-(2-hydroxybenzylidene)nicotinohydrazide (H2Nic), were synthesised and complexed to Ln(III) acetates. The centrosymmetric, acetato-bridged dinuclear coordination compounds with the formulae, [La2(Hphen)2(OAc)4(H2O)2]·DMF·H2O (1), [Ln2(HNic)2(OAc)4(H2O)2]·DMF·H2O (Ln = La (2) and Nd (3)) and [Ln2(HNic)2(OAc)4(H2O)2]·DMF (Ln = Er (4) and Yb (5)) were isolated and characterised by elemental analyses, IR spectroscopy, UV-Vis spectroscopy, X-ray diffraction studies and SHAPE 2.1. The nine-coordinate complexes 1–3 crystallise in the triclinic space group P-1, with the metal centres having the distorted spherical capped square antiprism geometry (C4v), while the eight-coordinate Er(III) and Yb(III) complexes (monoclinic system, space group P21/c) display the geometry of distorted triangular dodecahedron (D2d). Geometry optimisation of the monoanionic forms of the hydrazone ligands (Hphen– and HNic– ) were performed using Density Functional Theory (DFT) with Becke’s three parameter hybrid method and correlation functional of Lee, Yang and Parr (B3LYP) with augcc-pVTZ basis set. Natural population analysis (NPA) and molecular electrostatic potential (MEP) maps indicated that the most preferred sites for electrophilic attack in the anionic ligands are the phenolate and carbonyl oxygens, and the azomethine nitrogens. The evaluation of the cytotoxic activity of the compounds on breast cancer (MCF-7), the endometrial carcinoma (HEC-1A) and the human monocytic (THP-1) cell lines using the 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl tetrazolium bromide (MTT) assay revealed that the hydrazone ligands and complexes 1–4 are partially cytotoxic against MCF-7 cells, while the Schiff bases and complexes 3–5 significantly inhibit cell growth in HEC-1A cells. The complexation reactions of Ce(III), Nd(III), Gd(III) and Er(III) with the chelating/ bridging monoanionic ligand N-(2,6-dimethylphenyl)oxamate (Hpma– ) in basic media were performed in view of the potential applications of oxamate derivatives as cytotoxic agents. The coordination compounds were characterised by different Abstract T. Madanhire Nelson Mandela University xxvi physico-chemical techniques: elemental analysis, conductivity measurements, IR, 1 H NMR and UV-Vis-NIR spectroscopy. The anionic Hpma– was obtained through conversion of the proligand ethyl (2,6-dimethylphenylcarbamoyl)formate (Hdmp). The reactions afforded lanthanide(III)–oxamate coordination polymers of formulae: {[Ln(Hpma)3(MeOH)(H2O)]∙2MeOH}n (Ln = Ce (1) and Nd (2)), {[Gd2(Hpma)6(MeOH)4]∙6MeOH}n (3), {[Er2(Hpma)6(MeOH)(H2O)3]∙2MeOH}n (4) and [Ln2Na2(Hpma)8(EtOH)(H2O)6]n (Ln = Nd (5) and Gd (6)). The polymeric complexes feature Ln-Hpma moieties bridged by μ2-η 1 :η 1 :η 1 Hpma– , giving onedimensional zig-zag chains of the –Ln–O–C–O–Ln– type. Atomic charge analysis and the MEP map of the Hpma– moiety done using the DFT/B3LYP method were found to be consistent with the chelating and bridging modes of the anionic ligand through all the oxygen atoms. The evaluation of the cytotoxic activities of the metal salts, the proligand and the novel lanthanide complexes on MCF-7, HEC-1A and THP-1 cell lines revealed that only the rare-earth metal salts [Ce(NO3)3∙6H2O] and [Nd(NO3)3∙6H2O] showed modest cytotoxicity against MCF-7 and HEC-1A cells, respectively.
- Full Text:
- Date Issued: 2020
Synthesis, crystal structures and molecular modelling of rare earth complexes with bis(2-pyridylmethyl)amine and its derivatives : a quantum chemical investigation of ligand conformational space, complex intramolecular rearrangement and stability
- Authors: Matthews, Cameron
- Date: 2020
- Subjects: Rare earths , Complex compounds , Ligands
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/46229 , vital:39517
- Description: Limited research has been performed on the coordination behaviour of hybrid aliphatic and heterocyclic polyamines with trivalent rare earth elements. The rare earth coordination properties of several Nalkylated derivatives of the tridentate ligand bis(2-pyridylmethyl)amine (DPA, HL1) backbone - involving the rare earth elements Y, La-Nd, Sm, Eu and Tb-Lu - have been investigated in this study However, the structural and energetic characteristics of DPA coordination with rare earth elements (REE) have not been studied thus far. Potential applications of DPA-based rare earth complexes are primarily dependent on their electronic and magnetic characteristics, which are affected weakly by the coordination environment, where potential applications may include use as Lanthanide Shift Reagents (LSR), Luminescence probes and small-molecule magnets (SMM). A systematic conformational search of DPA was carried out in this study in order to identify the global minimum conformer and for comparison of the free and coordinated geometries, using the M06(D3) functional belonging to the Density Functional Theory (DFT) family of model chemistries. An understanding of the aforementioned would play an important role in analysis of bulk characterization and the prediction of the reactivity of DPA. Final geometries and electronic energies were obtained from the ‘domain based local pair natural orbital’ (DLPNO)-Møller-Plesset and -coupled cluster theoretical methods, as follows: DLPNO-CCSD(T0)/aug-cc-pVQZ//DLPNO-MP2/aug-cc-pVTZ. Fifteen Single-crystal X-ray diffractometer (SC-XRD) crystal structures of mononuclear rare earth chloride coordination complexes with DPA (RE = La-Nd, Sm, Eu, Tb-Lu and Y) were obtained and geometrically analysed in this study. Three isostructural series of constitutional isomers were identified, consisting of one series of nine-coordinate molecule (M1) and two series of eight-coordinate ion pairs (M2 and M3). This conformational diversity is most likely due the flexible nature of the DPA backbone, as well as the additional stability gained from reduced coordination spheres as a function of decreasing rare earth ionic radii across the lanthanide series (Lanthanide contraction). A Quantum Theory of Atoms-in-Molecules (QTAIM) topological analysis was performed in order to identify and characterise potential hydrogen bonding interactions in H-optimised crystal structures. The crystal structures of five dinuclear (RE = Tb-Tm) and two tetranuclear (RE = Yb and Lu) rare earth chloride complexes with DPA have also been structurally analysed. Furthermore, one mononuclear (RE = Dy), two dinuclear complexes (RE = Dy and Lu) with EtDPA, and one mononuclear complex with the DPA-derivative HL4 (RE = Dy) were structurally characterised. A DFT study of the theoretical interconversion of one real- and two hypothetical- mononuclear lanthanum containing isostructural series (cf. aforementioned crystal structures) was undertaken in order to gain a deeper understanding of the processes involved, in terms of the participating minimum energy paths (MEPs), intermediates and transition states – as determined via the Nudged-Elastic Band (NEB) procedure. This hypothesis is supported by the well-known conformational lability of rare earth complexes, due to the weak/minor covalency of their coordination bonds. An attempt was made to determine the respective energies of one real- and two hypothetical diamagnetic or ‘closed-shell’ constitutional isomers containing the rare earth ions La3+(M1), Y3+(M2) and Lu3+ (M3). It was assumed that the most stable isomers have a greater probability of being observed as the asymmetric unit of the complex crystal structure – assuming weak contributions of lattice or intermolecular interactions towards the geometry of the asymmetric unit.
- Full Text:
- Date Issued: 2020
- Authors: Matthews, Cameron
- Date: 2020
- Subjects: Rare earths , Complex compounds , Ligands
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/46229 , vital:39517
- Description: Limited research has been performed on the coordination behaviour of hybrid aliphatic and heterocyclic polyamines with trivalent rare earth elements. The rare earth coordination properties of several Nalkylated derivatives of the tridentate ligand bis(2-pyridylmethyl)amine (DPA, HL1) backbone - involving the rare earth elements Y, La-Nd, Sm, Eu and Tb-Lu - have been investigated in this study However, the structural and energetic characteristics of DPA coordination with rare earth elements (REE) have not been studied thus far. Potential applications of DPA-based rare earth complexes are primarily dependent on their electronic and magnetic characteristics, which are affected weakly by the coordination environment, where potential applications may include use as Lanthanide Shift Reagents (LSR), Luminescence probes and small-molecule magnets (SMM). A systematic conformational search of DPA was carried out in this study in order to identify the global minimum conformer and for comparison of the free and coordinated geometries, using the M06(D3) functional belonging to the Density Functional Theory (DFT) family of model chemistries. An understanding of the aforementioned would play an important role in analysis of bulk characterization and the prediction of the reactivity of DPA. Final geometries and electronic energies were obtained from the ‘domain based local pair natural orbital’ (DLPNO)-Møller-Plesset and -coupled cluster theoretical methods, as follows: DLPNO-CCSD(T0)/aug-cc-pVQZ//DLPNO-MP2/aug-cc-pVTZ. Fifteen Single-crystal X-ray diffractometer (SC-XRD) crystal structures of mononuclear rare earth chloride coordination complexes with DPA (RE = La-Nd, Sm, Eu, Tb-Lu and Y) were obtained and geometrically analysed in this study. Three isostructural series of constitutional isomers were identified, consisting of one series of nine-coordinate molecule (M1) and two series of eight-coordinate ion pairs (M2 and M3). This conformational diversity is most likely due the flexible nature of the DPA backbone, as well as the additional stability gained from reduced coordination spheres as a function of decreasing rare earth ionic radii across the lanthanide series (Lanthanide contraction). A Quantum Theory of Atoms-in-Molecules (QTAIM) topological analysis was performed in order to identify and characterise potential hydrogen bonding interactions in H-optimised crystal structures. The crystal structures of five dinuclear (RE = Tb-Tm) and two tetranuclear (RE = Yb and Lu) rare earth chloride complexes with DPA have also been structurally analysed. Furthermore, one mononuclear (RE = Dy), two dinuclear complexes (RE = Dy and Lu) with EtDPA, and one mononuclear complex with the DPA-derivative HL4 (RE = Dy) were structurally characterised. A DFT study of the theoretical interconversion of one real- and two hypothetical- mononuclear lanthanum containing isostructural series (cf. aforementioned crystal structures) was undertaken in order to gain a deeper understanding of the processes involved, in terms of the participating minimum energy paths (MEPs), intermediates and transition states – as determined via the Nudged-Elastic Band (NEB) procedure. This hypothesis is supported by the well-known conformational lability of rare earth complexes, due to the weak/minor covalency of their coordination bonds. An attempt was made to determine the respective energies of one real- and two hypothetical diamagnetic or ‘closed-shell’ constitutional isomers containing the rare earth ions La3+(M1), Y3+(M2) and Lu3+ (M3). It was assumed that the most stable isomers have a greater probability of being observed as the asymmetric unit of the complex crystal structure – assuming weak contributions of lattice or intermolecular interactions towards the geometry of the asymmetric unit.
- Full Text:
- Date Issued: 2020
The characterization and application of novel xanthenyland thioxanthenyl-derived compounds as hosts in the presence of various applicable guest mixtures
- Authors: Jooste, Daniel Victor
- Date: 2020
- Subjects: Organic compounds -- Synthesis
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/49104 , vital:41601
- Description: In this work, the host potential of four novel, structurally-related compounds, trans- N,N’-bis(9-phenyl-9-xanthenyl)cyclohexane-1,2-diamine (1,2-DAX), trans-N,N’-bis(9- phenyl-9-thioxanthenyl)cyclohexane-1,2-diamine (1,2-DAT), trans-N,N’-bis(9-phenyl- 9-xanthenyl)cyclohexane-1,4-diamine (1,4-DAX), and trans-N,N’-bis(9-phenyl-9- thioxanthenyl)cyclohexane-1,4-diamine (1,4-DAT), were investigated for their possible employment in the separation of isomers and other related compounds using host– guest chemistry. These hosts were synthesized following a Grignard reaction with phenylmagnesium bromide on either xanthone or thioxanthone. The resultant alcohol was treated with perchloric acid, before finally being linked with either trans-1,2- or trans-1,4-diaminocyclohexane to afford the four title host compounds. Initially, the feasibility of these hosts for separating isomers and related compounds was investigated by recrystallizing each one independently from various guest compounds including the methylpyridines and pyridine, xylenes and ethylbenzene, heterocyclic six-membered ring compounds, aniline, N-methylaniline and N,Ndimethylaniline, and also the alkylbenzenes toluene, cumene and ethylbenzene. 1HNMR spectroscopy was used to ascertain if inclusion of the guest species had occurred in this manner and, if so, the host:guest ratio of the complex. Guest–guest competitions were subsequently conducted in order to establish the selectivity of the hosts when presented with a mixture of guests, and if the host would be able to discriminate between them. Gas chromatography-mass spectrometry (GC-MS) was the analytical method of choice here if 1H-NMR spectroscopy was not suitable owing to guest–guest signal overlap. Here, the host was recrystallized from binary, ternary and quaternary mixed guests, where each was present in equimolar amounts. Additionally, binary competition experiments were conducted in mixed guest solvents where the molar ratios of the guests were varied, and host selectivity for these guests evaluated visually by means of selectivity profiles. Single crystal X-ray diffraction (SC-XRD) and thermal analysis were performed on any complexes that afforded crystals of suitable quality in order to relate inherent host–guest interactions and thermal stability to the observed host selectivity from the competition experiments. The more prevalent interactions that were evident in these inclusion complexes between host and guest species were, more usually, and interactions, as well as other short contacts. Hydrogen bonding interactions were observed in only a few of the complexes. Guest accommodation type was also investigated, and these species resided in either discrete cavities or channels within the host crystal, depending on the guest. For the most part, traces obtained from thermal analyses were highly convoluted and difficult to interpret. As a result, guest release onset temperatures could not be determined for all of the inclusion complexes. In some cases, however, this temperature was successfully measured and correlated directly with the observed selectivity order of the host suggested by the competition experiments. Both enhanced and contrasting results were obtained for the four host compounds. In all cases, 1,2-DAX and 1,2-DAT successfully formed complexes with each of the guest species from each series. Interestingly, the 1,4-derived hosts, however, were more selective in that 1,4-DAX did not complex with only one of the guest solvents, while 1,4-DAT did not enclathrate as many as ten of these solvents. The competition investigations showed that, in most of these experiments, the host compounds displayed selectivity for one of the guests present in the mixture, and in some cases, this selectivity was pronounced, alluding to the feasibility of separating related guests from one another through host–guest chemistry principles. Computational calculations were, additionally, conducted on each of the host molecules in order to gain a better understanding of their geometries, and to compare these with the apohost crystal structures. Significant geometry differences were noted between the calculated and crystal structures.
- Full Text:
- Date Issued: 2020
- Authors: Jooste, Daniel Victor
- Date: 2020
- Subjects: Organic compounds -- Synthesis
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/49104 , vital:41601
- Description: In this work, the host potential of four novel, structurally-related compounds, trans- N,N’-bis(9-phenyl-9-xanthenyl)cyclohexane-1,2-diamine (1,2-DAX), trans-N,N’-bis(9- phenyl-9-thioxanthenyl)cyclohexane-1,2-diamine (1,2-DAT), trans-N,N’-bis(9-phenyl- 9-xanthenyl)cyclohexane-1,4-diamine (1,4-DAX), and trans-N,N’-bis(9-phenyl-9- thioxanthenyl)cyclohexane-1,4-diamine (1,4-DAT), were investigated for their possible employment in the separation of isomers and other related compounds using host– guest chemistry. These hosts were synthesized following a Grignard reaction with phenylmagnesium bromide on either xanthone or thioxanthone. The resultant alcohol was treated with perchloric acid, before finally being linked with either trans-1,2- or trans-1,4-diaminocyclohexane to afford the four title host compounds. Initially, the feasibility of these hosts for separating isomers and related compounds was investigated by recrystallizing each one independently from various guest compounds including the methylpyridines and pyridine, xylenes and ethylbenzene, heterocyclic six-membered ring compounds, aniline, N-methylaniline and N,Ndimethylaniline, and also the alkylbenzenes toluene, cumene and ethylbenzene. 1HNMR spectroscopy was used to ascertain if inclusion of the guest species had occurred in this manner and, if so, the host:guest ratio of the complex. Guest–guest competitions were subsequently conducted in order to establish the selectivity of the hosts when presented with a mixture of guests, and if the host would be able to discriminate between them. Gas chromatography-mass spectrometry (GC-MS) was the analytical method of choice here if 1H-NMR spectroscopy was not suitable owing to guest–guest signal overlap. Here, the host was recrystallized from binary, ternary and quaternary mixed guests, where each was present in equimolar amounts. Additionally, binary competition experiments were conducted in mixed guest solvents where the molar ratios of the guests were varied, and host selectivity for these guests evaluated visually by means of selectivity profiles. Single crystal X-ray diffraction (SC-XRD) and thermal analysis were performed on any complexes that afforded crystals of suitable quality in order to relate inherent host–guest interactions and thermal stability to the observed host selectivity from the competition experiments. The more prevalent interactions that were evident in these inclusion complexes between host and guest species were, more usually, and interactions, as well as other short contacts. Hydrogen bonding interactions were observed in only a few of the complexes. Guest accommodation type was also investigated, and these species resided in either discrete cavities or channels within the host crystal, depending on the guest. For the most part, traces obtained from thermal analyses were highly convoluted and difficult to interpret. As a result, guest release onset temperatures could not be determined for all of the inclusion complexes. In some cases, however, this temperature was successfully measured and correlated directly with the observed selectivity order of the host suggested by the competition experiments. Both enhanced and contrasting results were obtained for the four host compounds. In all cases, 1,2-DAX and 1,2-DAT successfully formed complexes with each of the guest species from each series. Interestingly, the 1,4-derived hosts, however, were more selective in that 1,4-DAX did not complex with only one of the guest solvents, while 1,4-DAT did not enclathrate as many as ten of these solvents. The competition investigations showed that, in most of these experiments, the host compounds displayed selectivity for one of the guests present in the mixture, and in some cases, this selectivity was pronounced, alluding to the feasibility of separating related guests from one another through host–guest chemistry principles. Computational calculations were, additionally, conducted on each of the host molecules in order to gain a better understanding of their geometries, and to compare these with the apohost crystal structures. Significant geometry differences were noted between the calculated and crystal structures.
- Full Text:
- Date Issued: 2020
The effect of gamma radiation on the physical and chemical properties of wool
- Authors: Rapakgadi, Jim
- Date: 2020
- Subjects: Semiconductors -- Effect of radiation on
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/47668 , vital:40263
- Description: Pathogens (e.g. viruses, spores and bacteria) remain one of the biggest threats to trading and the economy of natural fibre products, such as wool, mohair and cotton. One of the means to kill or eliminate pathogens is by radiation sterilisation, involving electromagnetic radiation (e.g. gamma radiation and UV light). The research, reported in this thesis, was aimed at establishing the effect of gamma irradiation dosage on the various quality related physical and chemical characteristics of four different types of merino wool when treated in situ in scoured and greasy state in mini-bale form. Specific properties studied include wool colour (yellowness and brightness), solubility (alkali solubility and urea bisulphite solubility), fibre diameter and tensile properties (staple strength and bundle tenacity). Standard analytical and diagnostic techniques were used to assess changes, if any, in the above mentioned properties due to the radiation treatment. Nominal radiation dosages, ranging from 25 kGy to 75 kGy (actual measured from 0 kGy to 92.2 kGy) were used, 50 kGy normally being regarded as sufficient to destroy the pathogens in question. It was found that nominial gamma radiation dosages of up to 92.2 kGy did not significantly affect fibre diameter or staple length, indicating that no irreversible swelling of the wool fibres occurred as a result of gamma irradiation. Alkali and urea bisulphite solubility as well as yellowness, on the other hand, showed a systematic increase with an increase in irradiation dose, indicating that changes occurred at molecular level. Fibre tensile properties (staple strength and bundle tenacity) were also affected but to a much lesser extent than solubility and colour. The various merino wool types studied showed similar reactions to the radiation treatment, irrespective of the wool quality and type and also whether they were treated in greasy or scoured form.
- Full Text:
- Date Issued: 2020
- Authors: Rapakgadi, Jim
- Date: 2020
- Subjects: Semiconductors -- Effect of radiation on
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/47668 , vital:40263
- Description: Pathogens (e.g. viruses, spores and bacteria) remain one of the biggest threats to trading and the economy of natural fibre products, such as wool, mohair and cotton. One of the means to kill or eliminate pathogens is by radiation sterilisation, involving electromagnetic radiation (e.g. gamma radiation and UV light). The research, reported in this thesis, was aimed at establishing the effect of gamma irradiation dosage on the various quality related physical and chemical characteristics of four different types of merino wool when treated in situ in scoured and greasy state in mini-bale form. Specific properties studied include wool colour (yellowness and brightness), solubility (alkali solubility and urea bisulphite solubility), fibre diameter and tensile properties (staple strength and bundle tenacity). Standard analytical and diagnostic techniques were used to assess changes, if any, in the above mentioned properties due to the radiation treatment. Nominal radiation dosages, ranging from 25 kGy to 75 kGy (actual measured from 0 kGy to 92.2 kGy) were used, 50 kGy normally being regarded as sufficient to destroy the pathogens in question. It was found that nominial gamma radiation dosages of up to 92.2 kGy did not significantly affect fibre diameter or staple length, indicating that no irreversible swelling of the wool fibres occurred as a result of gamma irradiation. Alkali and urea bisulphite solubility as well as yellowness, on the other hand, showed a systematic increase with an increase in irradiation dose, indicating that changes occurred at molecular level. Fibre tensile properties (staple strength and bundle tenacity) were also affected but to a much lesser extent than solubility and colour. The various merino wool types studied showed similar reactions to the radiation treatment, irrespective of the wool quality and type and also whether they were treated in greasy or scoured form.
- Full Text:
- Date Issued: 2020
The impact of the slum upgrading approach on the livelihoods of scheme participants in Uganda : using the sustainable livelihoods framework
- Authors: Karungi, Doreen
- Date: 2020
- Subjects: Slums -- Public welfare -- Uganda
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/48412 , vital:40874
- Description: International and local agencies recommend implementing slum upgrading projects in slum areas for slum improvement. However, with no model or guiding principles provided to slum upgrading project implementers on how to upgrade a slum, slum clearance and redevelopment are implemented rather than slum upgrading consequently projects fail to benefit the target group, slum dwellers. Data collection, took place in five areas in Uganda that implemented slum upgrading. In depth interviews were held with project beneficiaries, local council officials and association executives, and government officials at the Ministry of Lands, Housing and Urban Development. Field observation, and document analysis was carried out. Data analysis involved the use of the sustainable livelihoods framework. This study demonstrates that the slum upgrading approach adopted in Uganda exhibited low sustainability and undercut the sustainability of livelihoods of scheme participants. Slum dwellers faced vulnerabilities from loss of land, characterized by rights based actions resulting in minimal to no compensation or eviction and loss of existing livelihoods. Slum upgrading projects introduced new livelihood opportunities that were too few for slum dwellers, were not sustainable and ended during or with project implementation, leaving beneficiaries without employment or income opportunities. Increased vulnerability was observed as slum dwellers that did not benefit from the project were displaced to worse conditions while slum dwellers that benefitted, project beneficiaries remained poor and in debt, with project loans to repay. Although project beneficiaries were provided with access to assets, they were unable to draw on these to build their livelihoods. Some beneficiaries sold plots and project houses allocated to them and moved to adjacent slums, where they lived in worse conditions. At the time of the current research, good quality houses in the slum upgrading areas did not belong to the original beneficiaries old dilapidated structures were occupied by project beneficiaries. This study shows that income earning activities carried out at the time of the study were similar to those carried out prior to the introduction of slum upgrading in project areas and were destroyed during project implementation or banned by implementers. These activities were however carried out in incomplete houses, vacant plots, temporary structures in project areas and in adjacent swamps and forest. Generally, after completion of slum upgrading projects, project areas, adjacent swamps and forest were characterized by haphazard developments including: rentals, informal businesses, farming, brewing and distilling local alcohol used by beneficiaries to derive livelihoods, in other words, slums were recreated, a problem slum upgrading project implementers set out to solve - except in one study area where the project was implemented differently as a result of persist rights based actions.
- Full Text:
- Date Issued: 2020
- Authors: Karungi, Doreen
- Date: 2020
- Subjects: Slums -- Public welfare -- Uganda
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/48412 , vital:40874
- Description: International and local agencies recommend implementing slum upgrading projects in slum areas for slum improvement. However, with no model or guiding principles provided to slum upgrading project implementers on how to upgrade a slum, slum clearance and redevelopment are implemented rather than slum upgrading consequently projects fail to benefit the target group, slum dwellers. Data collection, took place in five areas in Uganda that implemented slum upgrading. In depth interviews were held with project beneficiaries, local council officials and association executives, and government officials at the Ministry of Lands, Housing and Urban Development. Field observation, and document analysis was carried out. Data analysis involved the use of the sustainable livelihoods framework. This study demonstrates that the slum upgrading approach adopted in Uganda exhibited low sustainability and undercut the sustainability of livelihoods of scheme participants. Slum dwellers faced vulnerabilities from loss of land, characterized by rights based actions resulting in minimal to no compensation or eviction and loss of existing livelihoods. Slum upgrading projects introduced new livelihood opportunities that were too few for slum dwellers, were not sustainable and ended during or with project implementation, leaving beneficiaries without employment or income opportunities. Increased vulnerability was observed as slum dwellers that did not benefit from the project were displaced to worse conditions while slum dwellers that benefitted, project beneficiaries remained poor and in debt, with project loans to repay. Although project beneficiaries were provided with access to assets, they were unable to draw on these to build their livelihoods. Some beneficiaries sold plots and project houses allocated to them and moved to adjacent slums, where they lived in worse conditions. At the time of the current research, good quality houses in the slum upgrading areas did not belong to the original beneficiaries old dilapidated structures were occupied by project beneficiaries. This study shows that income earning activities carried out at the time of the study were similar to those carried out prior to the introduction of slum upgrading in project areas and were destroyed during project implementation or banned by implementers. These activities were however carried out in incomplete houses, vacant plots, temporary structures in project areas and in adjacent swamps and forest. Generally, after completion of slum upgrading projects, project areas, adjacent swamps and forest were characterized by haphazard developments including: rentals, informal businesses, farming, brewing and distilling local alcohol used by beneficiaries to derive livelihoods, in other words, slums were recreated, a problem slum upgrading project implementers set out to solve - except in one study area where the project was implemented differently as a result of persist rights based actions.
- Full Text:
- Date Issued: 2020
The taxonomic status of dolphins Tursiops spp. and Delphinus in South African waters
- Authors: Ngqulana, Sibusisiwe Gloria
- Date: 2020
- Subjects: Dolphins -- Classification -- South Africa
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/48566 , vital:40891
- Description: There is a need for identifying important sites, taxonomic units or populations and the usual criterion used is species identification. The species is the most important taxonomic unit to define in taxonomy, as it is the primary unit of concern in biodiversity, conservation and other biological issues. Thus, taxonomy plays an important role in conservation biology. Additionally, taxonomy can contribute to the development of analytical systems that will provide information on priorities for reaching defensible targets for established conservation. This poses a problem for cetaceans as basic information is frequently deficient and therefore not suitable for conservation planning purposes. This leads to difficulties in identifying and prioritizing areas for the monitoring, management and for the protection of species or populations. Morphological and geographic data used to be the primary means used by taxonomists to define species and subspecies, and due to this, most extant species have been classified by morphological characteristics. However, in the recent years, molecular analyses have been introduced and subsequently stable isotopes were also included to assist in interpreting results. Thereafter, it became a common approach to use more than one technique or a combination of independent datasets to resolve taxonomic uncertainties. This has resulted to a proposed term known as integrative taxonomy. Integrative taxonomy has recently been used in most studies that deal with taxonomy or population structure. The use of multiple lines of evidence, which is sometimes referred to as integrative taxonomy, raises the level of confidence in clearly delineating species. Cetaceans are the most taxonomically diverse group of aquatic mammals, with 93 living species accepted. Previously, their taxonomy was based on morphological characters. However, cetaceans have recently been subjected to intense phylogenetic investigations using both morphological and molecular data. These problems are not only prevalent in larger whales, but the small toothed whales also have these challenges, particularly family Delphinidae. This family is known to be the most diverse among the toothed whales. This is because this family diverged recently, and the divergence occurred relatively fast. This study aimed to clarify the taxonomic status of Tursiops spp. and Delphinus in the family Delphinidae, inhabiting the South African waters, using a combination of morphological and molecular methods. For skull morphology, geometric morphometrics were used, while for molecular methods, the D-loop region in the mitochondrial DNA was used to detect any differences within each genus. For the genus Tursiops, this study showed the consistency of results in all analyses used to investigate the presence of two species in the genus Tursiops in South African waters. This study only analysed sequences from the mitochondrial D-loop region and has re-confirmed the presence of the two species in South Africa. Skull and external morphology both show T. truncatus to be bigger and more robust, and to have a shorter rostrum compared with T. aduncus. These external observations are helpful when identifying the species in the field. Providing evidence supporting the presence of two Tursiops species in South Africa does not only contribute to the global context of the taxonomy of this genus, but also in the local context. This study also gives insight on the distribution of the two species in the region. For the genus Delphinus, molecular analysis did not reveal any clustering in South African Delphinus specimens, which suggested a single species, D. delphis, and that this species is morphologically highly variable. Both skull and external morphology did not reflect the molecular results, supporting suggestions that morphology in this genus does not reflect genetic differences. It has been shown that morphological variation reflects feeding specialization rather than speciation. This may also reflect local adaptations. Both skull and external morphology in this study showed no significant differences. This study did not investigate the phylogeographic details of the species in South Africa and further molecular work including a larger sample size from the west coast of South Africa is needed. By attempting to unravel the taxonomy of these dolphins in South African waters, the findings of this study will be very useful in the development of effective conservation and management strategies for these dolphins.
- Full Text:
- Date Issued: 2020
- Authors: Ngqulana, Sibusisiwe Gloria
- Date: 2020
- Subjects: Dolphins -- Classification -- South Africa
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/48566 , vital:40891
- Description: There is a need for identifying important sites, taxonomic units or populations and the usual criterion used is species identification. The species is the most important taxonomic unit to define in taxonomy, as it is the primary unit of concern in biodiversity, conservation and other biological issues. Thus, taxonomy plays an important role in conservation biology. Additionally, taxonomy can contribute to the development of analytical systems that will provide information on priorities for reaching defensible targets for established conservation. This poses a problem for cetaceans as basic information is frequently deficient and therefore not suitable for conservation planning purposes. This leads to difficulties in identifying and prioritizing areas for the monitoring, management and for the protection of species or populations. Morphological and geographic data used to be the primary means used by taxonomists to define species and subspecies, and due to this, most extant species have been classified by morphological characteristics. However, in the recent years, molecular analyses have been introduced and subsequently stable isotopes were also included to assist in interpreting results. Thereafter, it became a common approach to use more than one technique or a combination of independent datasets to resolve taxonomic uncertainties. This has resulted to a proposed term known as integrative taxonomy. Integrative taxonomy has recently been used in most studies that deal with taxonomy or population structure. The use of multiple lines of evidence, which is sometimes referred to as integrative taxonomy, raises the level of confidence in clearly delineating species. Cetaceans are the most taxonomically diverse group of aquatic mammals, with 93 living species accepted. Previously, their taxonomy was based on morphological characters. However, cetaceans have recently been subjected to intense phylogenetic investigations using both morphological and molecular data. These problems are not only prevalent in larger whales, but the small toothed whales also have these challenges, particularly family Delphinidae. This family is known to be the most diverse among the toothed whales. This is because this family diverged recently, and the divergence occurred relatively fast. This study aimed to clarify the taxonomic status of Tursiops spp. and Delphinus in the family Delphinidae, inhabiting the South African waters, using a combination of morphological and molecular methods. For skull morphology, geometric morphometrics were used, while for molecular methods, the D-loop region in the mitochondrial DNA was used to detect any differences within each genus. For the genus Tursiops, this study showed the consistency of results in all analyses used to investigate the presence of two species in the genus Tursiops in South African waters. This study only analysed sequences from the mitochondrial D-loop region and has re-confirmed the presence of the two species in South Africa. Skull and external morphology both show T. truncatus to be bigger and more robust, and to have a shorter rostrum compared with T. aduncus. These external observations are helpful when identifying the species in the field. Providing evidence supporting the presence of two Tursiops species in South Africa does not only contribute to the global context of the taxonomy of this genus, but also in the local context. This study also gives insight on the distribution of the two species in the region. For the genus Delphinus, molecular analysis did not reveal any clustering in South African Delphinus specimens, which suggested a single species, D. delphis, and that this species is morphologically highly variable. Both skull and external morphology did not reflect the molecular results, supporting suggestions that morphology in this genus does not reflect genetic differences. It has been shown that morphological variation reflects feeding specialization rather than speciation. This may also reflect local adaptations. Both skull and external morphology in this study showed no significant differences. This study did not investigate the phylogeographic details of the species in South Africa and further molecular work including a larger sample size from the west coast of South Africa is needed. By attempting to unravel the taxonomy of these dolphins in South African waters, the findings of this study will be very useful in the development of effective conservation and management strategies for these dolphins.
- Full Text:
- Date Issued: 2020
A comparative study of artificial neural networks and physics models as simulators in evolutionary robotics
- Pretorius, Christiaan Johannes
- Authors: Pretorius, Christiaan Johannes
- Date: 2019
- Subjects: Neural networks (Computer science)
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/30789 , vital:31131
- Description: The Evolutionary Robotics (ER) process is a technique that applies evolutionary optimization algorithms to the task of automatically developing, or evolving, robotic control programs. These control programs, or simply controllers, are evolved in order to allow a robot to perform a required task. During the ER process, use is often made of robotic simulators to evaluate the performance of candidate controllers that are produced in the course of the controller evolution process. Such simulators accelerate and otherwise simplify the controller evolution process, as opposed to the more arduous process of evaluating controllers in the real world without use of simulation. To date, the vast majority of simulators that have been applied in ER are physics- based models which are constructed by taking into account the underlying physics governing the operation of the robotic system in question. An alternative approach to simulator implementation in ER is the usage of Artificial Neural Networks (ANNs) as simulators in the ER process. Such simulators are referred to as Simulator Neural Networks (SNNs). Previous studies have indicated that SNNs can successfully be used as an alter- native to physics-based simulators in the ER process on various robotic platforms. At the commencement of the current study it was not, however, known how this relatively new method of simulation would compare to traditional physics-based simulation approaches in ER. The study presented in this thesis thus endeavoured to quantitatively compare SNNs and physics-based models as simulators in the ER process. In order to con- duct this comparative study, both SNNs and physics simulators were constructed for the modelling of three different robotic platforms: a differentially-steered robot, a wheeled inverted pendulum robot and a hexapod robot. Each of these two types of simulation was then used in simulation-based evolution processes to evolve con- trollers for each robotic platform. During these controller evolution processes, the SNNs and physics models were compared in terms of their accuracy in making pre- dictions of robotic behaviour, their computational efficiency in arriving at these predictions, the human effort required to construct each simulator and, most im- portantly, the real-world performance of controllers evolved by making use of each simulator. The results obtained in this study illustrated experimentally that SNNs were, in the majority of cases, able to make more accurate predictions than the physics- based models and these SNNs were arguably simpler to construct than the physics simulators. Additionally, SNNs were also shown to be a computationally efficient alternative to physics-based simulators in ER and, again in the majority of cases, these SNNs were able to produce controllers which outperformed those evolved in the physics-based simulators, when these controllers were uploaded to the real-world robots. The results of this thesis thus suggest that SNNs are a viable alternative to more commonly-used physics simulators in ER and further investigation of the potential of this simulation technique appears warranted.
- Full Text:
- Date Issued: 2019
- Authors: Pretorius, Christiaan Johannes
- Date: 2019
- Subjects: Neural networks (Computer science)
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/30789 , vital:31131
- Description: The Evolutionary Robotics (ER) process is a technique that applies evolutionary optimization algorithms to the task of automatically developing, or evolving, robotic control programs. These control programs, or simply controllers, are evolved in order to allow a robot to perform a required task. During the ER process, use is often made of robotic simulators to evaluate the performance of candidate controllers that are produced in the course of the controller evolution process. Such simulators accelerate and otherwise simplify the controller evolution process, as opposed to the more arduous process of evaluating controllers in the real world without use of simulation. To date, the vast majority of simulators that have been applied in ER are physics- based models which are constructed by taking into account the underlying physics governing the operation of the robotic system in question. An alternative approach to simulator implementation in ER is the usage of Artificial Neural Networks (ANNs) as simulators in the ER process. Such simulators are referred to as Simulator Neural Networks (SNNs). Previous studies have indicated that SNNs can successfully be used as an alter- native to physics-based simulators in the ER process on various robotic platforms. At the commencement of the current study it was not, however, known how this relatively new method of simulation would compare to traditional physics-based simulation approaches in ER. The study presented in this thesis thus endeavoured to quantitatively compare SNNs and physics-based models as simulators in the ER process. In order to con- duct this comparative study, both SNNs and physics simulators were constructed for the modelling of three different robotic platforms: a differentially-steered robot, a wheeled inverted pendulum robot and a hexapod robot. Each of these two types of simulation was then used in simulation-based evolution processes to evolve con- trollers for each robotic platform. During these controller evolution processes, the SNNs and physics models were compared in terms of their accuracy in making pre- dictions of robotic behaviour, their computational efficiency in arriving at these predictions, the human effort required to construct each simulator and, most im- portantly, the real-world performance of controllers evolved by making use of each simulator. The results obtained in this study illustrated experimentally that SNNs were, in the majority of cases, able to make more accurate predictions than the physics- based models and these SNNs were arguably simpler to construct than the physics simulators. Additionally, SNNs were also shown to be a computationally efficient alternative to physics-based simulators in ER and, again in the majority of cases, these SNNs were able to produce controllers which outperformed those evolved in the physics-based simulators, when these controllers were uploaded to the real-world robots. The results of this thesis thus suggest that SNNs are a viable alternative to more commonly-used physics simulators in ER and further investigation of the potential of this simulation technique appears warranted.
- Full Text:
- Date Issued: 2019