A Functional Data Analysis approach to understand patterns imbedded within various data types
- Authors: Mangisa, Siphumlile
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Data mining
- Language: English
- Type: Doctoral theses , text
- Identifier: http://hdl.handle.net/10948/52091 , vital:43426
- Description: The thesis investigates the use of the novel Functional Data Analysis (FDA) methods in tackling various data types. Strong motivation is provided for the use of interesting opportunities offered by FDA to analyse not only economic data, but generally, data from any domain. The use of these methods is illustrated using three unique self-contained case-studies from econometrics. , Thesis (PhD) -- Faculty of Science, School of Computer Science, Mathematics, Physics and Statistics, 2021
- Full Text: false
- Date Issued: 2021-04
- Authors: Mangisa, Siphumlile
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Data mining
- Language: English
- Type: Doctoral theses , text
- Identifier: http://hdl.handle.net/10948/52091 , vital:43426
- Description: The thesis investigates the use of the novel Functional Data Analysis (FDA) methods in tackling various data types. Strong motivation is provided for the use of interesting opportunities offered by FDA to analyse not only economic data, but generally, data from any domain. The use of these methods is illustrated using three unique self-contained case-studies from econometrics. , Thesis (PhD) -- Faculty of Science, School of Computer Science, Mathematics, Physics and Statistics, 2021
- Full Text: false
- Date Issued: 2021-04
A study of the airflow on the windward slope of a transverse dune in the Alexandria coastal dunefield
- Authors: Burkinshaw, Jennifer Ruth
- Date: 2021-04
- Subjects: Port Elizabeth (South Africa) , Eastern Cape (South Africa) , South Africa
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52734 , vital:43883
- Description: Our understanding of the evolution of dune morphology has been hampered by a lack of empirical observations of airflow behaviour over dune forms. Sand dunes intrude into the atmospheric boundary layer and convergence of streamlines results in an acceleration of airflow up the windward slopes of dunes. This study examines the airflow structure and corresponding bedform development on the windward slope of a 7 m high transverse dune on the edge of the Alexandria coastal dunefield, Algoa Bay, South Africa. The Alexandria dunefield is subjected to a trimodal wind regime, consisting of the dominant south-westerly which blows all year round, summer easterlies and winter northwesterlies. The morphology of the study dune, Dune13, is controlled by the easterlies and north-westerlies, and reverses seasonally with respect to these two winds. Seven section lines 30 m apart and normal to the dune crest were surveyed regularly over the period of a year to monitor the reversal process. Three detailed topographic surveys were also done during this period. Airflow behaviour was monitored during the year. Wind speed profiles on the windward slope of the dune were measured using 4 to 5 vertical arrays of anemometers positioned from the base of the dune to the crest on a 1 selected section line. Usually 4 to 5 anemometers were deployed in each vertical array, from a height of 6 to 10 cm above the surface, up to a height of 150 cm above the surface. Initially 8 microanemometers were available; ultimately 28 anemometers were run simultaneously. An independent weather station at an elevation of 6 m recorded the unaccelerated flow. Local gradient measurements and erosion and deposition rates were recorded along selected section lines. Strong summer easterly winds (14 m/sec at 1.4 m above the dune crest) were measured on a dune slope in the process of being transformed from a slipface to a stoss slope. The following winter, light north-westerly winds (typically B m/sec at 1.6 m above the dune crest) were measured on the new windward slope already reversed by the prevailing winter wind. Airflow data confirm the compression of airflow against the windward slope resulting in a non-logarithmic wind speed profile. Compression results in an increased shear velocity within 30 cm of the dune surface, and the dune slope is eroded. Higher up in the wind speed profile, shear velocity decreases to 0.1 m/sec. It is not known at what height the wind speed profile recovers from the intrusion of the dune into the boundary layer. High values of shear velocity (1.6 m/sec) above the rounded crestal area of the dune record the recovery of the wind speed profile from flow divergence, which is a response to the rapid reduction of dune gradient and is accompanied by deposition of sand in this region. 2 The erosion pin data act as a simple and sensitive test for changes in gradient, reflecting the dune's response to changes in the airflow regime. The shape of the dune plays a major role in determining the extent of the compression and the distribution of shear velocity up the slope. Increased shear velocity is experienced on that part of the slope which is nonaerodynamic with respect to the prevailing wind. Under unidirectional conditions, feedback between flow and form results ultimately in a slope with a curvature such that shear velocity increases systematically upslope. The survey data and erosion pin data record the reversal process as the dune achieves a new steady state during each wind season. The existence of a non-logarithmic wind speed profile makes it difficult to know what relevant measure of shear velocity is to be used in sand transport equations. Future work should include wind speed measurements within 10 cm of the surface. An ideal study modelling aeolian bedform development would utilise wind tunnel measurements, combined with field measurements such as obtained in this study, for comparison with numerical modelling. The study needs to be extended to 3-dimensional airflow measurements. , Thesis (MSc) -- Faculty of Science, School of Environmental Sciences, 2021
- Full Text:
- Date Issued: 2021-04
- Authors: Burkinshaw, Jennifer Ruth
- Date: 2021-04
- Subjects: Port Elizabeth (South Africa) , Eastern Cape (South Africa) , South Africa
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52734 , vital:43883
- Description: Our understanding of the evolution of dune morphology has been hampered by a lack of empirical observations of airflow behaviour over dune forms. Sand dunes intrude into the atmospheric boundary layer and convergence of streamlines results in an acceleration of airflow up the windward slopes of dunes. This study examines the airflow structure and corresponding bedform development on the windward slope of a 7 m high transverse dune on the edge of the Alexandria coastal dunefield, Algoa Bay, South Africa. The Alexandria dunefield is subjected to a trimodal wind regime, consisting of the dominant south-westerly which blows all year round, summer easterlies and winter northwesterlies. The morphology of the study dune, Dune13, is controlled by the easterlies and north-westerlies, and reverses seasonally with respect to these two winds. Seven section lines 30 m apart and normal to the dune crest were surveyed regularly over the period of a year to monitor the reversal process. Three detailed topographic surveys were also done during this period. Airflow behaviour was monitored during the year. Wind speed profiles on the windward slope of the dune were measured using 4 to 5 vertical arrays of anemometers positioned from the base of the dune to the crest on a 1 selected section line. Usually 4 to 5 anemometers were deployed in each vertical array, from a height of 6 to 10 cm above the surface, up to a height of 150 cm above the surface. Initially 8 microanemometers were available; ultimately 28 anemometers were run simultaneously. An independent weather station at an elevation of 6 m recorded the unaccelerated flow. Local gradient measurements and erosion and deposition rates were recorded along selected section lines. Strong summer easterly winds (14 m/sec at 1.4 m above the dune crest) were measured on a dune slope in the process of being transformed from a slipface to a stoss slope. The following winter, light north-westerly winds (typically B m/sec at 1.6 m above the dune crest) were measured on the new windward slope already reversed by the prevailing winter wind. Airflow data confirm the compression of airflow against the windward slope resulting in a non-logarithmic wind speed profile. Compression results in an increased shear velocity within 30 cm of the dune surface, and the dune slope is eroded. Higher up in the wind speed profile, shear velocity decreases to 0.1 m/sec. It is not known at what height the wind speed profile recovers from the intrusion of the dune into the boundary layer. High values of shear velocity (1.6 m/sec) above the rounded crestal area of the dune record the recovery of the wind speed profile from flow divergence, which is a response to the rapid reduction of dune gradient and is accompanied by deposition of sand in this region. 2 The erosion pin data act as a simple and sensitive test for changes in gradient, reflecting the dune's response to changes in the airflow regime. The shape of the dune plays a major role in determining the extent of the compression and the distribution of shear velocity up the slope. Increased shear velocity is experienced on that part of the slope which is nonaerodynamic with respect to the prevailing wind. Under unidirectional conditions, feedback between flow and form results ultimately in a slope with a curvature such that shear velocity increases systematically upslope. The survey data and erosion pin data record the reversal process as the dune achieves a new steady state during each wind season. The existence of a non-logarithmic wind speed profile makes it difficult to know what relevant measure of shear velocity is to be used in sand transport equations. Future work should include wind speed measurements within 10 cm of the surface. An ideal study modelling aeolian bedform development would utilise wind tunnel measurements, combined with field measurements such as obtained in this study, for comparison with numerical modelling. The study needs to be extended to 3-dimensional airflow measurements. , Thesis (MSc) -- Faculty of Science, School of Environmental Sciences, 2021
- Full Text:
- Date Issued: 2021-04
A theoretical study of alkoxyl radical and radical cation mediated cyclisation reactions in unsaturated alkanols
- Authors: Lee, Peter Mattison Clarke
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Density functionals
- Language: English
- Type: Doctoral theses , text
- Identifier: http://hdl.handle.net/10948/52130 , vital:43430
- Description: A large part of this computational study focussed on selectivity patterns for 4-penten- 1-oxyl radical 1,5- and 1,6-cyclisations and for those of analogous 5-hexenyl, 4- penten-1-aminyl, 4-penten-1-iminyl and 3-phenylpropan-1-oxyl radicals. The study furthermore included cyclisation reactions of the radical cations of 4-penten-1-ol and 3-phenylpropan-1-ol. Calculations were performed for all applicable species involved in the two cyclisation modes studied, i.e. 1,5- and 1,6-cyclisation. These species included radical and radical cation precursors, as well as relevant radical transition state structures and radical adducts. All cyclisation systems investigated used the Density Functional Theory (DFT) functional B3LYP with a combination of eight basis sets. Each basis set used was a progressively higher level of theory than B3LYP/6-31G* which was used as a starting point. Cyclisations for unsubstituted species, e.g. the 4-penten-1-oxyl radical 1,5- and 1,6-cyclisations, were also simulated using the ωB97X-D and ωB97X-V functionals which account for dispersive effects. The same set of eight basis sets were used with these two functionals. , Thesis (PhD) -- Faculty of Science, School of Computer Science, Mathematics, Physics and Statistics, 2021
- Full Text: false
- Date Issued: 2021-04
- Authors: Lee, Peter Mattison Clarke
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Density functionals
- Language: English
- Type: Doctoral theses , text
- Identifier: http://hdl.handle.net/10948/52130 , vital:43430
- Description: A large part of this computational study focussed on selectivity patterns for 4-penten- 1-oxyl radical 1,5- and 1,6-cyclisations and for those of analogous 5-hexenyl, 4- penten-1-aminyl, 4-penten-1-iminyl and 3-phenylpropan-1-oxyl radicals. The study furthermore included cyclisation reactions of the radical cations of 4-penten-1-ol and 3-phenylpropan-1-ol. Calculations were performed for all applicable species involved in the two cyclisation modes studied, i.e. 1,5- and 1,6-cyclisation. These species included radical and radical cation precursors, as well as relevant radical transition state structures and radical adducts. All cyclisation systems investigated used the Density Functional Theory (DFT) functional B3LYP with a combination of eight basis sets. Each basis set used was a progressively higher level of theory than B3LYP/6-31G* which was used as a starting point. Cyclisations for unsubstituted species, e.g. the 4-penten-1-oxyl radical 1,5- and 1,6-cyclisations, were also simulated using the ωB97X-D and ωB97X-V functionals which account for dispersive effects. The same set of eight basis sets were used with these two functionals. , Thesis (PhD) -- Faculty of Science, School of Computer Science, Mathematics, Physics and Statistics, 2021
- Full Text: false
- Date Issued: 2021-04
An assessment of different methods for measuring mammal diversity in two Southern African arid ecosystems.
- Authors: Burt, Chloe
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Biodiversity -- South Africa
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52338 , vital:43591
- Description: A current critical environmental problem is the loss of biodiversity. To allow for adaptive management and effective conservation interventions, reliable data needs to be readily available and effectively collected. To gather reliable data, the most appropriate method must be used. The aim of this thesis was to compare the effectiveness and efficiency of different mammal diversity sampling techniques, including discussing specificities and complementarity of each. I also assessed the limitation of methods in terms of the influence of species traits on detectability. I used standard mammal field methods such as block transects, camera traps, track plates, spotlight surveys, mist nets, acoustic surveys, and live traps for a wide range of mammal taxa, and also included the use of local knowledge in the comparison. I assessed the cost-efficiency of each method by determining the cost per species. The efficiency of each method was assessed by observing when the species accumulation curve reached an asymptote. The effectiveness was determined by the species richness and the taxonomic groups the method detected. My results indicated that block transects were the most efficient and effective method to assess species richness. However, a combination of methods allowed for a more comprehensive survey and my results indicate that block transects and live traps were the best combination. It is important to note, not only the effectiveness and efficiency of methods be assessed, but also their limitations. I therefore also assessed the influence of species traits on the detectability of five methods. The species traits that I tested were body size, diet, activity, and if they lived in groups. On both study areas, body size was the main trait that influenced detectability with most field methods. Diet and activity had varying influence on detectability depending on the method. Group living did not significantly influence detectability. Based on my results, I provide guidelines for choosing the most appropriate survey method for targeted studies on mammals in arid ecosystems as well as guidelines for choosing the most appropriate methods for a terrestrial mammal inventory. Survey methods selection and guidelines provided here can allow for effective monitoring and therefore effective and adaptive management and having reliable and repeatable outcomes from surveys. , Thesis (MSc) -- Faculty of Science, School of Natural Resource Management, 2021
- Full Text: false
- Date Issued: 2021-04
- Authors: Burt, Chloe
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Biodiversity -- South Africa
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52338 , vital:43591
- Description: A current critical environmental problem is the loss of biodiversity. To allow for adaptive management and effective conservation interventions, reliable data needs to be readily available and effectively collected. To gather reliable data, the most appropriate method must be used. The aim of this thesis was to compare the effectiveness and efficiency of different mammal diversity sampling techniques, including discussing specificities and complementarity of each. I also assessed the limitation of methods in terms of the influence of species traits on detectability. I used standard mammal field methods such as block transects, camera traps, track plates, spotlight surveys, mist nets, acoustic surveys, and live traps for a wide range of mammal taxa, and also included the use of local knowledge in the comparison. I assessed the cost-efficiency of each method by determining the cost per species. The efficiency of each method was assessed by observing when the species accumulation curve reached an asymptote. The effectiveness was determined by the species richness and the taxonomic groups the method detected. My results indicated that block transects were the most efficient and effective method to assess species richness. However, a combination of methods allowed for a more comprehensive survey and my results indicate that block transects and live traps were the best combination. It is important to note, not only the effectiveness and efficiency of methods be assessed, but also their limitations. I therefore also assessed the influence of species traits on the detectability of five methods. The species traits that I tested were body size, diet, activity, and if they lived in groups. On both study areas, body size was the main trait that influenced detectability with most field methods. Diet and activity had varying influence on detectability depending on the method. Group living did not significantly influence detectability. Based on my results, I provide guidelines for choosing the most appropriate survey method for targeted studies on mammals in arid ecosystems as well as guidelines for choosing the most appropriate methods for a terrestrial mammal inventory. Survey methods selection and guidelines provided here can allow for effective monitoring and therefore effective and adaptive management and having reliable and repeatable outcomes from surveys. , Thesis (MSc) -- Faculty of Science, School of Natural Resource Management, 2021
- Full Text: false
- Date Issued: 2021-04
An investigation into the effects of nano-carbons on the negative electrode morphology of lead acid batteries under high rate partial state of capacity cycling
- Authors: Bolo, Lukanyo
- Date: 2021-04
- Subjects: Port Elizabeth (South Africa) , Eastern Cape (South Africa) , South Africa
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/53324 , vital:45132
- Description: New technological developments in the field of Hybrid Electric Vehicles (HEV’s), vehicle stop-start applications and electrical power utility devices have brought about new demands on the use of energy storage devices such as the Pb-acid battery. These applications require a new understanding and research into the material chemistry of the battery where it has to functionin a Partial State of Capacity Cycling (PSoCC) or High Rate Partial State of Capacity Cycling (HRPSoCC)duty. The Pb-acid battery is well known for its use in the automotive industry where it had worked for decades as a typical Starter, Lighting and Ignition (SLI) power source. However, the conventional Pb-acid battery is not suitable for the new types of vehicle demands and applications. When the Pb-lead acid battery is subjected to PSoCC conditions, very quickly excessive amount of non-conducting PbSO4crystals are formed on the negative plates of the battery that then drastically reduce its useful life in application. Over the years, researchers have shown that the small addition of certain carbon types to the negative plate active material (NAM) during battery manufacturing can reduce the sulfation effect thereby increasing its use the modern vehicle applications. This study investigated industrial manufactured cells that contained up to 0.5 % by weight two types of carbons in the NAM. These were compared to standard built cells in terms of their standard achievable capacities at various rates, their Cold Cranking Ability (CCA), their Dynamic Charge Acceptance (DCA) and a simplified PSoCC test. The carbons were classified as an activated carbon and a carbon nano-fiber and were included as a dry weight additive during the standard paste manufacturing process of the negative plate. Cells were built with a 9-plate configuration using standard expanded grid positive plates with a nominal rated capacity of 36 Ah. The cells were formed and subjected to a number of electrochemical tests, of which selected negative plates were then studied using Scanning Electron Microscopy (SEM) and Electrochemical in-situ Atomic Force Microscopy (EC-AFM). The microscopic study was then used to try and relate the observed material characteristics of the NAM with the carbons to the electrochemical results observed when testing a built cell. , Thesis (MA) --Faculty of Science, School of Biomolecular and Chemical Sciences, 2021
- Full Text:
- Date Issued: 2021-04
- Authors: Bolo, Lukanyo
- Date: 2021-04
- Subjects: Port Elizabeth (South Africa) , Eastern Cape (South Africa) , South Africa
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/53324 , vital:45132
- Description: New technological developments in the field of Hybrid Electric Vehicles (HEV’s), vehicle stop-start applications and electrical power utility devices have brought about new demands on the use of energy storage devices such as the Pb-acid battery. These applications require a new understanding and research into the material chemistry of the battery where it has to functionin a Partial State of Capacity Cycling (PSoCC) or High Rate Partial State of Capacity Cycling (HRPSoCC)duty. The Pb-acid battery is well known for its use in the automotive industry where it had worked for decades as a typical Starter, Lighting and Ignition (SLI) power source. However, the conventional Pb-acid battery is not suitable for the new types of vehicle demands and applications. When the Pb-lead acid battery is subjected to PSoCC conditions, very quickly excessive amount of non-conducting PbSO4crystals are formed on the negative plates of the battery that then drastically reduce its useful life in application. Over the years, researchers have shown that the small addition of certain carbon types to the negative plate active material (NAM) during battery manufacturing can reduce the sulfation effect thereby increasing its use the modern vehicle applications. This study investigated industrial manufactured cells that contained up to 0.5 % by weight two types of carbons in the NAM. These were compared to standard built cells in terms of their standard achievable capacities at various rates, their Cold Cranking Ability (CCA), their Dynamic Charge Acceptance (DCA) and a simplified PSoCC test. The carbons were classified as an activated carbon and a carbon nano-fiber and were included as a dry weight additive during the standard paste manufacturing process of the negative plate. Cells were built with a 9-plate configuration using standard expanded grid positive plates with a nominal rated capacity of 36 Ah. The cells were formed and subjected to a number of electrochemical tests, of which selected negative plates were then studied using Scanning Electron Microscopy (SEM) and Electrochemical in-situ Atomic Force Microscopy (EC-AFM). The microscopic study was then used to try and relate the observed material characteristics of the NAM with the carbons to the electrochemical results observed when testing a built cell. , Thesis (MA) --Faculty of Science, School of Biomolecular and Chemical Sciences, 2021
- Full Text:
- Date Issued: 2021-04
Antifungal activity of silver nanoparticles on Cryptococcus species
- Authors: Roman, Nasstasja
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Cryptococcus
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52089 , vital:43445
- Description: Cryptococcus neoformans is responsible for cryptococcal infections, which have become a great concern particularly in immunocompromised patients. This pulmonary disease causes damage to lung tissue and is presumed to be responsible for approximately 650,000 deaths annually. There is a major need for novel antimicrobial agents to avoid the emergence of resistant strains due to extended use of known antifungal drugs. Silver has become an important antimicrobial agent used in medical devices, cosmetics and even house-hold or clothing products. It is highly toxic and has a reduced risk of resistance. Silver nanoparticles have been broadly explored as an antimicrobial agent due to its high reactivity, large surface area and small size. They have been reported to display antimicrobial activity towards bacteria, viruses and fungi. C. neoformans is a pathogenic microorganism that possesses a number of virulence factors which increases the infectivity of the microorganism in immunocompromised individuals. Some of these include the presence of a polysaccharide capsule, melanin, urease and phospholipase production. Owing to limited availability of antifungal drugs and the possibility of drug resistance, this study focused on an alternative antimicrobial agent, namely silver nanoparticles, and the effects that silver nanoparticles had on fungal β-glucan, chitin and urease, phospholipase and melanin virulence factors. Results from our study show that silver nanoparticles inhibited the growth of C. neoformans at concentrations between 3.125 ƒM and 25 ƒM. Both 26.4 nm and 28.4 nm nanoparticles displayed growth inhibition, but were unable to kill the organism at concentrations of 200 ƒM and above. The silver nanoparticles caused a reduction in βglucan content. There was also an accumulation of intracellular ROS, and phospholipase activity, was reduced. In contrast, the organism displayed an increase in chitin content, melanin and urease activity upon exposure to the nanoparticles. This study shows that chemically synthesized silver nanoparticles are an effective antifungal alternative that greatly reduces or inhibits C. neoformans growth at very low concentrations (ƒM). For future studies it may be feasible to investigate the use of these nanoparticles in combination with known antifungal agents, to prevent the emergence of resistance and inhibit possible defense mechanisms displayed by the organism. , Thesis (MSc) -- Faculty of Science, School of Biomolecular and Chemical Sciences, 2021
- Full Text: false
- Date Issued: 2021-04
- Authors: Roman, Nasstasja
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Cryptococcus
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52089 , vital:43445
- Description: Cryptococcus neoformans is responsible for cryptococcal infections, which have become a great concern particularly in immunocompromised patients. This pulmonary disease causes damage to lung tissue and is presumed to be responsible for approximately 650,000 deaths annually. There is a major need for novel antimicrobial agents to avoid the emergence of resistant strains due to extended use of known antifungal drugs. Silver has become an important antimicrobial agent used in medical devices, cosmetics and even house-hold or clothing products. It is highly toxic and has a reduced risk of resistance. Silver nanoparticles have been broadly explored as an antimicrobial agent due to its high reactivity, large surface area and small size. They have been reported to display antimicrobial activity towards bacteria, viruses and fungi. C. neoformans is a pathogenic microorganism that possesses a number of virulence factors which increases the infectivity of the microorganism in immunocompromised individuals. Some of these include the presence of a polysaccharide capsule, melanin, urease and phospholipase production. Owing to limited availability of antifungal drugs and the possibility of drug resistance, this study focused on an alternative antimicrobial agent, namely silver nanoparticles, and the effects that silver nanoparticles had on fungal β-glucan, chitin and urease, phospholipase and melanin virulence factors. Results from our study show that silver nanoparticles inhibited the growth of C. neoformans at concentrations between 3.125 ƒM and 25 ƒM. Both 26.4 nm and 28.4 nm nanoparticles displayed growth inhibition, but were unable to kill the organism at concentrations of 200 ƒM and above. The silver nanoparticles caused a reduction in βglucan content. There was also an accumulation of intracellular ROS, and phospholipase activity, was reduced. In contrast, the organism displayed an increase in chitin content, melanin and urease activity upon exposure to the nanoparticles. This study shows that chemically synthesized silver nanoparticles are an effective antifungal alternative that greatly reduces or inhibits C. neoformans growth at very low concentrations (ƒM). For future studies it may be feasible to investigate the use of these nanoparticles in combination with known antifungal agents, to prevent the emergence of resistance and inhibit possible defense mechanisms displayed by the organism. , Thesis (MSc) -- Faculty of Science, School of Biomolecular and Chemical Sciences, 2021
- Full Text: false
- Date Issued: 2021-04
Antiproliferative activity of novel rhenium complexes and medicinal plant extracts
- Oosthuizen, Kenneth Thomas, Venables, Luanne
- Authors: Oosthuizen, Kenneth Thomas , Venables, Luanne
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Medicinal plants -- South Africa
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52028 , vital:43441
- Description: Cancer is a complex, multifactorial disease that affects millions of individuals every year. The adverse side effects and escalating costs of current therapies coupled with the increased incidence of resistance to these therapies make it imperative that we explore novel treatments for the disease. This study investigated two avenues for novel drug design namely, novel synthetic compounds and medicinal plant extracts. The benzimidazole ring system has shown potential as a scaffold for designing novel anticancer agents. Conjugation of the metal rhenium to novel variants of this ring system open up the possibility of designing novel drugs that serve both a diagnostic and a therapeutic function. This study investigated the in vitro anticancer potential of ten such complexes against selected breast and cervical cancer cell lines. The selectivity of the complexes for cancer cells over normal cells was also investigated while the mechanism of action of effective complexes was determined by exploring cell cycle arrest, biochemical markers of apoptosis and mitochondrial membrane disruption. All ten complexes were screened against MCF7 breast and HeLa cervical cancer cell lines with four showing antiproliferative activity against both cancer cell lines and one showing cell line specific toxicity against MCF7 breast cancer cells. When looking at the SAR of the compounds it was noted that activity was higher in compounds which contain two potentially bidentate benzothiazole ligands, while compounds with potentially tridentate ligands show good activity, but only if the benzothiazole moiety is not involved in coordination. The complexes showed greater antiproliferative activity against the MCF7 breast cancer cells with IC50 values ranging from 3.2 to 7.0 µM versus a range of 7.6 to 24.4 µM being obtained on the HeLa cervical cancer cells. Antiproliferative complexes were tested against confluent and log phase Vero cells at their respective IC50 values to determine their effect on “normal” cells. Confluent Vero cells showed less cell death than those in log phase indicating that the complexes show preference for proliferating cells. The mechanism of action of the complexes was studied on both cancer cell lines via cell cycle analysis and apoptosis assays investigating phosphatidylserine translocation, caspase -3 and -8 activation and mitochondrial membrane potential with most complexes inducing cell cycle arrest followed by cell death via both the extrinsic and intrinsic pathways of apoptosis. Since 1940, 49% of all the available anticancer drugs approved for cancer treatment were natural products or directly derived from natural products. Plants are an excellent source of secondary metabolites, many of which are unique chemical compounds that cannot be synthesized in a laboratory. Ethnobotanical surveys conducted in conjunction with Traditional Health Practitioners of the Mkuranga and Same districts in Tanzania identified 25 plants that are used for the treatment of cancer. Four of these plants (A. mossambicensis, C. adenocaule, C. pseudopulchelusis and R. natalensis) with ethnobotanically reported anticancer usage showed cytotoxic activity against brine shrimp and were selected for further in vitro anticancer studies. All four plants were found to have antiproliferative activity against HeLa cervical cancer cells with IC50 values ranging from 3.4 to 50.8 µg/mL. This study was also tasked with investigating the mechanism of action of C. pseudopulchelus on HeLa cervical cancer cells by exploring cell cycle arrest, biochemical markers of apoptosis and mitochondrial membrane x disruption. C. pseudopulchelus caused early M phase arrest followed by slippage and subsequent cell death via the extrinsic pathway of apoptosis. In conclusion, this study showed that both novel complexes as well as medicinal plant extracts represent an interesting avenue for the development of novel anticancer drugs that are cheaper and produce less side effects than current therapies. , Thesis (MSc) -- Faculty of Science, School of Biomolecular and Chemical Sciences, 2021
- Full Text: false
- Date Issued: 2021-04
- Authors: Oosthuizen, Kenneth Thomas , Venables, Luanne
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Medicinal plants -- South Africa
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52028 , vital:43441
- Description: Cancer is a complex, multifactorial disease that affects millions of individuals every year. The adverse side effects and escalating costs of current therapies coupled with the increased incidence of resistance to these therapies make it imperative that we explore novel treatments for the disease. This study investigated two avenues for novel drug design namely, novel synthetic compounds and medicinal plant extracts. The benzimidazole ring system has shown potential as a scaffold for designing novel anticancer agents. Conjugation of the metal rhenium to novel variants of this ring system open up the possibility of designing novel drugs that serve both a diagnostic and a therapeutic function. This study investigated the in vitro anticancer potential of ten such complexes against selected breast and cervical cancer cell lines. The selectivity of the complexes for cancer cells over normal cells was also investigated while the mechanism of action of effective complexes was determined by exploring cell cycle arrest, biochemical markers of apoptosis and mitochondrial membrane disruption. All ten complexes were screened against MCF7 breast and HeLa cervical cancer cell lines with four showing antiproliferative activity against both cancer cell lines and one showing cell line specific toxicity against MCF7 breast cancer cells. When looking at the SAR of the compounds it was noted that activity was higher in compounds which contain two potentially bidentate benzothiazole ligands, while compounds with potentially tridentate ligands show good activity, but only if the benzothiazole moiety is not involved in coordination. The complexes showed greater antiproliferative activity against the MCF7 breast cancer cells with IC50 values ranging from 3.2 to 7.0 µM versus a range of 7.6 to 24.4 µM being obtained on the HeLa cervical cancer cells. Antiproliferative complexes were tested against confluent and log phase Vero cells at their respective IC50 values to determine their effect on “normal” cells. Confluent Vero cells showed less cell death than those in log phase indicating that the complexes show preference for proliferating cells. The mechanism of action of the complexes was studied on both cancer cell lines via cell cycle analysis and apoptosis assays investigating phosphatidylserine translocation, caspase -3 and -8 activation and mitochondrial membrane potential with most complexes inducing cell cycle arrest followed by cell death via both the extrinsic and intrinsic pathways of apoptosis. Since 1940, 49% of all the available anticancer drugs approved for cancer treatment were natural products or directly derived from natural products. Plants are an excellent source of secondary metabolites, many of which are unique chemical compounds that cannot be synthesized in a laboratory. Ethnobotanical surveys conducted in conjunction with Traditional Health Practitioners of the Mkuranga and Same districts in Tanzania identified 25 plants that are used for the treatment of cancer. Four of these plants (A. mossambicensis, C. adenocaule, C. pseudopulchelusis and R. natalensis) with ethnobotanically reported anticancer usage showed cytotoxic activity against brine shrimp and were selected for further in vitro anticancer studies. All four plants were found to have antiproliferative activity against HeLa cervical cancer cells with IC50 values ranging from 3.4 to 50.8 µg/mL. This study was also tasked with investigating the mechanism of action of C. pseudopulchelus on HeLa cervical cancer cells by exploring cell cycle arrest, biochemical markers of apoptosis and mitochondrial membrane x disruption. C. pseudopulchelus caused early M phase arrest followed by slippage and subsequent cell death via the extrinsic pathway of apoptosis. In conclusion, this study showed that both novel complexes as well as medicinal plant extracts represent an interesting avenue for the development of novel anticancer drugs that are cheaper and produce less side effects than current therapies. , Thesis (MSc) -- Faculty of Science, School of Biomolecular and Chemical Sciences, 2021
- Full Text: false
- Date Issued: 2021-04
Assessing community engagement programmes to mitigate uncontrolled, deliberately ignited fires in forestry plantation-communal land interface, Kwazulu-Natal, South Africa
- Authors: Ramantswana, Nontethelelo
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Forest fires -- South Africa
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52008 , vital:43439
- Description: Forestry plantations in South Africa are surrounded by poor local communities that normally depend on the forestry resources for subsistence. This interface area is faced with human induced uncontrolled fires. There is a lack of understanding regarding the reasons for deliberate ignition of uncontrolled fires. There is also lack of information on the success of community engagement programs (CEPs) as fire protection intervention and / or their implementation barriers. , Thesis (MSc) -- Faculty of Science, School of Natural Resource Management, 2021
- Full Text: false
- Date Issued: 2021-04
- Authors: Ramantswana, Nontethelelo
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Forest fires -- South Africa
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52008 , vital:43439
- Description: Forestry plantations in South Africa are surrounded by poor local communities that normally depend on the forestry resources for subsistence. This interface area is faced with human induced uncontrolled fires. There is a lack of understanding regarding the reasons for deliberate ignition of uncontrolled fires. There is also lack of information on the success of community engagement programs (CEPs) as fire protection intervention and / or their implementation barriers. , Thesis (MSc) -- Faculty of Science, School of Natural Resource Management, 2021
- Full Text: false
- Date Issued: 2021-04
Bayesian tolerance intervals and their application to ballistic grouping consistency in competitive shooting
- Authors: Gardiner, Marcell
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Bayesian statistical decision theory -- South Africa
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52456 , vital:43662
- Description: Competitive shooting has become immensely popular since the first modern Games held in 1896. Competitive shooting remains the fastest growing segment in the firearm industry, catering to individuals of all ages. Elite shooters are capable of reaching extreme levels of precision and accuracy, despite many factors existing that affect performance in Olympic shooting. One such factor is speed variability, which is dependent on the mass of the bullet, how constant is the lead being used, and, how constant the gunpowder is. The study focusses on the Bayesian simulation method proposed by Wolfinger (1998) for determining three kinds of tolerance intervals when applied to competitive shooting data, particularly examining speed variability. These tolerance intervals included the (𝛼𝛼,𝛿𝛿) tolerance interval (where 𝛼𝛼 is the content and 𝛿𝛿 is the confidence), the 𝛼𝛼 - expectation tolerance interval (where 𝛼𝛼 is the expected coverage of the interval) and the fixed-in advance tolerance interval in which the interval is held fixed and the proportion of process measurements it contains, is estimated. Speed measurements of individual shots fired were measured with either an Oeler 35P or Magneto Speed V3 chronograph and the data was fitted to a one-way random effects model. , Thesis (MSc) -- Faculty of Science, School of Computer Science, Mathematics, Physics and Statistics, 2021
- Full Text:
- Date Issued: 2021-04
- Authors: Gardiner, Marcell
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Bayesian statistical decision theory -- South Africa
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52456 , vital:43662
- Description: Competitive shooting has become immensely popular since the first modern Games held in 1896. Competitive shooting remains the fastest growing segment in the firearm industry, catering to individuals of all ages. Elite shooters are capable of reaching extreme levels of precision and accuracy, despite many factors existing that affect performance in Olympic shooting. One such factor is speed variability, which is dependent on the mass of the bullet, how constant is the lead being used, and, how constant the gunpowder is. The study focusses on the Bayesian simulation method proposed by Wolfinger (1998) for determining three kinds of tolerance intervals when applied to competitive shooting data, particularly examining speed variability. These tolerance intervals included the (𝛼𝛼,𝛿𝛿) tolerance interval (where 𝛼𝛼 is the content and 𝛿𝛿 is the confidence), the 𝛼𝛼 - expectation tolerance interval (where 𝛼𝛼 is the expected coverage of the interval) and the fixed-in advance tolerance interval in which the interval is held fixed and the proportion of process measurements it contains, is estimated. Speed measurements of individual shots fired were measured with either an Oeler 35P or Magneto Speed V3 chronograph and the data was fitted to a one-way random effects model. , Thesis (MSc) -- Faculty of Science, School of Computer Science, Mathematics, Physics and Statistics, 2021
- Full Text:
- Date Issued: 2021-04
Benthic habitat mapping using marine geophysics and machine learning on the continental shelf of South Africa
- Authors: Pillay, Talicia
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Marine geophysics
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52061 , vital:43452
- Description: A method to map seafloor substrates using machine learning, based primarily on hydroacoustic data including multibeam bathymetry, backscatter, and side-scan sonar, has been developed. The aim was to produce a customdesigned benthic habitat classification method that digitally integrates marine geophysics and biological science data, with relevance to all elements of the local substrate, and this was the first time it was attempted in a South African context. The algorithm developed is able to produce bio-physical benthic habitat maps and this can be extended along the continental shelf of South Africa as new data setsare collected and the algorithm is supplemented. At the outset, this work has focused on broad categories of rock and detailed categories of sediment. Four study sites with varying substrate were selected to holistically build the algorithm that followed a tiered approach of machine learning: Table Bay, Clifton, Koeberg Harbour and Cape St Francis. Table Bay was used to develop a new method of physical seafloor classification, by comparing and contrasting a number of statistical algorithms and software programs. Clifton was used to test the developed clustering algorithm, and Koeberg which is 35 km to the north was used to validate the algorithm because sediment samples, along with drop camera footage, were integrated to better define the results. The resultant verified algorithm was tested at Cape St Francis, where Remotely Operated Vehicle (ROV) footage was acquired in addition to hydroacoustic data. In the first phase of the process towards developing an algorithm, a customised tool was created within ArcGIS using python scripting language to classify seafloor bathymetry, which can be applied to any area of seafloor whatsoever. The tool was based on pioneering work done by the National Oceanic and Atmospheric Administration (NOAA) on a benthic terrain modelling toolbox and adapted to include side-scan sonar data. In the second phase of work, multibeam bathymetry, backscatter and side-scan sonar data that were processed using Qimera, Fledermaus Geocoder Toolbox, and Navlog processing software, were classified using different machine learning techniques including Decision Trees, Random Forests, and k-means clustering computer algorithms. The results from these algorithms were compared to manually-digitised polygons which were created to classify the seafloor substrate distribution by identification of different textures. Integrating all results facilitated a quantitative comparison that illuminated advantages and disadvantages of each machine learning technique and ultimately the k-means clustering techniques were found to be the simplest to implement and understand and worked most efficiently based on their seafloor segmentation capabilities in Table Bay, against all three hydroacoustic data sets (multibeam bathymetry, backscatter and side-scan sonar). In the third phase of work, ground-truthed seafloor characterisation maps were produced for the two study areas of Clifton and Koeberg Harbour. This applied multibeam bathymetry and backscatter data that were collected and processed with machine learning clustering techniques. , Thesis (MSc) -- Faculty of Science, Ocean Sciences, 2021
- Full Text: false
- Date Issued: 2021-04
- Authors: Pillay, Talicia
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Marine geophysics
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52061 , vital:43452
- Description: A method to map seafloor substrates using machine learning, based primarily on hydroacoustic data including multibeam bathymetry, backscatter, and side-scan sonar, has been developed. The aim was to produce a customdesigned benthic habitat classification method that digitally integrates marine geophysics and biological science data, with relevance to all elements of the local substrate, and this was the first time it was attempted in a South African context. The algorithm developed is able to produce bio-physical benthic habitat maps and this can be extended along the continental shelf of South Africa as new data setsare collected and the algorithm is supplemented. At the outset, this work has focused on broad categories of rock and detailed categories of sediment. Four study sites with varying substrate were selected to holistically build the algorithm that followed a tiered approach of machine learning: Table Bay, Clifton, Koeberg Harbour and Cape St Francis. Table Bay was used to develop a new method of physical seafloor classification, by comparing and contrasting a number of statistical algorithms and software programs. Clifton was used to test the developed clustering algorithm, and Koeberg which is 35 km to the north was used to validate the algorithm because sediment samples, along with drop camera footage, were integrated to better define the results. The resultant verified algorithm was tested at Cape St Francis, where Remotely Operated Vehicle (ROV) footage was acquired in addition to hydroacoustic data. In the first phase of the process towards developing an algorithm, a customised tool was created within ArcGIS using python scripting language to classify seafloor bathymetry, which can be applied to any area of seafloor whatsoever. The tool was based on pioneering work done by the National Oceanic and Atmospheric Administration (NOAA) on a benthic terrain modelling toolbox and adapted to include side-scan sonar data. In the second phase of work, multibeam bathymetry, backscatter and side-scan sonar data that were processed using Qimera, Fledermaus Geocoder Toolbox, and Navlog processing software, were classified using different machine learning techniques including Decision Trees, Random Forests, and k-means clustering computer algorithms. The results from these algorithms were compared to manually-digitised polygons which were created to classify the seafloor substrate distribution by identification of different textures. Integrating all results facilitated a quantitative comparison that illuminated advantages and disadvantages of each machine learning technique and ultimately the k-means clustering techniques were found to be the simplest to implement and understand and worked most efficiently based on their seafloor segmentation capabilities in Table Bay, against all three hydroacoustic data sets (multibeam bathymetry, backscatter and side-scan sonar). In the third phase of work, ground-truthed seafloor characterisation maps were produced for the two study areas of Clifton and Koeberg Harbour. This applied multibeam bathymetry and backscatter data that were collected and processed with machine learning clustering techniques. , Thesis (MSc) -- Faculty of Science, Ocean Sciences, 2021
- Full Text: false
- Date Issued: 2021-04
Cluster analysis for group selection in launch sales predictions
- Authors: Watchurst, Lee
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Cluster analysis
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52003 , vital:43447
- Description: One way for businesses to stay ahead in a competitive market is through the launch of new products and planning for these launches optimally. This includes ordering the correct quantity of stock in advance as well as maintaining these stock levels while the item launches. However, holding too much stock in warehouses can affect the business costs adversely. This research proposes the use of cluster analysis techniques to determine the up-front purchase quantity by identifying similar items and using their initial quantities sold. Products will be grouped based on their numerical and categorical attributes. Once the data is clustered, the Bass model will be used to obtain a sales profile for the new item. The Bass model is a popular choice for product life cycle planning due to the emphasis placed on the timing of adoption. The study will make use of data from a retail and wholesale company that sells, in part, single use items. With the planning for new launches being a key problem point in many companies, this research aims to optimise the planning process and ensure product launch success across stores. , Thesis (MSc) -- Faculty of Science, School of Computer Science, Mathematics, Physics and Statistics, 2021
- Full Text: false
- Date Issued: 2021-04
- Authors: Watchurst, Lee
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Cluster analysis
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52003 , vital:43447
- Description: One way for businesses to stay ahead in a competitive market is through the launch of new products and planning for these launches optimally. This includes ordering the correct quantity of stock in advance as well as maintaining these stock levels while the item launches. However, holding too much stock in warehouses can affect the business costs adversely. This research proposes the use of cluster analysis techniques to determine the up-front purchase quantity by identifying similar items and using their initial quantities sold. Products will be grouped based on their numerical and categorical attributes. Once the data is clustered, the Bass model will be used to obtain a sales profile for the new item. The Bass model is a popular choice for product life cycle planning due to the emphasis placed on the timing of adoption. The study will make use of data from a retail and wholesale company that sells, in part, single use items. With the planning for new launches being a key problem point in many companies, this research aims to optimise the planning process and ensure product launch success across stores. , Thesis (MSc) -- Faculty of Science, School of Computer Science, Mathematics, Physics and Statistics, 2021
- Full Text: false
- Date Issued: 2021-04
Deep neural networks for robot vision in evolutionary robotics
- Authors: Watt,Nathan
- Date: 2021-04
- Subjects: Evolutionary robotics -- South Africa , Neural networks
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/61601 , vital:71473
- Description: Description Advances in electronics manufacturing have made robots and their sensors cheaper and more accessible. Robots can have a variety of sensors, such as touch sensors, distance sensors and cameras. A robot’s controller is the software which interprets its sensors and determines how the robot will behave. The difficulty in programming robot controllers increases with complex robots and complicated tasks, forming a barrier to deploying robots for real-world applications. Robot controllers can be automatically created with Evolutionary Robotics (ER). ER makes use of an Evolutionary Algorithm (EA) to evolve controllers to complete a particular task. Instead of manually programming controllers, an EA can evolve controllers when provided with the robot’s task. ER has been used to evolve controllers for many different kinds of robots with a variety of sensors, however the use of robots with on-board camera sensors has been limited. The nature of EAs makes evolving a controller for a camera-equipped robot particularly difficult. There are two main challenges which complicate the evolution of vision-based controllers. First, every image from a camera contains a large amount of information, and a controller needs many parameters to receive that information, however it is difficult to evolve controllers with such a large number of parameters using EAs. Second, during the process of evolution, it is necessary to evaluate the fitness of many candidate controllers. This is typically done in simulation, however creating a simulator for a camera sensor is a tedious and timeconsuming task, as building a photo-realistic simulated environment requires handcrafted 3-dimensional models, textures and lighting. Two techniques have been used in previous experiments to overcome the challenges associated with evolving vision-based controllers. Either the controller was provided with extremely low-resolution images, or a task-specific algorithm was used to preprocess the images, only providing the necessary information to the controller. , Thesis (MSc) -- Faculty of Science, School of Computer Science, Mathematics, Physics and Statistics, 2021
- Full Text:
- Date Issued: 2021-04
- Authors: Watt,Nathan
- Date: 2021-04
- Subjects: Evolutionary robotics -- South Africa , Neural networks
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/61601 , vital:71473
- Description: Description Advances in electronics manufacturing have made robots and their sensors cheaper and more accessible. Robots can have a variety of sensors, such as touch sensors, distance sensors and cameras. A robot’s controller is the software which interprets its sensors and determines how the robot will behave. The difficulty in programming robot controllers increases with complex robots and complicated tasks, forming a barrier to deploying robots for real-world applications. Robot controllers can be automatically created with Evolutionary Robotics (ER). ER makes use of an Evolutionary Algorithm (EA) to evolve controllers to complete a particular task. Instead of manually programming controllers, an EA can evolve controllers when provided with the robot’s task. ER has been used to evolve controllers for many different kinds of robots with a variety of sensors, however the use of robots with on-board camera sensors has been limited. The nature of EAs makes evolving a controller for a camera-equipped robot particularly difficult. There are two main challenges which complicate the evolution of vision-based controllers. First, every image from a camera contains a large amount of information, and a controller needs many parameters to receive that information, however it is difficult to evolve controllers with such a large number of parameters using EAs. Second, during the process of evolution, it is necessary to evaluate the fitness of many candidate controllers. This is typically done in simulation, however creating a simulator for a camera sensor is a tedious and timeconsuming task, as building a photo-realistic simulated environment requires handcrafted 3-dimensional models, textures and lighting. Two techniques have been used in previous experiments to overcome the challenges associated with evolving vision-based controllers. Either the controller was provided with extremely low-resolution images, or a task-specific algorithm was used to preprocess the images, only providing the necessary information to the controller. , Thesis (MSc) -- Faculty of Science, School of Computer Science, Mathematics, Physics and Statistics, 2021
- Full Text:
- Date Issued: 2021-04
Descriptive analysis of organic farming systems and factors affecting adoption and disadoption in Zambia : a case study
- Authors: Munthali, Robert
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Organic farming -- Zambia
- Language: English
- Type: Doctoral theses , text
- Identifier: http://hdl.handle.net/10948/52111 , vital:43434
- Description: The last 30 years since the early 1990s has seen the heightened promotion of organic farming as a sustainable farming option in Zambia. The main objectives of this research study were to carry out an analysis of organic farming systems in order to understand factors affecting adoption, as well as reasons why some organic farmers discontinue using the approach. The study involved four districts in Zambia: Mazabuka, Chikankata, Lusaka and Chongwe. Specifically, the research investigated challenges and limitations of organic agronomic practices, and analysed factors that influence adopters and disadopters of organic agriculture. Primary data were collected through interviews and by administering semi-structured questionnaires. The research population comprised approximately 250 farmers adopters and disadopters, with the sample size being 50 farmers chosen across four districts using systematic random sampling methods. Additionally, five observation research field trials were conducted in order to gather practical data on individual farmer’s crop management practices. Farmers adopting and maintaining organic methods reported several reasons for embracing the system, some of which are: organic farming is regenerative; it is environmentally responsible; it uses manure, believed to help build healthy soils rich in organic matter with better nutrient and moisture holding capacity; pesticides and synthetic fertilizers are not used, reducing the levels of toxic residues in organic products which ultimately promotes health. Labour constraints, pest and disease challenges and weed problems were some of the reasons for disadoption. Data showed that disadopters had lower education (48% had primary education), were older and with slightly more children. Although there were more women adopting organic agriculture compared to men, the difference was small 56% and 44% women and men respectively. Among disadopters there were more women (67%) compared to men (33%). Overall, the study results suggest that promotion of organic agriculture should be encouraged through liaison with government departments such as the Zambian Department of Agriculture to include organic agriculture when developing policies. In general, organic farmers have positive opinions concerning the benefits resulting from organic agriculture. Organic farmers should enjoy equal opportunities to conventional agriculture in the provision of extension services, input support, infrastructural development, access to financial resources to procure farm machinery, and equipment and most important of all, they should get fair prices for their organic produce and products. , Thesis (PhD) -- Faculty of Science, School of Environmental Sciences, 2021
- Full Text: false
- Date Issued: 2021-04
- Authors: Munthali, Robert
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Organic farming -- Zambia
- Language: English
- Type: Doctoral theses , text
- Identifier: http://hdl.handle.net/10948/52111 , vital:43434
- Description: The last 30 years since the early 1990s has seen the heightened promotion of organic farming as a sustainable farming option in Zambia. The main objectives of this research study were to carry out an analysis of organic farming systems in order to understand factors affecting adoption, as well as reasons why some organic farmers discontinue using the approach. The study involved four districts in Zambia: Mazabuka, Chikankata, Lusaka and Chongwe. Specifically, the research investigated challenges and limitations of organic agronomic practices, and analysed factors that influence adopters and disadopters of organic agriculture. Primary data were collected through interviews and by administering semi-structured questionnaires. The research population comprised approximately 250 farmers adopters and disadopters, with the sample size being 50 farmers chosen across four districts using systematic random sampling methods. Additionally, five observation research field trials were conducted in order to gather practical data on individual farmer’s crop management practices. Farmers adopting and maintaining organic methods reported several reasons for embracing the system, some of which are: organic farming is regenerative; it is environmentally responsible; it uses manure, believed to help build healthy soils rich in organic matter with better nutrient and moisture holding capacity; pesticides and synthetic fertilizers are not used, reducing the levels of toxic residues in organic products which ultimately promotes health. Labour constraints, pest and disease challenges and weed problems were some of the reasons for disadoption. Data showed that disadopters had lower education (48% had primary education), were older and with slightly more children. Although there were more women adopting organic agriculture compared to men, the difference was small 56% and 44% women and men respectively. Among disadopters there were more women (67%) compared to men (33%). Overall, the study results suggest that promotion of organic agriculture should be encouraged through liaison with government departments such as the Zambian Department of Agriculture to include organic agriculture when developing policies. In general, organic farmers have positive opinions concerning the benefits resulting from organic agriculture. Organic farmers should enjoy equal opportunities to conventional agriculture in the provision of extension services, input support, infrastructural development, access to financial resources to procure farm machinery, and equipment and most important of all, they should get fair prices for their organic produce and products. , Thesis (PhD) -- Faculty of Science, School of Environmental Sciences, 2021
- Full Text: false
- Date Issued: 2021-04
Estimating activity patterns in camera trap studies, and the overlap between domestic and indigenous ungulates
- Authors: Peral, Christopher Michael
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Ungulates -- Behavior -- South Africa
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52019 , vital:43437
- Description: Animal activity can reveal responses to social and ecological factors such as the presence of food, water, shelter, predators, competitors or conspecifics. Thus, an understanding of an animal’s activity patterns provides insights into how and why they are using landscapes in the manner in which they do. Camera trap surveys, typically used to record the occurrence of species of interest, are also an efficient way to collect data regarding animal activity patterns but require robust analytical methods. The analysis of camera trap data requires assumptions that may differ for camera trap surveys that estimate occurrence or activity. One such assumption is temporal independence. This refers to the removal of images within a specified time interval, as these images are assumed to be images of the same individual. This assumption stems from the use of-time-to-independence in occurrence or density-based studies, where it reduces overestimation of animal counts. In studies of animal activity, this assumption is generally used, even though animal behaviour or activity occurs continuously over time and as such should be measured over time. I analysed 112 467 images of eight cooccurring species, to test the effect across various times to independence, these being set at 1, 5, 10, 15, 30 and 60 minutes. Time-to-independence resulted in significant loss of data and biased the interpretation of the activity patterns of species and interactions amongst species, particularly for herbivorous social species. In light of these findings, activity pattern estimates based on camera trap studies should be cautious when using time-to-independence. I then explore ‘habitat use’ in space and time by domestic and indigenous ungulates using a camera trap survey. Domestic ungulates (livestock) have become common place on rangelands and may compete with native species. It was hypothesised that there will be overlap in habitat use by livestock and indigenous ungulates. My results show that indigenous ungulate species, particularly larger species such as kudu (Tragelaphus strepsiceros) and springbok (Antidorcas marsupialis), changed their spatial and temporal use as a result of the presence of domestic ungulates. Other smaller indigenous ungulates, such as steenbok (Raphicerus campestris) and duiker (Sylvicapra grimmia), were less affected by the presence of livestock. Kudu were found to use more open habitats at different times whilst springbok used more dense habitats at different times in the presence of livestock. This supports the hypothesis of competition between livestock and indigenous ungulates. This study provides novel insight into the analysis and interpretation of activity patterns based on camera trap surveys. Additionally, this study demonstrates the competitive displacement between domestic and indigenous ungulate species. , Thesis (MSc) -- Faculty of Science, School of Environmental Sciences, 2021
- Full Text: false
- Date Issued: 2021-04
- Authors: Peral, Christopher Michael
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Ungulates -- Behavior -- South Africa
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52019 , vital:43437
- Description: Animal activity can reveal responses to social and ecological factors such as the presence of food, water, shelter, predators, competitors or conspecifics. Thus, an understanding of an animal’s activity patterns provides insights into how and why they are using landscapes in the manner in which they do. Camera trap surveys, typically used to record the occurrence of species of interest, are also an efficient way to collect data regarding animal activity patterns but require robust analytical methods. The analysis of camera trap data requires assumptions that may differ for camera trap surveys that estimate occurrence or activity. One such assumption is temporal independence. This refers to the removal of images within a specified time interval, as these images are assumed to be images of the same individual. This assumption stems from the use of-time-to-independence in occurrence or density-based studies, where it reduces overestimation of animal counts. In studies of animal activity, this assumption is generally used, even though animal behaviour or activity occurs continuously over time and as such should be measured over time. I analysed 112 467 images of eight cooccurring species, to test the effect across various times to independence, these being set at 1, 5, 10, 15, 30 and 60 minutes. Time-to-independence resulted in significant loss of data and biased the interpretation of the activity patterns of species and interactions amongst species, particularly for herbivorous social species. In light of these findings, activity pattern estimates based on camera trap studies should be cautious when using time-to-independence. I then explore ‘habitat use’ in space and time by domestic and indigenous ungulates using a camera trap survey. Domestic ungulates (livestock) have become common place on rangelands and may compete with native species. It was hypothesised that there will be overlap in habitat use by livestock and indigenous ungulates. My results show that indigenous ungulate species, particularly larger species such as kudu (Tragelaphus strepsiceros) and springbok (Antidorcas marsupialis), changed their spatial and temporal use as a result of the presence of domestic ungulates. Other smaller indigenous ungulates, such as steenbok (Raphicerus campestris) and duiker (Sylvicapra grimmia), were less affected by the presence of livestock. Kudu were found to use more open habitats at different times whilst springbok used more dense habitats at different times in the presence of livestock. This supports the hypothesis of competition between livestock and indigenous ungulates. This study provides novel insight into the analysis and interpretation of activity patterns based on camera trap surveys. Additionally, this study demonstrates the competitive displacement between domestic and indigenous ungulate species. , Thesis (MSc) -- Faculty of Science, School of Environmental Sciences, 2021
- Full Text: false
- Date Issued: 2021-04
Estimating Bayesian tolerance intervals for a two - factor factorial model
- Authors: Besele, Kagiso Francis
- Date: 2021-04
- Subjects: Gqenerha (South Africa) , Eastern Cape (South Africa) , Mathematical statistics
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52302 , vital:43587
- Description: Quality improvement efforts have become the cornerstone of all manufacturing processes. Quality can be defined in terms of variability reduction, and since variability is a statistical concept, statistical techniques such as statistical quality control present techniques for assessing process variation. Methods such as experimental design provide a way to ascertain factor relationships and give a basis for computing variability that arises from each process variable, ultimately providing a way of calculating total process variability. This in turn results in variance components and eventually variance component estimation. As with any statistical model, estimates may be classified in any one of two ways, point estimates or interval estimates. Interval estimates that provide information about an entire population, and not only information on a few observations from a sample or knowledge about only a population parameter, are known as tolerance intervals. Wolfinger (1998) provided a Bayesian simulationbased approach for ascertaining three types of tolerance intervals using a balanced one-way random effects model. In this study, the method initially proposed by Wolfinger (1998), is extended in order to estimate tolerance intervals for the balanced two-way crossed classification random effects model with interaction. The suggested and derived techniques will be applied to the thermal impedance data initially collected by Houf and Berman (1988), and the method presented by Wolfinger (1998) will be expanded to also include the estimation of tolerance intervals for averages of observations from new or unknown measurements. This Bayesian approach provides a thorough but yet simplistic paradigm to using tolerance intervals in manufacturing settings. , Thesis (MSc) -- Faculty of Science, Statistics, 2021
- Full Text: false
- Date Issued: 2021-04
- Authors: Besele, Kagiso Francis
- Date: 2021-04
- Subjects: Gqenerha (South Africa) , Eastern Cape (South Africa) , Mathematical statistics
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52302 , vital:43587
- Description: Quality improvement efforts have become the cornerstone of all manufacturing processes. Quality can be defined in terms of variability reduction, and since variability is a statistical concept, statistical techniques such as statistical quality control present techniques for assessing process variation. Methods such as experimental design provide a way to ascertain factor relationships and give a basis for computing variability that arises from each process variable, ultimately providing a way of calculating total process variability. This in turn results in variance components and eventually variance component estimation. As with any statistical model, estimates may be classified in any one of two ways, point estimates or interval estimates. Interval estimates that provide information about an entire population, and not only information on a few observations from a sample or knowledge about only a population parameter, are known as tolerance intervals. Wolfinger (1998) provided a Bayesian simulationbased approach for ascertaining three types of tolerance intervals using a balanced one-way random effects model. In this study, the method initially proposed by Wolfinger (1998), is extended in order to estimate tolerance intervals for the balanced two-way crossed classification random effects model with interaction. The suggested and derived techniques will be applied to the thermal impedance data initially collected by Houf and Berman (1988), and the method presented by Wolfinger (1998) will be expanded to also include the estimation of tolerance intervals for averages of observations from new or unknown measurements. This Bayesian approach provides a thorough but yet simplistic paradigm to using tolerance intervals in manufacturing settings. , Thesis (MSc) -- Faculty of Science, Statistics, 2021
- Full Text: false
- Date Issued: 2021-04
Estimating the population size of three large carnivore species and the diet of six large carnivore species, in Madikwe Game Reserve
- Authors: Honiball, Terry-Lee
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Carnivores
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52139 , vital:43421
- Description: Large carnivores play a vital role in ecosystems, providing ecosystem services such as herbivore regulation. If not properly monitored, carnivores could place significant pressure on certain prey species, and in extreme cases, result in the local extinction of certain species. A lack of demographic data can indirectly threaten carnivores, as policies and management decisions based on demographic data can result in the over harvesting or unsustainable hunting of large carnivores. Furthermore, their conservation is becoming confined to small, fenced, protected areas. Whereby species are intensively managed, making demographic and ecological data a fundamental part of conservation decisions. Therefore, it is essential for conservation practitioners to understand large carnivore demography and their role, to conserve them effectively. To provide Madikwe Game Reserve with the necessary data for effective long-term monitoring and management of the large carnivore guild, I estimated the population size of the cryptic carnivores. I then investigated the efficacy of different camera trap survey designs on improving the detection probability of leopard (Panthera pardus), brown hyaena (Parahyaena brunnea), and spotted hyaena (Crocuta crocuta). I found that brown hyaena and spotted hyaena detection probability increased with a higher number of camera traps, whereas leopard detection probability did not increase with an increased number of camera traps. I further investigated the diet, prey preferences and dietary overlap of lion (Panthera leo), cheetah (Acinonyx jubatus), African wild dog (Lycaon pictus), leopard, spotted hyaena and brown hyaena. I found that all species displayed a degree of dietary overlap, however the most notable level of competition occurred between leopards, wild dogs, and cheetahs. This iv information will assist MGR to make informed decisions about large carnivore management in the reserve. I recommend continued long-term monitoring for the establishment of trends over time, allowing for holistic management of the entire large carnivore guild. , Thesis (MSc) -- Faculty of Science, School of Natural Resource Management, 2021
- Full Text: false
- Date Issued: 2021-04
- Authors: Honiball, Terry-Lee
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Carnivores
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52139 , vital:43421
- Description: Large carnivores play a vital role in ecosystems, providing ecosystem services such as herbivore regulation. If not properly monitored, carnivores could place significant pressure on certain prey species, and in extreme cases, result in the local extinction of certain species. A lack of demographic data can indirectly threaten carnivores, as policies and management decisions based on demographic data can result in the over harvesting or unsustainable hunting of large carnivores. Furthermore, their conservation is becoming confined to small, fenced, protected areas. Whereby species are intensively managed, making demographic and ecological data a fundamental part of conservation decisions. Therefore, it is essential for conservation practitioners to understand large carnivore demography and their role, to conserve them effectively. To provide Madikwe Game Reserve with the necessary data for effective long-term monitoring and management of the large carnivore guild, I estimated the population size of the cryptic carnivores. I then investigated the efficacy of different camera trap survey designs on improving the detection probability of leopard (Panthera pardus), brown hyaena (Parahyaena brunnea), and spotted hyaena (Crocuta crocuta). I found that brown hyaena and spotted hyaena detection probability increased with a higher number of camera traps, whereas leopard detection probability did not increase with an increased number of camera traps. I further investigated the diet, prey preferences and dietary overlap of lion (Panthera leo), cheetah (Acinonyx jubatus), African wild dog (Lycaon pictus), leopard, spotted hyaena and brown hyaena. I found that all species displayed a degree of dietary overlap, however the most notable level of competition occurred between leopards, wild dogs, and cheetahs. This iv information will assist MGR to make informed decisions about large carnivore management in the reserve. I recommend continued long-term monitoring for the establishment of trends over time, allowing for holistic management of the entire large carnivore guild. , Thesis (MSc) -- Faculty of Science, School of Natural Resource Management, 2021
- Full Text: false
- Date Issued: 2021-04
Evolutionary robotics controllers with location perception facilitated by neural network-based simulators
- Authors: Phillips, Antin Paul
- Date: 2021-04
- Subjects: Evolutionary Robots -- South Africa , Neural networks (Computer science) -- South Africa
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/61613 , vital:71474
- Description: Humans impressively maintain a real-time approximation of their bodily form. For instance, one knows where one’s arm is, relative to the body, without needing to directly observe it. This ability, in part, allows humans to interact with the environment without direct observation. This bodily sense is referred to as ”proprioception“. The human body contains various proprioceptors, sensory neurons which provide information about the physical state of the body. This information, along with internal body representations that humans develop over time, allows one to maintain an approximation of their bodily form. Humans also possess an impressive sense of direction and navigation ability. For instance, a blindfolded human can move around a familiar environment and maintain an approximate sense of where they are within that environment. This ability is, in part, enabled by proprioception as it provides one with an approximation of the effects their actions have on their body. The field of Evolutionary Robots (ER) makes extensive use of robotic simulators to carry out simulated robotic evaluations. Research has been conducted into alternate forms of simulation and Simulator Neural Networks (SNNs) were subsequently developed. The speed and accuracy of these SNNs, relative to more typical simulation techniques, is what inspired the approach explored in this research. Robots do not necessarily possess the appropriate hardware to sense their position within an environment. Thus, it was proposed that SNNs could be incorporated into ER controllers to approximate the position of the robot. These SNNs would be executed in parallel to the robot and provide a constant approximation of the robot’s position. This would provide controllers with information that they would not otherwise have, albeit approximate information. Various experiments were carried out which examined both typical ER controllers as well as those which were augmented in the proposed fashion. The augmented controllers were found to outperform typical controllers as well as develop more advanced and efficient behaviours. Furthermore, the augmented controllers demonstrated the ability to solve tasks that regular controllers could not. A potential criticism of the approach suggested in this research is that ER controllers could hypothetically be trained in such a way that the proposed augmentation would be unnecessary. This possibility was investigated and it was found that successfully training controllers in such a manner would be unlikely. Furthermore, the effort involved in fine-tuning this training process would be greater than simply following the approach suggested in this research. Another potential drawback of the suggested approach involved the accuracy of the information that SNNs could provide to controllers. The approximated information was found to diverge over time and negatively affected controller performance. A method to address this issue was proposed and subsequently implemented. This method was demonstrated to be an effective means of reducing the divergence of the SNNs outputs and, in turn, improved controller performance. , Thesis (MSc) -- Faculty of Science, School of Computer Science, Mathematics, Physics and Statistics, 2021
- Full Text:
- Date Issued: 2021-04
- Authors: Phillips, Antin Paul
- Date: 2021-04
- Subjects: Evolutionary Robots -- South Africa , Neural networks (Computer science) -- South Africa
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/61613 , vital:71474
- Description: Humans impressively maintain a real-time approximation of their bodily form. For instance, one knows where one’s arm is, relative to the body, without needing to directly observe it. This ability, in part, allows humans to interact with the environment without direct observation. This bodily sense is referred to as ”proprioception“. The human body contains various proprioceptors, sensory neurons which provide information about the physical state of the body. This information, along with internal body representations that humans develop over time, allows one to maintain an approximation of their bodily form. Humans also possess an impressive sense of direction and navigation ability. For instance, a blindfolded human can move around a familiar environment and maintain an approximate sense of where they are within that environment. This ability is, in part, enabled by proprioception as it provides one with an approximation of the effects their actions have on their body. The field of Evolutionary Robots (ER) makes extensive use of robotic simulators to carry out simulated robotic evaluations. Research has been conducted into alternate forms of simulation and Simulator Neural Networks (SNNs) were subsequently developed. The speed and accuracy of these SNNs, relative to more typical simulation techniques, is what inspired the approach explored in this research. Robots do not necessarily possess the appropriate hardware to sense their position within an environment. Thus, it was proposed that SNNs could be incorporated into ER controllers to approximate the position of the robot. These SNNs would be executed in parallel to the robot and provide a constant approximation of the robot’s position. This would provide controllers with information that they would not otherwise have, albeit approximate information. Various experiments were carried out which examined both typical ER controllers as well as those which were augmented in the proposed fashion. The augmented controllers were found to outperform typical controllers as well as develop more advanced and efficient behaviours. Furthermore, the augmented controllers demonstrated the ability to solve tasks that regular controllers could not. A potential criticism of the approach suggested in this research is that ER controllers could hypothetically be trained in such a way that the proposed augmentation would be unnecessary. This possibility was investigated and it was found that successfully training controllers in such a manner would be unlikely. Furthermore, the effort involved in fine-tuning this training process would be greater than simply following the approach suggested in this research. Another potential drawback of the suggested approach involved the accuracy of the information that SNNs could provide to controllers. The approximated information was found to diverge over time and negatively affected controller performance. A method to address this issue was proposed and subsequently implemented. This method was demonstrated to be an effective means of reducing the divergence of the SNNs outputs and, in turn, improved controller performance. , Thesis (MSc) -- Faculty of Science, School of Computer Science, Mathematics, Physics and Statistics, 2021
- Full Text:
- Date Issued: 2021-04
Evolutionary robotics controllers with location perception facilitated by neural network-based simulators
- Authors: Phillips, Antin Paul
- Date: 2021-04
- Subjects: Grahamstown (South Africa) , Eastern Cape (South Africa) , Neural networks (Computer science) -- South Africa
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52137 , vital:43444
- Description: Humans impressively maintain a real-time approximation of their bodily form. For instance, one knows where one’s arm is, relative to the body, without needing to directly observe it. This ability, in part, allows humans to interact with the environment without direct observation. This bodily sense is referred to as ”proprioception“. The human body contains various proprioceptors, sensory neurons which provide information about the physical state of the body. This information, along with internal body representations that humans develop over time, allows one to maintain an approximation of their bodily form. Humans also possess an impressive sense of direction and navigation ability. For instance, a blindfolded human can move around a familiar environment and maintain an approximate sense of where they are within that environment. This ability is, in part, enabled by proprioception as it provides one with an approximation of the effects their actions have on their body. The field of Evolutionary Robots (ER) makes extensive use of robotic simulators to carry out simulated robotic evaluations. Research has been conducted into alternate forms of simulation and Simulator Neural Networks (SNNs) were subsequently developed. The speed and accuracy of these SNNs, relative to more typical simulation techniques, is what inspired the approach explored in this research. Robots do not necessarily possess the appropriate hardware to sense their position within an environment. Thus, it was proposed that SNNs could be incorporated into ER controllers to approximate the position of the robot. These SNNs would be executed in parallel to the robot and provide a constant approximation of the robot’s position. This would provide controllers with information that they would not otherwise have, albeit approximate information. Various experiments were carried out which examined both typical ER controllers as well as those which were augmented in the proposed fashion. The augmented controllers were found to outperform typical controllers as well as develop more advanced and efficient behaviours. Furthermore, the augmented controllers demonstrated the ability to solve tasks that regular controllers could not. A potential criticism of the approach suggested in this research is that ER controllers could hypothetically be trained in such a way that the proposed augmentation would be unnecessary. This possibility was investigated and it was found that successfully training controllers in such a manner would be unlikely. Furthermore, the effort involved in fine-tuning this training process would be greater than simply following the approach suggested in this research. Another potential drawback of the suggested approach involved the accuracy of the information that SNNs could provide to controllers. The approximated information was found to diverge over time and negatively affected controller performance. A method to address this issue was proposed and subsequently implemented. This method was demonstrated to be an effective means of reducing the divergence of the SNNs outputs and, in turn, improved controller performance. , Thesis (MSc) -- Faculty of Science, School of Computer Science, Mathematics, Physics and Statistics, 2021
- Full Text: false
- Date Issued: 2021-04
- Authors: Phillips, Antin Paul
- Date: 2021-04
- Subjects: Grahamstown (South Africa) , Eastern Cape (South Africa) , Neural networks (Computer science) -- South Africa
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52137 , vital:43444
- Description: Humans impressively maintain a real-time approximation of their bodily form. For instance, one knows where one’s arm is, relative to the body, without needing to directly observe it. This ability, in part, allows humans to interact with the environment without direct observation. This bodily sense is referred to as ”proprioception“. The human body contains various proprioceptors, sensory neurons which provide information about the physical state of the body. This information, along with internal body representations that humans develop over time, allows one to maintain an approximation of their bodily form. Humans also possess an impressive sense of direction and navigation ability. For instance, a blindfolded human can move around a familiar environment and maintain an approximate sense of where they are within that environment. This ability is, in part, enabled by proprioception as it provides one with an approximation of the effects their actions have on their body. The field of Evolutionary Robots (ER) makes extensive use of robotic simulators to carry out simulated robotic evaluations. Research has been conducted into alternate forms of simulation and Simulator Neural Networks (SNNs) were subsequently developed. The speed and accuracy of these SNNs, relative to more typical simulation techniques, is what inspired the approach explored in this research. Robots do not necessarily possess the appropriate hardware to sense their position within an environment. Thus, it was proposed that SNNs could be incorporated into ER controllers to approximate the position of the robot. These SNNs would be executed in parallel to the robot and provide a constant approximation of the robot’s position. This would provide controllers with information that they would not otherwise have, albeit approximate information. Various experiments were carried out which examined both typical ER controllers as well as those which were augmented in the proposed fashion. The augmented controllers were found to outperform typical controllers as well as develop more advanced and efficient behaviours. Furthermore, the augmented controllers demonstrated the ability to solve tasks that regular controllers could not. A potential criticism of the approach suggested in this research is that ER controllers could hypothetically be trained in such a way that the proposed augmentation would be unnecessary. This possibility was investigated and it was found that successfully training controllers in such a manner would be unlikely. Furthermore, the effort involved in fine-tuning this training process would be greater than simply following the approach suggested in this research. Another potential drawback of the suggested approach involved the accuracy of the information that SNNs could provide to controllers. The approximated information was found to diverge over time and negatively affected controller performance. A method to address this issue was proposed and subsequently implemented. This method was demonstrated to be an effective means of reducing the divergence of the SNNs outputs and, in turn, improved controller performance. , Thesis (MSc) -- Faculty of Science, School of Computer Science, Mathematics, Physics and Statistics, 2021
- Full Text: false
- Date Issued: 2021-04
Factors affecting the performance of the wooden furniture manufacturing industry in the Southern Cape
- Authors: Vheriwa, Prayer
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Furniture making -- South Africa
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52017 , vital:43449
- Description: The furniture manufacturing industry in South Africa is facing several challenges that affect the performance of firms in this sector. A literature review indicates several factors such as availability of resources; location; access to local markets; access to export markets; and development programmes that may affect the performance of wooden furniture manufacturing firms in the Southern Cape. The purpose of the current study was to fill the knowledge gap on why the performance of wooden furniture manufacturing firms in the Southern Cape has declined. The study intends to contribute to the limited formal information resources on the Southern Cape furniture manufacturing industry. A conceptual framework was developed and employed for investigating factors hypothesised to affect the performance of furniture manufacturing firms in the Southern Cape. The research conducted was explorative. An interpretive paradigm was selected to collect and interpret qualitative data and a positivist paradigm was used for the quantitative component of the study. The quantitative data was subjected to a chi-square test using IBM SPSS statistics software and qualitative data was analysed for meanings and context using ATLAS.ti software. Statistical tests on qualitative and quantitative data were conducted to determine whether sufficient evidence existed to make conclusions on the hypotheses, relating to significant relationships between the selected independent variables for the study and the performance of wooden furniture manufacturing firms in the Southern Cape. The results show that existing development programmes and availability of resources, namely: timber, skilled human resources and availability of finance, influence the performance of furniture manufacturing firms in the Southern Cape. Results further reveal that location; access to local market; and access to export market have a partial influence on the performance of the furniture manufacturing firms in the Southern Cape. The study presents information which serves to improve the capacity for growth of wooden furniture manufacturing firms. Based on the results, recommendations were made to overcome the factors affecting the performance of the furniture manufacturing firms in the Southern Cape. , Thesis (MSc) -- Faculty of Science, School of Natural Resource Management, 2021
- Full Text: false
- Date Issued: 2021-04
- Authors: Vheriwa, Prayer
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Furniture making -- South Africa
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52017 , vital:43449
- Description: The furniture manufacturing industry in South Africa is facing several challenges that affect the performance of firms in this sector. A literature review indicates several factors such as availability of resources; location; access to local markets; access to export markets; and development programmes that may affect the performance of wooden furniture manufacturing firms in the Southern Cape. The purpose of the current study was to fill the knowledge gap on why the performance of wooden furniture manufacturing firms in the Southern Cape has declined. The study intends to contribute to the limited formal information resources on the Southern Cape furniture manufacturing industry. A conceptual framework was developed and employed for investigating factors hypothesised to affect the performance of furniture manufacturing firms in the Southern Cape. The research conducted was explorative. An interpretive paradigm was selected to collect and interpret qualitative data and a positivist paradigm was used for the quantitative component of the study. The quantitative data was subjected to a chi-square test using IBM SPSS statistics software and qualitative data was analysed for meanings and context using ATLAS.ti software. Statistical tests on qualitative and quantitative data were conducted to determine whether sufficient evidence existed to make conclusions on the hypotheses, relating to significant relationships between the selected independent variables for the study and the performance of wooden furniture manufacturing firms in the Southern Cape. The results show that existing development programmes and availability of resources, namely: timber, skilled human resources and availability of finance, influence the performance of furniture manufacturing firms in the Southern Cape. Results further reveal that location; access to local market; and access to export market have a partial influence on the performance of the furniture manufacturing firms in the Southern Cape. The study presents information which serves to improve the capacity for growth of wooden furniture manufacturing firms. Based on the results, recommendations were made to overcome the factors affecting the performance of the furniture manufacturing firms in the Southern Cape. , Thesis (MSc) -- Faculty of Science, School of Natural Resource Management, 2021
- Full Text: false
- Date Issued: 2021-04
Fish ecology and ecophysiology in an anthropogenically-altered temporarily open/closed estuary in the warm-temperate region of the Eastern Cape Province, South Africa
- Authors: Mpinga, Mfundo
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Fishes--Ecology--South Africa
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52122 , vital:43431
- Description: Estuaries are globally recognized for their contribution to the coastal marine environment by acting as nursery grounds for numerous marine estuarine-dependent fish species. These linking ecosystems between marine and riverine environments, are ideal for these species due to the habitat diversity, nutrient and food availability, shallower and calmer environments they provide relative to the ocean. Abrupt hydrodynamic and physico-chemical changes which frequently occur in estuaries impose physiological stress and requires superior adaptability in the fish that reside within them. This is exacerbated by anthropogenic modifications to estuaries which affects their natural ecological function. The Seekoei Estuary is one of 192 temporarily open/closed estuaries (TOCEs) in South Africa and these estuaries are characterised by their periodical connection to the sea due to the formation of a mouth berm. These estuaries are the dominant type in South Africa. A lot of research has been done in these kinds of estuaries, however, it is still limited due to their vast numbers. These estuaries are popular for human settlement and this has led to mismanagement due to the lack of ecological information. Furthermore, because of their inability to continuously flush out water, temporarily open closed estuaries are more vulnerable to pollution and anthropogenic modifications. The aim of this study was to assess the small fish community in relation to their environment in the hypersaline anthropogenically-altered Seekoei Estuary, by analysing the spatio-temporal trends of physico-chemical properties, nutrient loads, phytoplankton biomass and fish community structure. In addition, histological adaptation in gills of fish that survive hypersaline water was investigated. The study found that physico-chemical properties were influenced by anthropogenic alterations, in particular adverse water abstraction, instream water obstructions and artificial breaching. Nutrient concentration was found to be influenced by the intense agricultural practices in the upper reaches. It was evident that nutrient concentration (dissolved inorganic phosphorus) coupled with high salinity and optimum temperature (15 oC - 20 oC) affected the phytoplankton biomass. This estuary’s inability to naturally breach makes it vulnerable to phytoplankton blooms. The accumulation of cohesive silt caused by low river inflow, due to water abstraction in the upper reaches, was attributed to this estuary’s prolonged mouth closure. This was exacerbated by the presence of the causeway, which acts as a barrier between the middle and lower estuary. The abundance of marine fish species in the estuary, despite its long closure, indicates that the larvae and juveniles of these species make use of overwash events and brief artificial breaching events to recruit into the estuary. The high species richness in iii sites below the causeway is indicative of the barrier effect, which prevents the migration of some species to the upper reaches. This was evident during a massive fish kill, recorded during this study, which was more severe in sites below the causeway than above it. Chloride cell hyperplasia, mucus secretion and thickening of the epithelial layer in gills of indicator fish species from this estuary was the histological adaptation, which enables these fishes to be tolerant to high salinity. , Thesis (MSc) -- Faculty of Science, School of Environmental Sciences, 2021
- Full Text: false
- Date Issued: 2021-04
- Authors: Mpinga, Mfundo
- Date: 2021-04
- Subjects: Gqeberha (South Africa) , Eastern Cape (South Africa) , Fishes--Ecology--South Africa
- Language: English
- Type: Master's theses , text
- Identifier: http://hdl.handle.net/10948/52122 , vital:43431
- Description: Estuaries are globally recognized for their contribution to the coastal marine environment by acting as nursery grounds for numerous marine estuarine-dependent fish species. These linking ecosystems between marine and riverine environments, are ideal for these species due to the habitat diversity, nutrient and food availability, shallower and calmer environments they provide relative to the ocean. Abrupt hydrodynamic and physico-chemical changes which frequently occur in estuaries impose physiological stress and requires superior adaptability in the fish that reside within them. This is exacerbated by anthropogenic modifications to estuaries which affects their natural ecological function. The Seekoei Estuary is one of 192 temporarily open/closed estuaries (TOCEs) in South Africa and these estuaries are characterised by their periodical connection to the sea due to the formation of a mouth berm. These estuaries are the dominant type in South Africa. A lot of research has been done in these kinds of estuaries, however, it is still limited due to their vast numbers. These estuaries are popular for human settlement and this has led to mismanagement due to the lack of ecological information. Furthermore, because of their inability to continuously flush out water, temporarily open closed estuaries are more vulnerable to pollution and anthropogenic modifications. The aim of this study was to assess the small fish community in relation to their environment in the hypersaline anthropogenically-altered Seekoei Estuary, by analysing the spatio-temporal trends of physico-chemical properties, nutrient loads, phytoplankton biomass and fish community structure. In addition, histological adaptation in gills of fish that survive hypersaline water was investigated. The study found that physico-chemical properties were influenced by anthropogenic alterations, in particular adverse water abstraction, instream water obstructions and artificial breaching. Nutrient concentration was found to be influenced by the intense agricultural practices in the upper reaches. It was evident that nutrient concentration (dissolved inorganic phosphorus) coupled with high salinity and optimum temperature (15 oC - 20 oC) affected the phytoplankton biomass. This estuary’s inability to naturally breach makes it vulnerable to phytoplankton blooms. The accumulation of cohesive silt caused by low river inflow, due to water abstraction in the upper reaches, was attributed to this estuary’s prolonged mouth closure. This was exacerbated by the presence of the causeway, which acts as a barrier between the middle and lower estuary. The abundance of marine fish species in the estuary, despite its long closure, indicates that the larvae and juveniles of these species make use of overwash events and brief artificial breaching events to recruit into the estuary. The high species richness in iii sites below the causeway is indicative of the barrier effect, which prevents the migration of some species to the upper reaches. This was evident during a massive fish kill, recorded during this study, which was more severe in sites below the causeway than above it. Chloride cell hyperplasia, mucus secretion and thickening of the epithelial layer in gills of indicator fish species from this estuary was the histological adaptation, which enables these fishes to be tolerant to high salinity. , Thesis (MSc) -- Faculty of Science, School of Environmental Sciences, 2021
- Full Text: false
- Date Issued: 2021-04